Foreword Acknowledgement

Eleventh World Congress on Neutron Capture TherapyMonday, October 11, 2004 - PMParallel Session 1 - B i o l o g y a n d C h e m i s t r yneutron beams in patients with malignant glioma. Eighteen patients with malignantglioma underwent mixed epithermal- and thermal neutron beam and sodium borocaptatebetween 1998 and 2004. The radiation dose (i.e. physical dose of boron n-alphareaction) in the protocol used between 1998 and 2000 (Protocol A, n=8) prescribed amaximum tumor volume dose of 15 Gy. In 2001, a new dose-escalated protocol wasintroduced (Protocol B, n=4); it prescribes a minimum tumor volume dose of 18 Gy or,alternatively, a minimum target volume dose of 15 Gy. Since 2002, the radiation dosewas reduced to 80-90% dose of Protocol B because of acute radiation injury. A newProtocol was applied to 6 glioblastoma patients (Protocol C, n=6).The average values of the maximum vascular dose of brain surface in Protocol A, Band C were 11.4±4.2 Gy, 15.7±1.2 Gy and 13.9±3.6 Gy, respectively. Acute radiationinjury such as a generalized convulsion within 1 week after BNCT was recognizedin three patients of Protocol B. Delayed radiation injury such as a neurological deteriorationafter BNCT was appeared 3-6 months after BNCT, and it was recognized 1patient in Protocol A, 5 patients in Protocol B. According to acute radiation injury, themaximum vascular dose was 15.8±1.3 Gy in positive, it was 12.6±4.3 Gy in negative.There was no significant difference between them. According to the delayed radiationinjury, the maximum vascular dose was 13.8±3.8 Gy in positive, it was 13.6±4.9 Gy innegative. There was no significant difference between them. The dose escalation islimited because most patients in Protocol B suffered from acute radiation injury. Weconclude that the maximum vascular dose does not exceed over 12 Gy to avoid thedelayed radiation injury, especially, it should be limited under 10 Gy in the case thattumor exists in speech center.4:55 PM -Pathological analysis of the effects of BNCT: Markeddecrease of Ki-67 labeling index and preservation ofHealthy Neurons. Shinji Kawabata a, *, Shin-Ichi Miyatake a , FumiharuAkai b , Nobuhiro Nakagawa b , Yoshinaga Kajimoto a , Kunio Yokoyama a ,Toshihiko Kuroiwa a , Mamoru Taneda b , Shin-Ichirou Masunaga c , YoshinoriSakurai d , Akira Maruhashi d , Yoshio Imahori e , Mitsunori Kirihata f , Koji Ono ca Department of Neurosurgery, Osaka Medical College, Takatsuki, Osaka, b Departmentof Neurosurgery, Kinki University, Sayama, Osaka, c Radiation Oncology ResearchLaboratory, Kyoto University Research Reactor Institute, Kumatori, Osaka, dDivision of Radiation Life Science, Kyoto University Research Reactor Institute, Kumatori,Osaka, e Department of Neurosurgery, Kyoto prefectual university of Medicine,Kyoto, f Department of Agriculture, Osaka prefecture university, Sakai, Osaka,JapanIntroduction: We experienced sufficient initial tumor reduction effects of boron neutroncapture therapy (BNCT) for malignant gliomas. However, in some cases we also experiencedaggravation of the radiographic image after the treatment. To analyze theeffects and its mechanism of BNCT on malignant gliomas, we investigated the pathologicalchanges before and after BNCT of human glioma patients.Materials and Methods: Nineteen patients of malignant gliomas were treated 22 timeswith BNCT in Osaka Medical College and Kinki University Hospital with the same protocol.Almost all cases showed radiographic improvement as initial effects, however4 patients (3 glioblastoma multiforme, GB and 1 anaplastic oligoastrocytoma, AOA)received recraniotomy for the aggravation in the follow-up MRI. These 4 patients andanother 1 GB patient died from the hepatic disease were analyzed by pathologicalfindings with H&E staining and labeling index.5:15 PM -First BNCT treatment of a skin melanoma in Argentina:dosimetric analysis and clinical outcome. S.J. González a,c, *,M.R. Bonomi b,d , G.A. Santa Cruz a,c , H.R. Blaumann a,e , O.A. CalzettaLarrieu a,e , P. Menéndez b,d , R. Jiménez Rebagliati a,f , J. Longhino a,e , D.B.Feld a,g , M.A. Dagrosa a,g , C. Argerich h , S.G. Castiglia a,i , D.A. Batistoni a,f , S.J.Liberman a,f , B.M.C. Roth b,da Comisión Nacional de Energía Atómica, Av. Del Libertador 8250, (1429) Ciudadde Buenos Aires, Argentina, b Instituto de Oncología Ángel H. Roffo, Av. San Martín5481 (1417), Ciudad de Buenos Aires, Argentina, c UARyCN, Centro Atómico Constituyentes,CNEA, Argentina, d Dpto. de Terapia Radiante, Instituto Roffo, Argentina, eCentro Atómico Bariloche, Río Negro, CNEA, Argentina, f Departamento de Química,Centro Atómico Constituyentes, CNEA, Argentina, g Departamento de Radiobiología,Centro Atómico Constituyentes, CNEA, Argentina, h Hospital Privado Regional, 20de Febrero 598, S. C. de Bariloche, Río Negro, Argentina, i UAR, Centro AtómicoEzeiza, CNEA, ArgentinaPurpose: This article reports on the first BNCT treatment in Argentina, which initiatesthe phase I/II clinical trials for skin melanomas, designed by the Comisión Nacionalde Energía Atómica and Instituto Roffo BNCT group. The protocol-based procedure,treatment regime, treatment planning and dosimetry, patient irradiation, retrospectiveanalysis and clinical outcome are described.Materials and Methods: A 54-year-old woman, affected by a skin melanoma with multiplenodes progression on the external face of her right thigh, was the subject treatedby BNCT at the RA-6 reactor, Bariloche, Argentina. The treatment regime involvesintravenous administration of boronophenylalanine (BPA) at a dose of 14 g m-2 infusedover 90 minutes. A maximum normal skin dose of 16.5 RBE Gy and a minimumdose for tumor control of 20 RBE Gy are assumed. A pre-planning dosimetry wasaccomplished based on the results of a biodistribution performed on the same patienttwo months before the irradiation. Samples of blood and tissues were analyzed forboron content by inductively coupled plasma atomic emission spectrometry. Patientpositioning was determined in a simulation room in Buenos Aires, that reproduces thegeometry of the irradiation room. Subsequently, a planning CT study was performed,reproducing the same position by means of radio-opaque fiducial markers and positioningdevices. Dose calculations and treatment plan were obtained using NCTPlan v.1.3 and the Dose-Volume Histogram tool developed by the BNCT treatment planninggroup at CNEA. An open 2-comparment model was used to fit blood boron concentrationvalues. On October 9, 2003, the patient was treated. Blood samples were takenduring infusion and patient positioning, and after treatment. The total irradiation timewas about 50 minutes.Results: The retrospective analysis revealed that the actual maximum and averageskin doses were 15 and 7.4 RBE Gy respectively. Eight weeks after BNCT, 21 out of25 nodules included within the isodose curve of 15 RBE Gy were in complete clinicalresponse. The tumor response was assessed clinically by inspection and palpationby three different radiation oncology physicians. A grade 1 RTOG/EORTC skin acutereaction was observed and was in resolution after eight weeks.Conclusion: The first melanoma treatment by BPA-based BNCT was performed successfullyin Argentina, with low toxicity and good clinical response. The same patientreceived a second BNCT treatment in a different location of the extremity and theresults are still under evaluation.Results: Case 1 (GB) was received BNCT and additional radiosurgery 1 month afterBNCT for the shortage of the absorbed dose at deepest part of the tumor. 5 monthsafter BNCT the lesion showed aggravation on MRI and recraniotomy was applied forhistological confirmation, which revealed tumor necrosis. The patient showed stabledisease after the recraniotomy. Case 2 (GB) showed marked initial effects by BNCT.The tumor volume showed 71.0% reduction 1 week after BNCT, however the tumorshowed regrowth 2 month after BNCT. Recraniotomy showed viable tumor cellsamong the necrosis. Labeling index analysis measured by Ki-67 monoclonal antibodywas applied to these patients before and after BNCT. The first craniotomy revealed10 to 25% positivity of labeling index, while recraniotomy showed less than 0.5% of it.Histology of the recraniotomy of case 3 (GB) showed interestDiscussion and Conclusion: Histology obtained from the stable disease after BNCTshowed almost all necrotic tissue. Labeling index analysis revealed marked inhibitionof DNA synthesis by BNCT. Some viable tumor cells in the necrotic tissue may beascribed to uneven distribution of the boron compound in the tumor tissue. Marginalrecurrence occurred from the deeper part of the tumor, which seemed to be shortageof the neutron flux and limitation of this treatment. Selective or specific tumor cell killingof BNCT were observed by the histological analysis. This selective killing of BNCT wasalso observed by magnetic resonance spectroscopic analysis.10

11Eleventh World Congress on Neutron Capture TherapyMonday, October 11, 2004 - PMParallel Session 2 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sSession Chairs: Andres Kreiner, Tom Blue2:00 PM -Microdistributions of prospective BNCT-compoundCuTCPH in tissue sections with a heavy ion microbeam.P. Stoliar a , A.J. Kreiner a,b,c, *, M.E. Debray a,b , M.E. Caraballo a , A.A. Valda b , J.Davidson a,c , M. Davidson a,c , J.M. Kesque a , H. Somacal b , H. DiPaolo a , A.A.Burlon a,b , M.J. Ozafrán a , M.E. Vázquez a , D. Minsky a,b , E.M. Heber d , V.A.Trivillin d , A.E. Schwint c,da Departamento de Física, Argentine Atomic Energy Commission, CNEA, Av. Gral.Paz 1499, CP 1650, Bienos Aires, Villa Martelli, Argentina, b Escuela de Ciencia yTecnología, Universidad de San Martín, Villa Ballester, Argentina, c CONICET, BuenosAires, CP 1917, Argentina, d Departamento de Radiobiología, CNEA, Av. Gral.Paz 1499, CP 1650, Villa Martelli, ArgentinaIntroduction: The success of BNCT depends on our ability to deliver boron to tissuesusing one or more boron compounds to exert a therapeutic effect on tumor with scarcedamage to normal tissue. The efficacy of a boron compound will depend on its localization.Thus, the need for information on the microdistribution of the boron-carrier attissue and cellular level is critical in terms of testing new drugs, evaluating their therapeuticpotential and optimizing drug delivery strategies. The aim of the present studywas to evaluate the microdistribution of the carborane-containing tetraphenylporphyrinCuTCPH with one Cu atom in its molecular structure.Materials and Methods: The analytic technique employed in the present study is knownas micro-PIXE, namely Particle Induced X-ray Emission with beams of micrometer dimensions.It is based on the detection of characteristic X-rays of the elements in thesample, following the irradiation with a focused 16O ion beam. Focusing is performedwith a high-precision magnetic quadrupole triplet on freeze-dried cryosections of tumorand normal tissue from hamsters injected with the drug and of non-injected controls.Since B X-rays are extremely difficult to measure, advantage has been taken of thepresence of Cu in the porphyrin molecule. Exophytic tumors were induced on theright Cheek Pouch of Syrian Hamsters (HCP) in keeping with the HCP carcinogenesismodel of oral cancer previously validated for BCNT research. Samples from tumor,precancerous tissue surrounding tumor, normal contralateral pouch and liver (both forinjected and non-injected animals as controls) were taken.Results: Two-dimensional maps of elemental concentration were obtained by scanningthe beam over the sample. The results for Cu can be translated into B concentrations(the B/Cu mass ratio in CuTCPH is 6.805). The average B concentration in a tumorsample over a large 2 x 2 mm2 field was 116±46 ug/g with “hot spots” of up to 422ug/g. Maximum and minimum concentrations in a series of 1 x 1 mm2 tumor sectionswere 184 and 68 ug/g. Concentration in liver was high (694 ug/g). For precanceroustissue we obtained two widely different values for different regions (13 and 299 ug/g ata hot spot). A value for a section of normal contralateral HCP tissue was 35 ug/g.Discussion: Very inhomogeneous concentrations were found in all sections (exceptin liver tissue). The Cu (and hence the CuTCPH and the associated B to the extentthat the drug is stable) localizes preferentially in tumor as compared to normal tissue.Within tumor, it appears to localize selectively in tumor stroma (connective tissue)rather than in tumor parenchyma. In particular, it may be associated to blood vessels.Conclusions: The microPIXE technique, as implemented with an 16O microbeam, hasbeen shown to be a powerful analytical tool for the study of multielemental distributionsin histological sections at the few ug/g level. In particular its application for thedetermination of microdistributions of a drug of possible relevance for BNCT has beendemonstrated. CuTCPH appears to be localized preferentially in tumor stroma.2:20 PM -Improvement of the high-resolution alpha auto-radiographywith contact UV microscopy. Y. Kajimoto a , K. Amemiya a ,H. Takahashi b , M. Nakazawa a , Y. Nakagawa c , H. Yanagie d , T. Hisa d , M.Eriguchi d , T. Majima e , T. Kageji f , S. Miyatake g , S. Kawabata g , Y. Sakurai h , T.Kobayashi h , N. Yasuda i , M. Kagawa j , K. Ogura ka Department of Quantum Engineering and Systems Science, The University of Tokyo,Tokyo 113-8656, Japan, b Research into Artifacts, Center of Engineering, TheUniversity of Tokyo, Tokyo 153-8904, Japan, c Department of Neurosurgery, NationalKagawa Children’s Hospital, Kagawa 765-8501, Japan. d Research Center for AdvancedScience and Technology, The University of Tokyo, Tokyo 153-8094, Japan, ePhotonics Research Institute, National Institute of Advanced Industrial Science andTechnology, Ibaraki 305-8568, Japan, f Department of Neurological Surgery, TheUniversity of Tokushima, Tokushima 770-8503, Japan, g Department of Neurosurgery,Osaka Medical College, Osaka, 569-8686, Japan, h Research Reactor Institute,Kyoto University, Osaka 590-0494, Japan, i Research Center for Radiation Safety,National Institute of Radiological Sciences, Chiba 263-8555, Japan, j Department ofPathology, Osaka Medical College, Osaka 569-8686, Japan, k College of industrialTechnology, Nihon University, Chiba 275-8575, JapanIn boron neutron capture therapy (BNCT), it is significant to know the microdistributionof a boron drug inside a cell to evaluate efficacy of that drug, because chargedparticles from 10B(n, alpha) reaction need to hit nucleus to bring tumor cells death.We have developed a novel high-resolution neutron induced alpha-autoradiography(NIAR) with contact ultra-violet (UV) microscopy for the measurement of microdistributionof boron compounds. In this technique, sectioned biological specimens mountedon CR-39 plastic track detectors are irradiated by thermal neutrons. At this stage,tracks for alpha and lithium particles from boron-neutron reaction are recorded alongtheir paths in the CR-39 plates. Those tracks in cellular histology on the CR-39 areimaged using contact UV-ray microscopy. In contact UV-ray microscopy, specimensmounted on the CR-39 are exposed to UV from low-pressure mercury lamp whosepeak wavelength is 254nm. This wavelength region, photons are mainly absorbedby nucleic acid, therefore transmission UV image of mounted sample is recorded onthe CR-39 surface. After etching process, alpha/lithium particles tracks are etchedinto conical shaped holes (etch pits); on the other hand, cell images are appeared asrelief on the CR-39 surface. The cell images and alpha/lithium tracks can be observedsimultaneously with AFM at high resolution, therefore we can know where 10B are atthe intercellular structure level.In the high-resolution NIAR, there are two points that must be discussed: imagingresolution and reliability of sample preparation. In this study we have investigatedthe high-resolution NIAR with contact UV microscopy technique from those points ofview.To investigate achievable resolution of the NIAR, we used the biological specimensprepared in the same way of electron microscopy; several kinds of tissues dissectedout from tumor-transplanted rats are fixed by glutaraldehyde and osmium tetroxide,dehydrated and then embedded in epoxy resin. Those specimens were sliced intovarious thick sections with ultramicrotome (200 nm – 1 mm). The resolution of contactUV microscopic image of cells was improved better by slicing sections thinner. Anotherexperiments using standard sample revealed that the resolution of contact UV imagingwas < 100 nm. That is sufficient to resolve fine structures of cellular organelles.Using samples prepared for electron microscopy, there is a possibility of the redistributionof boron drugs in fixation and hydration because boron compound drugs aresoluble in water. For more reliable measurements, we also used the sections of rapidfreezing fixation specimens. When specimens are frozen rapidly and are freeze-dried,the chance of redistribution will be reduced. We applied our NIAR technique with contactUV-ray microscopy to rapidly frozen specimens, and observed high-resolution cellimages and alpha tracks simultaneously. The track counts and boron concentration ofstandard concentration samples showed good liner correlation. We will improve thequality of sections in order to obtain better cell image.These results promise the high-resolution boron mapping in cellular histology withreliability.2:40 PM -The microdosimetry of boron neutron capture therapyin a ellipsoidal cell geometry. T. L. Nichols a, *, L. F. Miller a , andG. W. Kabalka ba The University of Tennessee, Department of Nuclear and Radiological Engineering,Knoxville, Tennessee, 37996, U.S.A., b The University of Tennessee, Departments ofChemistry, and Radiology, Knoxville, Tennessee, 37996, U.S.A.Two reactions deliver the majority of local dose in boron neutron capture therapy(BNCT). The ionized particles (protons, alpha particles, and lithium nuclei) producedin the two reactions, 10B(n,a g)7 Li and 14N(n,p) 14O, have short ranges that are lessthan ~14mm (which is the order of a cell diameter of a typical human cell). The ionizedparticles are heavy and in the 2+ charge state in the case of the boron reactions.These heavy 2+ ions will do significant damage to molecules near their tracks. Thus,the distribution of nitrogen and, in particular, of boron determines the spatial characteristicsof the radiation field. Since the distribution of nitrogen is nearly homogeneous inthe brain and is not easily altered for the purpose of radiotherapy, the spatial variationin the radiation dose is due mainly to the spatial distribution of boron. This implies thatthe spatial distribution of boron determines the microscopic energy deposition andtherefore the spatial characteristics of the microscopic dose. The microscopic dosefrom the (n,a) and (n,p) reactions have been examined in detail and as averred, theproton dose is relatively homogeneous except for statistical variability. The statisticalvariability in essence adds a false spatial variability that would not be seen if a largenumber of histories were performed. Since the majority spatial variability in the borondistribution, the (n,p) reaction may be suppressed to better understand the spatialdistribution effects on the microscopic dose. Programs have been written in FORTRANusing Monte-Carlo techniques to model ellipsoidal cells that are either randomly sizedand located in the region of interest or the cells are arranged in a face centered cubicand are identical except for the location of the nuclei which may be random. It is then

Eleventh World Congress on Neutron Capture TherapyMonday, October 11, 2004 - PMParallel Session 2 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sbe shown that closely packing prolate ellipsoidal cells with a large eccentricity in onedimension will receive a larger nuclear doses than those cases in which the cells aremore sparsely packed. This is shown to demonstrate that the boron content of a celland its’ nucleus can have a significant impact upon the dose to neighboring cells. Thelocal boron distribution in a region of interest can be shown to impact the macrodosimetricdose with possible implications for clinical outcomes.3:00 PM -BNCT dose calculation in irregular fields using the sectorintegration method. H.R. Blaumann a, *, D.E. Sanz b , J.M. Longhinoa , O.A. Calzetta Larrieu aa Centro Atómico Bariloche, S. C. de Bariloche, Río Negro, Argentina, b FundaciónEscuela de Medicina Nuclear, Mendoza, ArgentinaIntroduction: Irregular fields for Boron Neutron Capture Therapy (BNCT) have beenalready proposed with the aim to spare normal tissue in the treatment of superficialtumors. Preliminary calculations using Monte Carlo techniques show that the neutronthermal flux entering the patient can have a strong dependence on field size andshape. This added dependence would require individual measurements prior to treatmentin order to have a redundant result thus incrementing the confidence of the deliveredtreatment. Since these measurements are very time-consuming and complex itis appealing to develop a secondary calculation system.In this work we apply, from first principles, the sector-integration method for dose calculationin irregular fields in a homogeneous medium and on the central beam axis.This is an extension of the Clarkson’s method used in external beam radiotherapy. Thevarious dosimetric responses produced by the irregular field, i.e. the fast neutron dose,the photon dose and the thermal flux, are calculated by sector integrating the relatedmeasured responses corresponding to circular fields over the field boundary.Material and Methods: The experimental work was carried out at our BNCT facility,in the RA-6 reactor at Centro Atómico Bariloche, in Bariloche, Argentina, which hasrecently begun to be used for human clinical trials applied to malignant melanomas.To apply the model, circular field input data were acquired from dose and neutron fluxmeasurements along the central axis in a full scatter water phantom, in contact withthe beam port. Each port set up was defined by a shielding plate with a different innerdiameter inserted in the beam outlet thus defining different circular fields. Shieldingplates were designed to reduce the 15 cm outer beam diameter by simulating thebeam port with the Monte Carlo MCNP-IV-C code in order to assess its compositionand thickness. A plate with a rectangular aperture was used as test field and the associateddata were measured. Dosimetric responses were measured by using the foilactivation method and the paired ionisation chambers technique for each shieldedbeam configuration, corresponding to circular fields with radii ranging from 0 to 7.5 cm,at different depths in the water phantom.Given the currents of both ionisation chambers and the thermal flux for the circularfields, the same quantities were calculated for the rectangular field shape by the sectorintegration method. Physical doses were calculated for the rectangular field, from thevalues of the chambers currents and thermal flux determined by the sector-integrationmethod and by direct measurements. Equivalent doses were calculated by using appropriateRBEs values and boron concentration for the case of skin melanomas.Results: Differences between the direct measurements and the sector integrationmethod are less than 3% in equivalent dose at any depth.Discussion and Conclusions: This result indicates that the tool is suitable as a redundantcalculation system. A further step is the extension of the method to off axis points,fact that is under current research.3:20 PM -Brain Equivalent Plastic for Ephitermal Neutorn BeamDosimetry. K. J. Riley 1 , P. J. Binns 1 and O. K. Harling 21 Nuclear Reactor Laboratory, Massachusetts Institute of Technology, Cambridge, MA02139, USA, 2 Dept. of Nuclear Engineering, Massachusetts Institute of Technology,Cambridge, MA 02139, USAClinical studies of neutron capture therapy (NCT) at the Massachusetts Institute ofTechnology (MIT) have focused on intracranial metastatic melanoma and glioblastomato establish the absorbed dose required with BPA mediated therapy to reach normaltissue tolerance of the brain. This entails a determination of the absorbed dose to braintissue that has an elemental composition significantly different from the A-150 plasticcommonly used as wall material in ionization chambers for neutron dosimetry.Dosimetry of the neutron beams employed in NCT requires an assessment of thethermal flux as well as the fast neutron and photon absorbed doses as each componentcan possess a markedly different relative biological effectiveness (RBE). This isaccomplished at MIT using a combination of passive and active dosimeters. To assessthe total (thermal plus fast) neutron dose component the twin chamber method is utilizedwith separate graphite and A-150 walled ionization chambers. The fast neutroncomponent is determined from the total neutron dose by subtracting the kerma associatedwith thermal neutrons determined from flux measurements.Low energy neutrons have little energy to impart through elastic scattering and theensuing tissue kerma principally results from capture reactions with nitrogen that consequentlydepends upon the elemental nitrogen concentration in the tissue of interest.The nitrogen content of muscle equivalent A-150 plastic is 3.5% (by weight) asopposed to 2.2% for brain. To avoid the corrections necessary for relating the totalmeasured absorbed dose in A-150 plastic to kerma in brain, a new tissue equivalentplastic for brain (A-181) was previously formulated. The feasibility of utilizing this materialfor routine calibrations under reference conditions pertinent to clinical irradiationsin an epithermal neutron beam was investigated.A commercial ionization chamber was constructed with walls made of the brain equivalentplastic and a methane based flush gas with a matching nitrogen composition wasprepared. The sensitivity of the A-181 walled ionization chamber relative to photonsfor all neutrons in a clinical epithermal beam was calculated to vary between 0.79 ±0.04 in-air and 0.95 ± 0.01 at depths of 4 cm and greater in-phantom. Differences inthe total neutron doses measured with A-150 and A-181 plastic walled chambers wereattributed, within experimental error, to the dose produced by thermal neutron capturereactions from the different concentrations of nitrogen in the two tissue substitutes.The response of the A-181 chamber was converted to total neutron dose with an uncertaintyincreasing with depth in-phantom from 13 to 23% the magnitude of which isdetermined by a relatively large photon dose subtraction. The use of A-181 in place ofA-150 plastic will no longer require partitioning the measured neutron dose by energyand should simplify dose reporting in BNCT.3:55 PM -Benchmark experiments for cyclotron-based neutronsource for BNCT. S. Yonai a, *, T. Itoga a , M. Baba a , T. Nakamura a , H.Yokobori b , Y. Tahara ca Cyclotron and Radioisotope Center, Tohoku University, Aoba, Aramaki, Aoba-ku,Sendai 980-8578, Japan, b Advanced Reactor Technology Co, Ltd., 15-1, Tomihisa,Shinjuku-ku, Tokyo, Japan, c Mitsubishi Heavy Industries, Ltd., 3-1, Minatomirai 3-chome, Nishi-ku, Yokohama, JapanIntroduction: Accelerator-based neutron sources have not been realized yet in BNCTapplications, mainly because they require a very high beam current from an acceleratorwhich introduces difficulty in target cooling. In the previous study, we found thefeasibility of a cyclotron-based BNCT using the Ta(p,n) reaction bombarded by 50MeV protons of 300 uA by simulations using the MCNPX code code, which brings anadvantage in target cooling. In order to realize the cyclotron-based BNCT, it is necessaryto validate the simulations by the measurements. Here in this study, we constructedthe epithermal neutron field at CYRIC (Cyclotron and Radioisotope Center),Tohoku University, which simulates the treatment field for the cyclotron-based BNCTinvestigated in the previous study. We measured the neutron energy spectrum of thisfield by using a newly-developed multi-moderator spectrometer and the distribution ofthermal neutrons in an acrylic phantom by using the gold activation foils through the197Au(n,gamma)198Au reaction.Experimental Method: The measurements were performed at the TOF (Time-Of-Flight)room, which has very low background, at CYRIC. The neutrons were produced fromthe 3 mm thick (stopping-length) Ta target bombarded at an angle of 90 degree by 50MeV protons, and were extracted to the TOF room through the first and second collimators.The epithermal neutron field was constructed by assembling 30 cm thick iron,45.8 cm thick AlF3, 12.1 cm thick Al and 1.4 cm natLiF, in this order. We measured theneutron energy spectrum behind the moderator with our new multi-moderator spectrometerwhich has good sensitivity for epithermal neutrons of energies form 1 eV to 1MeV. This spectrometer consists of a spherical 3He counter surrounded with sphericalinner and outer polyethylene moderators inserting a 1 mm thick silicon rubber loadedwith natB between the polyethylene moderators. The measurement of the depth distributionof the reaction rates of 197Au(n,gamma)198Au in a acrylic phantom of 30 cm *30 cm * 30 cm cm was performed by the gold activation technique.Simulation Method: The calculations to compare with the measurements were performedusing the MCNPX code with the LA150 cross-section data library for the Ta(p,n)neutrons source. The calculational geometry included floor, ceiling, walls, tables formoderators, spectrometer and phantom faithfully.Results and Discussions: The calculations of neutron counts of the spectrometeragree with the measurements within ~10 %. Good agreement between the calculationand the measurement of neutron energy spectrum can be obtained within ~10 %. Thecalculations of 197Au(n,gamma)198Au reaction rates in the phantom agree with themeasurements within ~20 %.12

Eleventh World Congress on Neutron Capture TherapyMonday, October 11, 2004 - PMParallel Session 2 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sConclusion: The results obtained in this benchmark experiment indicate that wecan validate the accuracy of the calculation of the neutron energy spectrum passingthrough the moderator and the thermalization in a phantom. As the next step to realizethe BNCT at CYRIC, we have a plan to measure the absorbed dose distributionin a phantom with the same experimental arrangement, and perform the mock-upexperiment by fabricating an irradiation assembly for BNCT which includes the coolingsystem for the neutron-producing target.4:15 PM -Optimization of an accelerator-based epithermal neutronsource for neutron capture therapy. O.E. Kononov a, *,V.N. Kononov a , M.V. Bokhovko a , V.V. Korobeynikov a , A.N. Soloviev a , A.S.Sysoev a , I.A. Gulidov a , W.T. Chu b , D.W. Nigg ca State Scientific Center of Russian Federation, Institute for Physics and Power Engineering,Bondarenko sq. 1, 249033 Obninsk, Kaluga, Russia, b Lawrence BerkeleyNational Laboratory, Berkeley, USA, c Idaho National Engineering and EnvironmentalLaboratory, Idaho Falls, ID, USAAbstract: A beam shaping assembly optimization for epithermal neutron beam productionon accelerator based facility for neutron capture therapy are presented. Optimizationobjective is to cerate neutron beam with required for BNCT energy and intensity.Results of calculation were experimentally tested and are in good agreement withmeasurements.Materials and methods: For last ten years active discussed and investigated problemof neutron source creation for neutron capture therapy based on non-expansive protonaccelerator with beam energy 2-3 MeV and power 10-20 kW, which is possible toinstall in oncology clinics. In this paper presented investigation on chousing optimalmoderator material, its size optimization, calculation and measurement epithermalneutron beam characteristics on an output port of moderator assembly. Main investigationswere carried out for initial proton energy 2.3, 2.4, 2.8 MeV. Maximum neutronenergy in this case is 0.6-1 MeV and total calculated neutron yield 6,3_1012, 8,1_1012_ 1,37_1013 neutron per second for beam current 10 mA. Neutrons and gamma raystransport were calculated by using Monte Carlo computer codes S-95NCT and MCNP.As criteria for chousing material and optimum moderator size was taken 2 parameters- fepi – epithermal neutron leakage flux density (neutron energy more then 1 eV) fromsphere surface for proton current 10 mA and - neutron and gamma rays dose rate intissue per one epithermal neutron.Results: Presented results obtained in investigations for mostly prospective moderators– heavy water, magnesium fluorine, polytetrafluoroethylene and Fluental®.Shown, that the best characteristics for epithermal neutron beam creation has magnesiumfluorine. Optimized magnesium fluorine moderator thickness is 20 cm. Moredetail epithermal neutron beam characteristic was derived in calculation in-phantomabsorbed dose distribution. In terms of absorbed dose distribution rate value in tumor/healthytissue beam shaping assembly from magnesium fluorine in 1.25 timesmore effective then polytetrafluoroethylene and 1.8 times then Fluental®. For directverification computer modeling was performed thermal neutron flux measurementsinside water phantom. Thermal neutron beam flux measurement was performed bygold foil activation method. Calculation results and experimental measurements arein good agreement.Conclusion: By calculation in-phantom dose distribution shown, that with such moderator,2.3 MeV and 10 mA proton beam advanced depth is 9 cm, therapeutic ratio ondepth 3 cm is 6, advanced depth dose rate on depth 9 cm is ~1 RBE Gy per minute,which means that maximum treatment time will be about 10 minutes.4:35 PM -A shielding design for an accelerator-based neutronsource for boron neutron capture therapy. A.E. Hawk, T.E.Blue*, J.E. WoollardNuclear Engineering Program, The Ohio State University, 206 West, 18th Avenue,Columbus, Oh 43210-1189, USAResearch in boron neutron capture therapy(BNCT) at The Ohio State UniversityNuclearEngineering Department has been primarilyfocused on delivering a high qualityneutronfield for use in BNCT using an accelerator-based neutron source (ABNS).An ABNS for BNCT is composed of a proton accelerator, a high-energybeam transportsystem, a 7Li target, a target heat removal system (HRS), a moderator assembly,and a treatment room. The intent of this paper is to demonstrate the advantages of ashielded moderator assemblydesign, in terms of material requirements necessarytoadequatelyprotect radiation personnel located outside a treatment room for BNCT,over an unshielded moderator assemblydesign.4:55 PM -An optimized neutron-beam shaping assembly for accelerator-basedBNCT. A.A. Burlon a, *, A.J. Kreiner a,b,c , A.A. Valda a ,D.M. Minsky a,ba Escuela de Ciencia y Tecnología, Universidad de San Martín, Alem 3901, 1653 VillaBallester, Argentina, b Departamento de Física, CNEA, Av. Gral. Paz 1499, CP 1650Villa Martelli, Argentina, c CONICET, ArgentinaIntroduction: An easy to build and relatively cheap beam shaping assembly is studiedhere through an extensive MCNP investigation. The solution proposed consistsof successive stacks of Al, polytetrafluoroethylene (PTFE) and LiF as moderator andneutron absorber, and Pb as reflector. The 7Li(p,n)7Be reaction and proton bombardingenergies of 1.92, 2.0, 2.3 and 2.5 MeV have been considered for three moderatorthicknesses (18, 26 and 34 cm).Materials and Methods: In the geometry of the simulations, a whole-body phantom(with Snyder’s head model) was considered. The doses were evaluated within thehead phantom along its centerline. A 71.6 cm diameter lead reflector and a 15 cmdiameter Al/PTFE/LiF moderator were considered. Also, a thermal neutron shield wasplaced at the exit of the moderator and reflector. A 10B concentration of 40 ppm in tumorand 11.4 ppm in healthy tissue were adopted. The treatment times were calculatedassuming a 20 mA proton beam. In a second instance, the effect of the specific skin radiosensitivityand its 10B uptake were considered for the scalp. The dose assessmentwas performed through the calculation of Tumor Control Probabilities (TCP).Results: The TCP curve is a function of the total tumor RBE-dose. Here, we translatedthe total tumor RBE-dose axis into a maximum total healthy tissue RBE-dose axisfrom our simulated results. From these TCP curves it is possible to evaluate the treatmentcapability for a particular beam shaping assembly knowing the maximum dosedelivered to the healthy tissue and the treatment time. Figures of merit showing themaximum healthy tissue dose and treatment times (for a 98% TCP) were plotted forvarious beam shaping assemblies (in both cases, with and without the specific characteristicsof the scalp).Discussion: For 1.92 MeV protons, the maximum healthy tissue doses are the lowestfor most of the positions in the brain, but the treatment times are excessively long. Forthe 2.0 MeV case, the maximum healthy tissue doses for relatively deep tumors staybelow 12.5 RBE-Gy but with treatment times from one to about seven hours. For theresonance case (2.3 MeV) several points stay under the 12.5 RBEGy line and theshortest treatment times are near an hour. At 2.5 MeV, the best performance correspondsto the 34 cm moderator and treatment times less than 45 minutes (for tumorsalmost as deep as 5 cm). When the specific radiosensitivity and 10B uptake of the skinare assigned to the scalp the figures of merit show drastic changes. Nevertheless,some useful points remain under the 12.5 RBE-Gy limit (particularly for 2.3 MeV protons)and the treatment times are still reasonable (between 40 and 60 minutes).Conclusions: The MCNP simulations show an acceptable behavior of Al/PTFE/LiF andPb as beam-shaping assembly. They also show the advantage of irradiating the targetwith near-resonance-energy protons (2.3 MeV) because of the high neutron yield atthis energy and yet sufficiently small fast neutron production, leading to lowest treatmenttimes.5:15 PM -Lithium neutron producing target for BINP acceleratorbasedneutron source. B. Bayanov a , V. Belov a , V. Kindyuk b , E.Oparin c , S. Taskaev a, *a Budker Institute of Nuclear Physics, 11 Lavrentiev avenue, 630090 Novosibirsk, Russia,b Novosibirsk State University, 2 Pirogova str., 630090 Novosibirsk, Russia, cNovosibirsk State Technical University, 20 Marx avenue, 630092 Novosibirsk, RussiaThe Budker Institute of Nuclear Physics and the Institute of Physics and Power Engineering,Obninsk, have proposed an accelerator based neutron source for neutroncapture and fast neutron therapy at hospital. Innovative approach is based upon tandemelectrostatic accelerator with vacuum insulation and near threshold neutron generation.Pilot facility is under construction now at the BINP. One of the main elementsof the facility is lithium target, that produces neutrons via threshold 7Li(p,n)7Be reactionat 25 kW proton beam with energies 1.915 MeV or 2.5 MeV.In the present report, results of experiments and simulations on neutron producingtarget prototype are presented, choice of target for the source under construction issubstantiated, the necessary experiments are shown, and the conception of the targetis presented.The first model of neutron producing stationary target was tested under 20 kW electronbeam. This target consisted of 0.2 mm thick molybdenum foil, that diffusely welded onan ARMCO steel disk. Ten rectangular grooves were on the disk for cooling by water13

Eleventh World Congress on Neutron Capture TherapyMonday, October 11, 2004 - PMParallel Session 2 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sor liquid gallium. In the process of examination, heat removal up to 650 W cm–2 wasprovided using water, and liquid metal cooling resulted in the target destruction due tohigh chemical interaction of gallium with ARMCO steel.A new target prototype was made consisting of tungsten disk 80 mm in diameter, 3mm thick with thirteen cooling rectangular channels, pressed to titanium body withoutdiffuse welding. Laborious diffuse welding was refused, which allowed to obtain morehomogeneous temperature field on the surface of the target. Hydraulic resistance forheat carrier flow in the target and lithium layer temperature are calculated. Experimentsare planned for the near future to study hydraulic and thermal regimes of targetprototype. It is clear now that using water is possible for cooling of this target. This allowsto refuse using gallium for cooling the target, therefore not to solve the problemsof corrosion of target material and pump, arising from gallium influence.Neutron producing target for the neutron source under construction is proposed tocorrespond to the existing prototype with the following fundamental modifications: i)target diameter is to be increased up to 10 cm for 25 kW proton beam, ii) it will becooled with water only, iii) to provide efficient cooling at minimal water consumption,the target channels are to be spiral, iv) lithium layer is to be evaporated immediately inthe target unit. Calculation showed that the lithium target could run up to 10 mA protonbeam before melting. The material for proton beam absorber is to be determined afterplanned experiments on radiation blistering at the existing target prototype under availablepulse proton beam. Diagnostic equipment provides detecting a-particles insidethe vacuum chamber close to the target and detecting neutrons and g-rays outside.Other promising types of neutron producing target are also under development.Manufacturing the neutron producing target up to the end of 2004 and obtaining a neutronbeam on BINP accelerator based neutron source are planned during 2005.14

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - AMDosimetry - N u c l ea r E n g i n e e r i n g , P hy s i c s , & C l i n i ca l A p p l i ca t i o n sSession Chairs: Milan Marek, Stuart Green8:30 AM -In-phantom characterisation studies at the BirminghamAccelerator-Generated epIthermal Neutron Source(BAGINS) BNCT facility. Christopher N. Culbertson a , Stuart Green b, *,Anna J. Mason a , David Picton a , Gareth Baugh a , Richard P. Hugtenburg a,b ,Zaizhe Yin a , Malcolm C. Scott a , John M. Nelson aa School of Physics and Astronomy, University of Birmingham, Birmingham B15 2TT,UK, b Department of Medical Physics, University Hospital Birmingham, Queen ElizabethHospital, Edgbaston, Birmingham B15 2TH, UKThis revised paper is presented posthumously under the name of the originallead author Dr. Chris Culbertson. It is dedicated to his memory.A broad experimental campaign to validate the final, constructed epithermal neutronbeam for BNCT at the University of Birmingham concluded in November 2003. Thefinal moderator and facility designs are overviewed briefly, followed by a summary ofthe dosimetric results from this campaign. The dual ionisation chamber technique wasused together with foil activation to quantify the fast neutron, photon, and thermal neutronbeam dose components along the beam-axis and perpendicular to the beam-lineaxis at two depths in a large rectangular phantom exposed to the beam with a 12 cmdiameter beam delimiter in place.The Dynamitron accelerator at the University of Birmingham has demonstrated acceleratedproton currents in excess of 1 mA at a potential of 2.8 MV. The proton beamstrikes a thick natural lithium target positioned at the centre of a moderating structure,generating a neutron source intensity of 1.37x10^12n/s at 1 mA via the 7Li(p,n)7Bereaction. Designs utilizing this source for epithermal beam production have been proposedand investigated through simulations and experimental measurements. Thecompleted and commissioned facility design applies a region of FluentalTM to moderatethe neutron spectrum to appropriate therapy energies and a lithiated polyethylenesheet to provide a delimited neutron beam port. An impingent flow, heavy water coolingsystem is used to cool the copper backing plate of the thick lithium target, preventingliquification of the horizontally aligned target. This heat removal system has demonstratedadequate heat removal at up to 1 mA operation.Dose measurements agree with in-phantom MCNP4C predictions within 10% for thephoton dose, within 10% for thermal neutron dose, and within 25% for the protonrecoil dose along the main beam axis. The results of this measurement are comparedwith calculation and summarized in Table 1. Further beam studies utilized fission andionisation chambers for in-air scans of thermal neutron fluence and photon and fastneutron doses across the front of the moderator and found similar levels of agreementwith calculation. The relative contribution from each component was found to be highlysimilar to facilities already utilized in BNCT clinical trials.This is the first publication of dose characteristics of the intended therapy beam, measuredat clinical beam power. As such it is a unique set of measurements of what weanticipate will be the world’s first clinical accelerator-based BNCT beam.scaling factors to relate the separate dose components between the different institutesare being determined. Preliminary normalization of measured and calculateddosimetry for patients is nearing completion to enable the physical radiation doses thatcomprise a treatment prescription at a host institute to be directly related to the correspondingmeasured doses of a visiting group. This should serve as an impetus forthe direct comparison of patient data although the clinical requirements for achievingthis need to be clearly defined. This may necessitate more extensive comparisons oftreatment planning calculations through the solution of test problems and clarificationregarding the question of dose specification from treatment calculations in general.9:30 AM -Measurement of free beam neutron spectra at eightBNCT facilities worldwide. I. Auterinen a, *, T. Serén a , K. Anttila a , A.Kosunen b , S. Savolainen c,da VTT Processes, POB 1608, FIN-02044 VTT, Finland, b Radiation Metrology Laboratory,Radiation and Nuclear Safety Authority-STUK, POB 14, FIN-00881 Helsinki,Finland, c Department of Physical Sciences, University of Helsinki, POB 64, FIN-00014 Finland, d Helsinki Medical Imaging Center, University of Helsinki, P.O. Box340, FIN-00029 HUS, FinlandIntroduction: As part of the “International Dosimetry Exchange” project the dosimetrygroup from VTT has determined the neutron spectrum in BNCT beams using the multifoilactivation detector method. Measurements have been made on the already closedepithermal BNCT facility at the BMRR of the Brookhaven National Laboratory, on theHFR at JRC in Petten, The Netherlands, on the mixed mode beam at KURRI, Japan,on the fission converter beam (FCB) at MIT, USA, on the epithermal beam of the RA-6facility in Bariloche, Argentina, on the epithermal beam at WSU, USA, on the mixedmode beam at JRR-4 at JAERI, Japan, as well as on the epithermal beam at FiR 1 atVTT, Espoo, Finland.Materials and methods: 10 different detector isotopes, majority of them diluted into Alin order to minimize self-shielding effects, were irradiated both in a bare stack and ina cadmium capsule. Additionally, for the fast neutron activation reactions about 1 gplates of Al and In are irradiated in a Cd-capsule which is placed inside a sphere with10 mm thick walls of 10B giving very efficient shielding against thermal as well as mostepithermal neutrons. After irradiation the induced activities in the foils were measuredwith local gammaspectrometry systems. Saturation activities and reaction rates weredetermined. Beam monitor records were used, when available, to judge the correctirradiation time and to normalize the saturation activities from different irradiations toa nominal value. At each facility the calculated neutron spectrum from a beam modelused locally in the dosimetry work and/or in the treatment planning was used. Thecalculated neutron spectra were adjusted based on the measured reaction rates withthe generalized logarithmic least-squares code LSL-M2.Results: The calculated and adjusted 47-group neutron spectra at the eight facilitiesare presented. Also summary of the calculated and adjusted thermal, epithermal andfast fluxes as well as total neutron doses and fast (E > 10 keV) neutron doses to adultbrain tissue, as defined by ICRU in 1992, is shown.9:10 AM -Progress with the NCT international dosimetry exchange. P.J.Binns a, *, K.J. Riley a , O.K. Harling b , I. Auterinen c , M. Marek d , W.S. Kiger III ea Nuclear Reactor Laboratory, Massachusetts Institute of Technology, 138 AlbanyStreet, Cambridge, MA 02139, USA, b Nuclear Reactor Laboratory and Nuclear EngineeringDepartment, Massachusetts Institute of Technology, Cambridge, MA 02139,USA, c VTT Processes, Technical Research Center of Finland, P.O. Box 1608 FIN-02044 VTT, Finland, d Nuclear Research Institute, Rez 25068, Czech Republic, eDepartment of Radiation Oncology, Beth Israel Deaconess Medical Center, HarvardMedical School, Boston, MA 02215, USAThe international collaboration that was organized to undertake a dosimetry exchangefor purposes of combining clinical data from different facilities conducting neutron capturetherapy has continued since its founding at the 9th ISNCT symposium in October2000. The thrust towards accumulating physical dosimetry data for comparisonbetween different participants has broadened to include facilities in Japan and thedetermination of spectral descriptions of different beams. Retrospective analysis ofpatient data from the Brookhaven Medical Research Reactor is also being consideredfor incorporation into this study to increase the pool of available data. Meanwhile thenext essential phase of comparing measurements of visiting dosimetry groups withtreatment plan calculations from the host institutes has commenced. Host centers fromPetten, Finland and the Czech Republic in Europe and MIT in the USA have appliedthe regular calculations and clinical calibrations from their current clinical studies, togenerate treatment plans in the large standard phantom used for measurements byvisiting participants. These data have been exchanged between the participants andDiscussion: The adjustment ratios are a quality of the calculated spectra and weregenerally moderate. If in the first adjustment strongly inconsistent individual reactionrates were observed, they were omitted from the final adjustment. In all epithermalbeams (except the JAERI mixed beam with a higher thermal flux) some reactionsdisplayed a higher reaction rate with Cd capsule than bare. This is probably due todownscattering of neutrons by the Cd capsule into energy regions with strong resonances.The uncertainty of the determined epithermal flux has been about 4%, of thethermal and fast flux about 10%.Conclusions: The multifoil activation method is a powerful tool to measure the characteristicsof epithermal and mixed mode neutron beams for neutron capture therapy.It can be used at the full intensity of the beams. It gives detailed information on thespectrum. It is limited by the requirement of a calculated multigroup spectrum and alsoit requires a good quality gammaspectrometric system.9:50 AM -Spatial characterization of BNCT beams. M. Marek*, L.ViererblNuclear Research Institute at Rez, plc., Rez 25068, Czech RepublicIntroduction: In the framework of Code of Practice and “Dosimetry exchange” thebeams at MIT (FCB), Petten (HFR), Espoo (FiR), Studsvik and NRI were characterizedwith respect to basic parameters in-air as well as by the dose-depth measurementsin phantoms. The goal of this work was to complete the previous work withcomparison of the space distribution of the epithermal and/or thermal neutrons free in15

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - AMDosimetry - N u c l ea r E n g i n e e r i n g , P hy s i c s , & C l i n i ca l A p p l i ca t i o n sIntroduction: In order to optimize NCT treatment, it is very important to determine thermalneutron flux in specific zones of patients. In this work, a novel method to measureon-line incident thermal neutron flux is presented. It is based on the use of a specialself-powered neutron detector (SPND) that can be placed on or inside the patients.SPNDs are coaxial cylindrical assemblies consisting of a central emitter with a relativelyhigh neutron capture cross section material-, a surrounding insulator, and anexternal conductive sheath. Emitter captures neutrons and emits high-energy electronsthat reach the sheath after going through the insulator. A current proportional tothermal neutron flux can be collected using a current amplifier connecting the emitterto the sheath. Owing to the mixed nature of the reactor irradiation beams, detector designshould provide a very low gamma response compared to the neutron one. One ofthe advantages of this kind of detectors is that the sensitive volume can be significantlysmall-sized, which allows for both localized flux determination and, if required, invasiveuse, for example under the skin or inside the brain. Another advantage is that no highvoltageis required, which is very important for medical applications.Materials & Methods: Two detectors, SPND1 (diameter=1.7mm) and SPND2(diameter=1.9mm), were constructed with graphite-lined and Zircaloy-4 sheath respectively,acrylic isolation, and rhodium emitter (length=12.3mm). They were used tocompare their gamma response using a 60Co beam in a position of 1.5mGys-1. Forevaluating SPND2 neutron response, a third detector, SPND3 (diameter=1.9mm), withZircaloy-4 sheath, acrylic isolation and Zircaloy-4 emitter (length=15mm), was assembled.Both detectors were irradiated in the thermal beam of a local nuclear researchreactor, with a thermal flux of (1.6±0.2)E8 ncm-2s-1 in the irradiation position. Detectorswere connected to a low-noise coaxial cable of 2.5mm external diameter and 14mlength to drive the current to an electrometric amplifier located in a remote position,with no flux. The special design of the electrometer used allows obtaining a very stablereading, with 1fA resolution, by filtering high frequency fluctuations of the input.Results: System gamma sensitivity was (3.6±0.1)E-14A/mGys-1 for SPND1 and(2.0±0.1)E-14A/mGys-1 for SPND2. System thermal neutron sensitivity was(3.2±0.4)E-21A/ncm-2s-1 for SPND2 and (-2±1)E-23A/ncm-2s-1 for SPND3.Discussion: SPND2 showed a lower gamma response than SPND1 due to the betterelectronic compensation between emitter and sheath. Zircaloy-4 was then adopted assheath material.SPND2 presented an adequate neutron response. Its comparison with SPND3 responseleads to the conclusion that most of the signal was originated in the interactionbetween neutrons and rhodium. With these characteristics, values expected usingSPND2 for typical NCT flux levels are: 3pA (Thermal flux=1E9ncm-2s-1) and 0.02pA(1mGys-1) for neutron and gamma currents, respectively. In this case gamma contributionwould be less than 1%.Conclusion: An implantable SPND-based system was developed in compliance withall design requirements (materials, dimensions and sensitivity). It can be used to obtainon-line thermal neutron doses delivered to patients, and to recalculate treatmentparameters for their optimization and correction during irradiation.17

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - B i o l o g y182 - Approach to Magnetic Neutron Capture Therapy. AnatolyA. Kuznetsov, Sergey N. Podoynitsyn, Victor I. Filippov, Lubov Kh.Komissarova, Oleg A. KuznetsovInstitute of Biochemical Physics of RAS, Kosygin st 4 Moscow 119991 RussiaThe method of magnetic neutron capture therapy (MNCT) can be described as a combinationof two methods: magnetic localization of drugs using magnetic targeted carriersand neutron capture therapy itself.Delivery of the magnetic adsorbents suspension into the area of a tumor is made underthe control of angiography, and by its injection into a certain artery that brings bloodto the tumor. The applied external magnetic field (H~1000 Oe, grad H~100 Oe/cm) ona projection of the tumor causes agglomeration of particles of the magnetic adsorbentin fine vessels around the tumor. An active substance, which will destruct the tumor,will be extracted in the area of the tumor from surface of the magnetic ferro-carbonparticles in the process of desorption. Experiments and clinical tests demonstratedthat we can localize the majority of magnetic particles (more than 90%) in the vesselsof tumor using this method.In this work we made two-component ultradispersed particles (UDP), which wereformed from vapours of respective materials. In order to make the necessary nanopowderscontaining Fe and C, we injected the initial coarse-dispersed powder in plasmachemicalelectroarc reactors, into the area of long electroarc (up to 20cm), where itwas completely evaporated. Condensation into the UDP was realized in the area of avortex flow of a cold gas, in which the particles also split up into fractions.In case of magnetic neutron capture therapy (MNCT) we can use any non-toxic boroncompounds as drugs. We tested applicability of two compounds as medications:these were borax (Na2B4O7) and L- boronphenylalanine (BPA) that is traditional forneutron capture therapy. It was impossible to realize sorption of BPA completely becausethe process of desorption was almost totally absent due to bad dissolubility ofthis compound in case of neutral pH. This does not allow to use this form of BPA in theprocedure of MNCT.15% of the total amount of borax was absorbed on magnetic Fe-C particles in case of2% initial concentration of borax in the physiological solution (2g per 100g of the solution).The total amount of boron that was transferred into the solution was 6-8mg. Thusthe results indicate that 0,5 g of such particles in the area of 100g tumor is enough foran effective MNCT.Another researched variant of MNCT realization is introduction of the necessary neutron-capturingatoms of boron directly into the magnetic particles during their manufacturing.Two-componental ultradispersed particles Fe-B (10%) were formed from vaporsof appropriate materials in the inert gas (argon) by the method of plasma recondensingof mixture of powders of two materials. The preliminary results demonstrate that introductionof 0,2g of Fe-B (10%) particles into the area of a 100g tumor is enough foran effective MNCT.It is important to note that both types of the particles have high magnetization andmagnetic homogeneity, allow to form stable magnetic suspensions and have low toxicity.Besides joint and simultaneous application of Fe-C particles for magnetic chemotherapyand magnetic neutron capture therapy is possible.4 - Evaluation of Anti-EGF Receptor Monoclonal AntibodyCetuximab as a Potential Delivery Agent for NeutronCapture Therapy.Gong Wu a , Rolf F. Barth a, *, Weilian Yang a , Madhumita Chatterjee b ,WernerTjarks b , Michael J. Ciesielski c , and Robert A. Fenstermaker ca Department of Pathology and b College of Pharmacy, The Ohio State University,Columbus, Ohio 43210, and c Department of Neurosurgery, Roswell Park CancerInstitute, Buffalo, New York 14263The gene encoding EGFR often is amplified in human gliomas, and the receptor itselfhas been considered as a potential target for the specific delivery of therapeutic agentsto brain tumors. The purpose of the present study was to investigate the use of thechimeric MoAb cetuximab (IMC-C225), which is directed against EGFR and EGFRvIII,as a boron delivery agent for neutron capture therapy (NCT) of brain tumors. As determinedby 125I-cetuximab radioligand binding assays, F98 rat glioma cells, which hadbeen transfected with the gene encoding EGFR (F98EGFR), expressed 1.60±0.13 105receptor sites/cell with a Ka=1.64±0.32 108 M-1. F98 cells transfected with the geneencoding a mutant form of EGFR, designated the F98EGFRvIII glioma, expressed1.07±0.10 105 receptor sites/cell with a Ka=2.18±0.54 09 M-1 compared to backgroundlevels expressed on F98 wild type cells (F98WT). A heavily boronated, 5th generationpolyamidoamine (PAMAM or “starburst”) dendrimer, G5-B1100, was linked to oligosaccharidemoieties, which were distant from antigen binding sites of cetuximab, bymeans of the heterobifunctional reagents N-succinimidyl-3-(2-pyridyldithio)propionate(SPDP) and N-(k-maleimidoundecanoic acid) hydrazide (KMUH). The resulting bioconjugate,designated C225-G5-B1100, was separated from the unconjugated dendrimerusing a Sephacryl S-300 column. Based on the relative concentration ratios ofboron and protein, there were ~1100 boron atoms per molecule of cetuximab with onlya slight reduction of Ka. The localization of C225-G5-B1100 or G5-B1100 in rats bearingintracerebral implants of either F98EGFR or F98WT gliomas was determined 24 hfollowing direct intratumoral (i.t.) injection at which time 92.3±23.3 µg B/g tumor waslocalized in F98EGFR gliomas versus 36.5±18.8 µg B/g tumor in F98WT gliomas and13.4±6.1 µg in normal brain. In contrast, only 6.7±3.6 µg B/g tumor of G5-B1100 waslocalized in F98EGFR gliomas following i.t. injection, thereby demonstrating specificmolecular targeting of EGFR. Based on these data, BNCT studies will be initiated inF98EGFR glioma bearing rats to evaluate C225-G5-B1100 for the treatment of intracerebralbrain tumors.6 - DNA damage of the rat 9L gliosarcoma cell after BoronNeutron Capture Therapy by use of the single-cellgel electrophoresis (Comet) assay. K. Nakai a,b, *, H. Kumada a,b ,F. Yoshida a,b , K. Endo a,b , T. Yamamoto a,c , A. Matsumura a,ba Department of Neurosurgery, Institute of Clinical Science, University of Tsukuba,Tenoudai1-1-1 , Tsukuba, Ibaraki, 305-8575, Japan, b University of Tsukuba, GraduateSchool of Comprehensive Human Sciences, Functional and Regulatory MedicalSciences, c Ibaraki Prefecture Central Hospital, Koibuchi 6528, Tomobe, Ibaraki,309-1793, Japan, d Japan Atomic Energy Research Institute, Shirakata-shirane 2-4,Tokai-mura, Ibaraki, 319-1195, JapanThe Single-cell electrophoresis gel (Comet) assay was used to investigate DNA damageafter boron neutron capture therapy (BNCT). The basic principle of the cometassay is the migration of DNA in an agarose matrix under electrophoretic conditions.Among the various versions of the assay, the alkaline (pH of the unwinding >13) methodsenables detection of the broadest spectrum of DNA damage. It can detect doubleand single strand breaks, alkali-labile sites and DNA crosslinking. On the other hand,the neutral methods detect mainly double strand break (DSB) and seems to be usefulfor assessing the DNA fragmentation after BNCT. The aim of this study is to investigatethe DNA damage precisely in order to find optimum and effective condition for BNCT.The 9L-gliosarcma cell lines were used. Borocaptate (BSH) was administrated inmedium 48hours before neutron radiation. 10B concentration of treatment groupswas kept at 5ppm (low dose group) and at 30ppm (high dose group) through out theprocedures. The cells were filled with medium in cryovials and neutron radiated atJRR-4 with rotating irradiation system 30minutes and 90minutes. The samples wereperformed comet assay under alkaline conditions and neutral conditions respectively.The comet assay analysis was carried out with epifluorescence microscope and Publicdomain image-analysis programs. Therefore we use the simplified definition of tailmoment: (Tail length ) * (Percent tail DNA). In each experiment 30-50 cells were measuredper slide.The neutron fluence was 8.65E+11n/cm2 in 30min radiation group and 2.57E+12n/cm2in 90min radiation group, respectively. The neutral comet assay showed increase thetail moment in boronated cells. In the distribution histogram of tail moment of neutralassay, large moment cells were seen in especially in boronated cells. In the alkalinecondition which shows single and double strand breaks, no difference was observedamong these groups as can be seen.The difference in tail moment between irradiated boronated and non-boronated cellsshow that double strand break was induced by boron existence. However, the differencebetween low dose boronated group and high dose boronated group was not significant.This result leave room for variety of interpretations. It is possible to build somehypothesis as follows. 1. The subcellular boron distribution is not parallel to the mediumboron concentration. 2. recovery of strand break 3. cell-cycle dependent transportwas influenced. These possibilities should be investigated in the future study.Biological response of BNCT highly depends on intracellular boron neutron capture reactioni.e. microdistribution and intracellular concentration of boron compound. Cometassay represents direct measurements of DNA damage which is basis of any biologicalresponses to radiations. Further investigation should be required to clarify themechanism of cell damage in BNCT and its difference among boron compounds. Asconclusion, this assay is very simple and quick, so it can be used for screening of newboron compounds or new irradiation facility.8 - Ascorbic acid reduced mutagenicity at the HPRT locusin CHO cells against thermal neutron radiation. YukoKinashi*, Yoshinori Sakurai, Shinichiro Masunaga, Minoru Suzuki, KenjiNagata, Koji OnoResearch Reactor Institute, Kyoto University, Kumatori-cho, Sennan-gun, Osaka 590-0494, JapanPurpose: The purpose of this study is to compare the biological effect of neutroninducedlong-lived radicals with that of neutron-induced short-lived radicals. Someinvestigator reported that radiation-induced long-lived radicals were scavenged bypost-irradiation treatment of ascorbic acid. We studied the effects of ascorbic acidacting as a long-lived radical scavenger on cell killing and mutagenicity in Chinese

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - B i o l o g yhamster ovary (CHO) cells against thermal neutrons produced at the Kyoto UniversityResearch (KUR) reactor.Materials and Methods: Ascorbic acid (5mM) was added to cells 30 min. after irradiationand removed 150 min. after neutron irradiation. CHO cells were irradiated in theTeflon tubes by thermal neutrons with or without boron at 10 ppm. The biological endpoint of cell survival was measured by colony formation assay. The mutagenicity wasmeasured by the mutant frequency in the HPRT locus.Results and Discussion: The cell killing effect of neutrons with or without boron did notchanged after ascorbic acid post-irradiation treatment. However, ascorbic acid showedan apparent protective effect against mutagenesis of the HPRT locus induced by thermalneutron irradiation. After ascorbic acid post-irradiation treatment, the mutagenicitywas decreased, especially when the cells were irradiated with 10ppm of boron. At the2Gy of neutron irradiation, mutation frequency was reduced from 6 to 2 per 105 (onehundred and thousand) cells with no boron, from 10 to 2 per 105 (one hundred andthousand) cells with 10 ppm of boron. Previously, we reported that the dimethyl sulphoxide(DMSO) showed the protective effect on cell killing and mutagenicity againstthe neutron irradiation. DMSO is well known as a short-lived radical scavenger. Wehere compare the effect of mutation induction of short-lived radicals and long livedradicals.The treatment with DMSO, which acts as a short-lived radical scavenger, reducedmutation induction of neutrons. Furthermore, the ascorbic acid post-irradiationtreatment, which acts as a long-lived radical scavenger, reduced much more mutationinduction of neutrons than the short-lived radical scavenger did. This protective effectof ascorbic acid was dominant in the presence of 10 ppm of boron. These observationsindicated that long-lived radical scavenger has a much more radio-protectiveeffect against the mutagenic actions of 10B(n,alpha)7Li reaction than that of 1H(n,gamma)2D or 14N(n,rho)14C reactions induced by neutron irradiation.Conclusion: Our results suggest that post-radiation treatment of ascorbic acid do notchange the cell killing effect of neutrons. Ascorbic acid scavenge long-lived radicalseffectively and cause much more protective effects than short-lived radical scavengeragainst mutagenicity of normal tissue in the BNCT (boron neutron capture therapy).10 - BNCT of spontaneous melanoma at dogs clinicalexperimental Study. V.N.Kulakov a , V.F.Khokhlov a , I.N.Sheino a ,A.A.Portnov b , K.N.Zaitsev b , V.N.Mitin c , N.G.Kozlovskaya c , I.A.Shikunova ca SSC–Institute of Biophysics, 46Moscow, Russia, b Moscow Engineering PhysicsInstitute, Moscow, Russia, c Russian Cancer Research Center - RAMS, Moscow,RussiaThe purpose of the clinical experimental study was to assess the effectiveness of BNCTfor spontaneous melanoma in dogs with tumors of various grades. The effectivenessof BNCT was compared with -therapy and radiation therapy in the neutron beam ofthe MEPhI reactor, for systemic and regional (in the tumor-feeding artery) routes ofadministration of the neutron capture agent to the animals. The artery was selectedunder X-ray control with use of an X-ray contrast agent. Borate ethers of [10B]-Lboronphenylalanine(BPA) were used in doses: 140 - 330 mg/kg (intravenously) and 50- 200 mg/kg (regionally in the tumor-feeding artery) as converted to BPA. The studywas carried out in three control and two experimental groups of animals: control #1(irradiation of tumor-free animals in the reactor neutron beam), control #2 (irradiationof the tumor in the neutron beam), control #3 (3a - 10-14 sessions of - irradiation ofthe tumor, total dose - 40-50 Gy; 3b - surgical removal of the tumor, immunotherapy),experimental #4 (neutron irradiation of the tumor with BPA injected: 4a -intravenousadministration, 4b - regional administration). The dogs were selected basing on theresults of clinical examination: a survey by an anesthesiologist, blood count, cytologicanalysis of puncture biopsy material, lung X-ray, ultrasonic scan of abdomen, in somecases ECG. Irradiation was performed at the MEPhI reactor in a specially designedroom with systems of telemetric monitoring, medicine delivery, and ventilation.The studies followed in accordance with the laws protecting the rights of animals. Irradiationwas carried out under general anesthesia and temporary immobilization of thedogs; the drugs used for narcosis are permitted by the current legislation of the Russiafor application in the veterinary centers in scientific experiments. Irradiation time was90 minutes at doses of 20-30 Gy (without 10 ) and 60-90 Gy (with 10 ); the ratio of 10Bconcentrations in tumor and normal tissues before irradiation was ~2.5, post irradiation- ~1.5. The hemopoiesis of the animals was controlled at 3, 7, 14 days, and 1, 3, and6 months post irradiation.BNCT resulted in complete involution of tumor in 86% of cases. The blood countshowed a minor decrease of lymphocytes by the 7th -14th day, which recovered in 5-7days. Recurrence of the tumor was observed in 13% of the animals. Complications:radiation stomatitis in the animals with neoplastic process in the mouth, alopecia intumor sites, with hair regrowth in 2-3 months. The treatment of melanoma in the neutronbeam without administration of boron-containing drugs results in a low therapeuticeffect. The survival of dogs with melanoma was 18 months after BNCT, 7 months after-therapy, and 6 months after neutron irradiation without BPA; this corresponds to 6, 2.3and 2 years of human life.The work is carried out with financial support ISTC, Project # 1951.12 - Cobaltacarborane derivatives for NCT. V.N.Kulakov a ,V.I.Bregadze b , V.F.Khokhlov a , T.P.Klimova b , V.V.Mesherikova c ,T.A.Nasonova a , I.A.Dobrinina a , I.B.Sivaev b , O.M.Khitrova ba Russian State Research Center - Institute of Biophysics, Moscow, Russia, bA.N.Nesmeyanov Institute of Organoelement Compounds, Moscow, Russia, cN.N.Blokhin Russian Cancer Center of the Russian Academy of Medical Sciences,Moscow, RussiaCobaltacarborane [3,3’-Co(1,2-C2B9H11)2]- is proposed as an alternative boron moietyfor synthesis of BNCT agents. This moiety contains more boron atoms per a moleculethan carboranes or [B12H12]2- and displays an extraordinary stability due to thedelocalized cluster bonding of cobalt with the dicarbollide ligand orbitals1. Synthesis ofa number of functional derivatives of cobaltacarborane has been described recently2.In this contribution we report results of our study of comparative efficiency ofbis(dicarbollide)cobalt (DCC) and its amino acid derivative (DCC-AA) in in vitro and invivo experiments. The study was performed using the epithermal neutron beam of theresearch nuclear reactor at Moscow Engineering Physical Institute.The in vitro experiments were carried out on mouse B-16 melanoma cells. The cellswerepropagated as a monolayer in plastic flasks on RPMI-1640 growth media with 10%fetal calf serum and 80 µg/ml of Gentamicin. The cells were irradiated in suspension inthe full growth media in concentration 0.5·106/ml in 0.5 ml aliquots in Eppendorf tubes.After the irradiation cells were seeded in plastic Petry dishes for assay of cell platingefficiency (cell survival). These values were found to be 11.7±0.5 and 5.7±0.3 for DCCand DCC-AA, respectively.In addition for cell survival, assayed after 12 days of cultivation, we also measuredthe initial growth rate of the cell colonies by counting cell number in 30 randomlychosen colonies once per day during several days of colonies growth. The averagecell number per growing colony at 3-rd day after irradiation, taken as percentage ofcell number in the colonies grown from untreated cells, was used as the end-pointsand designated as a cell growth rate. They are 10.6±0.6 and 5.4±0.2 for DCC andDCC-AA, respectively.The results obtained demonstrate that DCC and DCC-AA increase efficiency of neutronirradiation, DCC-AA being more efficient than DCC. Both compounds themselvesdid not change cell survival (colony forming ability), when they were present in thespecified concentration during the preparation and irradiation of cell suspension (afterirradiation cell suspension was greatly diluted with fresh medium). However, DCC-AAdecrease cell multiplication in the initial days of the colony growth up to 55% of thecontrol values.The in vivo experiments were performed on Black/6 mice with B-16 tumors implantedin shanks. The treatment was carried out on the 8-th day after cell implantation, whenthe tumor size was ~ 1 cm3. For this purpose 0.1 ml aqueous solution of boron compound,containing 20 µg 10B, was introduced intratumorally as 4 injections 20-30 minbefore the irradiation. It was found that DCC increases suppression of tumor growthby neutron irradiation, while DCC-AA practically does not increase effect of neutronirradiation. This is in contradiction with results of the in vivo tests and can be explainedby faster removal of DCC-AA from the tumor in comparison with DCC.1. I.B.Sivaev, V.I.Bregadze. Chemistry of cobalt bis(dicarbollides). A review. Collect.Czech. Chem.Commun., 1999, 64(5), 783-805.2. I.B.Sivaev, Z.A.Starikova, S.Sjöberg, V.I.Bregadze. Synthesis of functional derivativesof the [3,3° -Co(1,2-C2B9H11)2]- anion. J.Organomet.Chem., 2002, 649(1), 1-8.14 - Boron distribution in each structure of normal ratbrain after intravenous injection of boronophenylalanine-fructose.Y. Shibata a, *, R. G. Zamenhof a , H. Patel a , W. S. KigerIII ba Department of Radiology, Beth Israel Deaconess Medical Center, Harvard MedicalSchool, 330 Brookline Ave. Boston MA, 02215 USA, b Department of Radiation Oncology,Beth Israel Deaconess Medical Center, Harvard Medical School, 330 BrooklineAve. Boston MA, 02215 USAMicro-distribution of boron compound is critical to determine the radiation effect for19

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - C h e m i s t r y2 - Synthesis and In Vivo Evaluation of BPA-Gd-DTPAComplex as an MRI Contrast Agent and as a Carrier forNeutron Capture Therapy. Kazunori Takahashi a , Hiroyuki Nakamurab , Shozo Furumoto c , Kazuyoshi Yamamoto d , Akira Matsumura e , HiroshiFukuda c , and Yoshinori Yamamoto aa Department of Chemistry, Graduate School of Science, Tohoku University, Sendai,b Faculty of Science, Gakushuin University, Tokyo, c Department of Nuclear Medicineand Radiology, IDAC, Tohoku University, Sendai, d Department of Research Reactor,Tokai Research Establishment, Japan Atomic Energy Research Institute, Tokai-mura,e Department of Neurosurgery, Institute of Clinical Medicine, Tsukuba University,Ibaraki, Japan.We synthesized a BPA-Gd-DTPA compound, as a carrier for neutron capture therapyto be used for MRI contrast media. Pinanediol was used as the protective groupfor B(OH)2 group of BPA, and the BPA unit was connected to the DTPA frameworkthrough an amide bond. The structure of the BPA-Gd-DTPA was confirmed by theESI-MS and the elemental analysis. The biodistribution studies were performed afterinjection of the compound into AH109A hepatoma bearing Donryu rats. The concentrationsof Gd and boron were measured by prompt g-ray analyses. The tumor uptakes(% ID/g) were 1% and 0.3% at 20 min and 60 min after injection, respectively and werehigher than that of carborane-Gd-DTPA, which we previously reported. However, liverand kidney uptake was very high and tumor/blood ratio was very low (0.38) comparedto that of BPA itself (ca. 3.0). Alfa autoradiogram of a tumor bearing rat showed higherconcentration of boron in the tumor compared to surrounding muscle and very highin the intestine. Although tumor selectivity of the compound was higher than that ofcarborane-Gd-DTPA, further studies of the synthesis and in vivo evaluation of betterbinary compounds are continuing.The particle sizes were determined as the Z-average means by the photon correlationspectroscopy using a Zetasizer 3000HS (Malvern Instruments Ltd., UK). Gd-nanoMIC-SA consisted of 300 mg of Gd-DTPA-SA, 1200 mg of HCO-60 and 10 mg of stearicacid in 10 ml of water with a particle size of 83 nm. Subsequently heating at 70C for 20min reduced the size of Gd-nanoMIC-SA to 38 nm. Hamsters bearing Green’s melanoma(melanotic No.179 cell, D1-179) were obtained by subcutaneous transplantationof the melanoma cell fragments on the left thigh 10 days before use. The Gd-nanoLPor Gd-nanoMIC-SA suspension (4.5 mg Gd/ml) was twice injected intravenously at 24-hour interval at a dose of 1ml per hamster each. At the predetermined time after injection,blood was collected from heart after the hamsters being anesthetized with ether.After the hamsters were sacrificed, tumor, liver, spleen, kidney, lung, skin and musclewere removed immediately. These tissues were directly wet-ashed with nitric acid,and then the Gd concentration in each tissue was measured by inductively coupledplasma-atomic emission spectrometry (ICP-AES).The profiles in blood indicated the faster elimination with Gd-nanoMIC-SA. In tumor,Gd concentration was 127 micro-g/g wet tissue with Gd-nanoMIC-SA and 189 microg/gwet tissue with Gd-nanoLP at 12 hours. The Gd concentrations in liver and spleenwith Gd-nanoLP leveled at 833 and 687 micro-g/g wet tissue, respectively, at 12 hours.However, those with Gd-nanoMIC-SA exhibited the peaks of 759 micro-g/g wet tissueat 72 hours and 728 micro-g/g wet tissue at 24 hours, respectively, in spleen, followedby decrease to 407 micro-g/g wet tissue at 72 hours. Thus, the unexpectedly high Gdaccumulations in liver and spleen with Gd-nanoMIC-SA seemed only temporary. TheGd levels in the other organs were very low. The tumor/blood (T/B) ratio at 12 hourswas 0.9 with Gd-nanoLP, and 2.6 with Gd-nanoMIC-SA. But, T/B ratio of Gd-nanoMIC-SA was increased to 6.2 at 24 hours. These results suggested that Gd-nanoMIC-SA,which was prepared in a simpler formulation with a smaller particle size, making itpossible to effectively modify the particle-surface without excessive size-enlargement,might be an alternative to Gd-nanoLP.4 - Synthesis and Vesicle Formation of a nido-CarboraneCluster Lipid: Application to Boron Delivery System forBNCT. Hiroyuki Nakamura a, * Yusuke Miyajim a , Toshiaki Takei a , SatoshiKasaoka b , Kazuo Maruyama ba Department of Chemistry, Faculty of Science, Gakushuin University, Tokyo 171-8588, Japan, b Department of Biopharmaceutics, School of Pharmaceutical Sciences,Teikyo University, Kanagawa, 199-0195, JapanThe nido-carborane lipid, which has a double-tailed moiety, was synthesized fromheptadecanol by 5 steps. According to the analysis in a Transmission Electron Microscopyby negatively stained with uranyl acetate, the lipid formed a stable vesicle inwhich calcein was encapsulated. The nido-carborane lipid was also incorporated intodistearoylphosphatidylcholine (DSPC) liposome at a very high concentration. The liposomeprepared from DSPC, PEG-DSPE and the lipid exhibited a similar distribution inmice in comparison with the original PEG-liposome.226 - Gadolinium-Containing Micellar Nanoparticles ForNeutron Capture Therapy. Y. Kaneseki, H. Ichikawa, Y. FukumoriFaculty of Pharmaceutical Sciences and High Technology Research Center, KobeGakuin University, JapanIn the trials of gadolinium neutron capture therapy (Gd-NCT), gadolinium diethylenetriaminepentaaceticacid (Gd-DTPA), being used clinically as MRI contrast agent, hasbeen traditionally employed so far. However, Gd-DTPA is rapidly eliminated from thesystemic circulation and hardly accumulates in tumor because of its high hydrophilicity,while Gd concentration in tumor required to obtain the tumor inactivation effect isestimated to be more than 100 micro-g Gd-157/ml. Thus, gadolinium diethylenetriaminepentaaceticacid distearylamide (Gd-DTPA-SA) loaded lipid nanoparticles (GdnanoLPs)were prepared for Gd-NCT by a thin-film hydration method combined witha sonication. Gd-nanoLPs had a core of soybean oil and a surface layer containinghydrogenated egg yolk phosphatidylcholine (HEPC), occupying much volume of particlewith a particle size of 77 nm. In order to simplify the formulation and reduce theparticle size, micellar nanoparticles (Gd-nanoMIC-SA) were prepared without soybeanoil and HEPC.

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - N u c l ea r E n g i n e e r i n gThe GE PETtrace was chosen for this investigation because this type of cyclotron ispopular among nuclear pharmacies and clinics in many countries (about 75 PETtraceshave been sold worldwide), in part because it’s compact and reliable. This cyclotronproduces protons with energies large enough to produce neutrons with appropriate energyand flux for BNCT and it does not require significant changes in design to providethese neutrons. In particular, using the standard PETtrace 18O target is considered.The 18O target was studied because about 90-95% of time these cyclotrons use thistarget already for fluorodeoxyglucose (FDG) production. The efficiency of cyclotronuse may be significantly increased if unused neutrons produced during FDG productioncould be utilized for other medical modalities such as BNCT in the same time asradioisotope production.The resulting dose from the radiation emitted from the target is evaluated using theMonte Carlo radiation transport code MCNP. The input geometry includes a moderatorassembly surrounded by a 30 cm thick graphite reflector. The heavy water was chosenas a moderator. Various thicknesses of moderating material were used in this study.The thickness of 30 cm was chosen because it provided the lowest fast-neutron contributionto the total flux in comparison to the low energy component and still allowed deliveryof the necessary dose in a reasonable time. The tally cells were 0.5 cm diameterspheres at depths of 1, 2, 3, 4, 5, 6, 7 and 8 cm inside the phantom brain. The flux tallywas used to calculate photon and neutron dose, by applying flux-to-dose conversionfactors. The results suggest that it is possible to use this particular cyclotron with thistarget for tumor treatment. The data presented here suggest that this type of neutronsource should provide acceptable doses and treatment times for tumor irradiation atdepths of up to 4 cm inside the brain. Treatment of the tumor at greater depths requiressignificant increase of the treatment time.Since some modifications are proposed by GE in cyclotron design (increased targetcurrent, new ion source development, etc.) further research will be continued in orderto improve current results.10 - Development of In-Hospital Neutron Irradiator byZhou Yongmao. Zhou Yongmao a , Gao Zhixian b , Xia Pu c , Zhu Jianshida China Zhongyuan Engineering Cooperation, Beijing China, b Beijing NeurosurgeonResearch Institute, Beijing Tiantan Hospital, Beijing,China, c China Atomic EnergyInstitute, Beijing, China, d Beijing Application Physics & Calculation Mathematics Institute,Beijing, China“Towards hospital-based NCT” has become one of major discussion topics in eachsession of International Symposium on Neutron Capture Therapy for Cancer theseyears. Based on our successful and ripe experience in developing miniature researchreactors, we are engaged in the development of new NCT facilities series i.e. In-hospitalNeutron Irradiator (IHNI) that is to be installed in the hospital and operated byneurosurgeon incharge. Research on the first set of IHNI for brain tumor treatmentconcentrates on selection of LEU core design parameters, optimal comparison of adoptedmaterials and sizes of neutron beam irradiation devices as well as developmentof BNCT treatment planning software. To date, the comparison to schemes has beendone, major design parameters have been worked out, facilities installation site hasbeen decided and engineering design and construction is in progress. The facility isscheduled to generate neutron beam in 2006.12 - The SPES-BNCT project: An accelerator based neutronbeam facility at INFN Legnaro. J. Esposito* ,a , P. Colautti a ,A. Pisent a , L. Tecchio a , S. Agosteo b , L. De Nardo f , G. Jori c , G. Sotti d , G.Fortuna a , R.Tinti ea Istituto Nazionale di Fisica Nucleare, INFN-LNL, via Dell’Università 2, I-35020 Legnaro(PD), Italy, b Dipartimento di Ingeneria Nucleare, Politecnico di Milano, via Ponzio34/3, I-20133 Milano, Italy, c Dipartimento di Biologia, Università di Padova, via UgoBassi 58/B, I-35121 Padova, Italy, d U. O. Radioterapia -Azienda Ospedaliera -Universitàdi Padova, via Giustiniani, 2, I-35128, Padova Italy, e Ente per le nuove tecnologie,l’energia e l’ambiente, ENEA/ERG-SIEC, via Martiri di Monte Sole 4, I-40129Bologna, Italy, f Dipartimento di Fisica, Università di Padova, via F. Marzolo, 8 I-35131Padova, ItalyIntroduction: An accelerator-based thermal neutron source, aimed at the BNCT treatmentof extended skin melanoma, is foreseen to be installed in the next years at theLaboratori Nazionali di Legnaro (LNL), in the framework of SPES (Study and Productionof Exotic nuclear Spieces) project aimed at the investigation of the new frontierin nuclear physics research of exotic, not stable nuclei. The SPES-BNCT project willexploit the beam delivered by the 5 MeV, 30 mA (i.e. 150 kW) Radio-Frequency Quadrupole(RFQ) proton driver currently under construction. The LNL-BNCT project willtherefore mean to represent a challenge to provide an intense thermal neutron beamfacility and a fundamental test bench for an operative, accelerator-based BNCT facilityconcept, which could provide in perspective a possible spin-off in a hospital institutioninstead of the more complex, even low power, dedicated reactor-based systems.Materials and Methods: A combined boron neutron capture plus photodynamic(BNCT+PDT) therapy approach will be investigated, due to the promising photosensitizersuptake selectivity in tumour tissues A new, boron loaded phthalocyanine compound(B-Pc), has been recently synthesized by Molteni Pharmaceuticals (Florence,Italy) in order to investigate this possibility. Moreover a preliminary beryllium neutronproduction target is being designed in collaboration with the STC Sintez of EfremovInstitute in S. Petersburg as the best neutron converter solution consistent with theSPES design specifications The first, full-scale prototype is foreseen to be constructedby the end of 2004 which will then undergo a series of operative as well as criticalpower test conditions.Irradiation beam quality monitoring: A first mini-TEPC prototype with a small sensitivevolume able to monitor the fluctuation of the absorbed energy, namely the microdosimetricspectra, both in a 1 µm and in a 50 nm simulated size site (at density of 1g/cm3),which are respectively of chromosome and of mitochondrion typical sizes, has beendesigned and constructed. The goal is to provide an on-line therapeutic neutron beammonitor, quite complex in BNCT treatment because of the large LET spectrum in theirradiated tissues. Preliminary TEPCs in air measurements have been performed at 1µm simulated size, using both 0 ppm and 100 ppm 10B loaded TEPC cathode shells ina mixed radiation field (gamma, fast and slow neutrons) separating the different dosecontributions, both at the LNL demonstration facility, with a 5 MeV, low current (1µA)proton beam and on a thick beryllium target as well as at irradiation cavity of TAPIROENEA fast reactor thermal column.Conclusion: The main next steps of SPES-BNCT project will be focused on the neutronconverter prototype construction and test, the final thermal beam shaping facility designand the construction of a new mini TEPC for monitoring and processing microdosimetricdata in high flux radiation fields. Moreover the remodelling of TAPIRO thermalirradiation facility, in order to minimize the gamma ray contamination for in vitro and invivo studies, as well as the feasibility of a new boron phthalocyanine compound with60 10B atoms clusters will be carried on.14 - Performance enhancements of MCNP4B, MCNP5,and MCNPX for Monte Carlo Radiotherapy Planning Calculationsin Lattice Geometries. W.S. Kiger, III a , J.R. Albrittonb , A.G. Hochberg a,c , and J.T. Goorley da Department of Radiation Oncology, Beth Israel Deaconess Medical Center, HarvardMedical School, 330 Brookline Avenue, Boston, MA 02215, USA, b Nuclear EngineeringDepartment, Massachusetts Institute of Technology, 77 Massachusetts Avenue,Cambridge, MA 02139, USA, c Institut National des Sciences et Techniques Nucléaires(Commissariat à l’Energie Atomique), Paris, France, d Los Alamos NationalLaboratory, X-5, Mail Stop F663, Los Alamos, NM 87545, USAAchieving reasonable computation times for Monte Carlo-based radiotherapy planningcalculations while simulating enough histories to maintain acceptable statistical precisioncan be difficult, especially for the computationally expensive, scatter-dominatedneutron transport problems required for Neutron Capture Therapy (NCT). Several NCTtreatment planning systems (TPS) employ the general-purpose Monte Carlo radiationtransport code MCNP as their dose computation engine because of its many advantages.This paper examines the issue of computational speed for 3 versions of theMCNP code, MCNP4B, MCNP5, and MCNPX, in the context of NCT treatment planningcalculations using a voxel phantom produced by the NCTPlan TPS. In addition tothe standard versions of these codes, patched versions of MCNP4B and MCNP5 speciallyaccelerated for calculations in a lattice geometry were assessed. Furthermore,the influence of different geometric representations (cell or lattice representations ofthe voxel model) and tallying techniques, including the recently developed mesh tally,on computation efficiency was assessed. For certain combinations of geometric representationand tally techniques, the computations are prohibitively slow, taking morethan 8,000 minutes per million source neutrons and photons. For the problem studied,the minimum total computation times of 12.3 and 16.2 min were obtained using thepatched versions of MCNP4B and MCNP5, respectively, with a lattice geometry for 106neutron and 106 photon histories. Using the standard, unpatched versions of MCNP,computation times only 23-71% slower can be obtained by using a judicious combinationof geometric representation and tally technique to avoid prohibitively slow computations.Compared to the slowest calculations, calculations using the patched versionof MCNP4B and MCNP5 represent 530- to 660-fold improvements in speed. Thesestudies may provide useful guidance for others who are using MCNP for radiotherapyplanning calculations or other applications with similar voxel or lattice geometries.24

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - N u c l ea r E n g i n e e r i n g16 - Moderator Assembly Design Assessment Usingnewly-designed Neutron Field Assessment Parameters.Chenguang Li*, Thomas E. BlueNilendu Gupta Nuclear Engineering Program, Department of Mechanical Engineering,The Ohio State University, Columbus, OH,43202The purpose of this article is to use the newly developed neutron field assessmentparameters (NFAPs) to evaluate Boron Neutron Capture Therapy (BNCT) moderatorassembly design, and at the same time, to demonstrate the viability of using ahigh-resolution voxel-based head phantom (the Zubal phantom) with the designedmoderator assemblies for Accelerator Based Neutron Source (ABNS) for BNCT in ourMCNP calculations.Boron Neutron Capture Therapy (BNCT) for the malignant brain tumor in Japan hasbeen performed using a thermal neutron beam since 1968. In the conventional BNCT,the irradiation under craniotomy has been performed in order to compensate the doseinsufficiency by the depths pathological defects as a decrease characteristic of a thermalneutron beam. Since the conventional dosimetry technique cannot be used forthe epithermal neutron beam, it is necessary to perform faithfully the irradiation planfor preparation in order to be realized virtual irradiation plan. In materialization of irradiationwithout craniotomy, a numerical simulation conditions must be equal to actualirradiation conditions, which has the positioning data and beam intensity data. Themethod of monitoring on epithermal neutron beam was selected the foil activationmethod, the scaling factors were derived from calibration experiments with the reactionrate of the gold wire for the reactor power. In this report, we also will explain ourprocedure of clinical irradiation with epithermal neutron using the gold wire monitors.To carry out the BNCT using the epithermal neutron, the epithermal neutron beamintensity was measured by using Au-197 reaction rate activated on the resonanceabsorption peak (4.9eV). Two scaling factors, which are the reactor power calibrationfactor and the calculation / experiment (C/E) scaling factor, are necessary in order tocorrect with the simulation and actual irradiation experiment. First, an optimum detectorposition was investigated using MCNP code. The result of MCNP calculationshowed that the influence of large subject placed at the collimator was below 1% whenthe detector was placed in the distance of over 20cm from the collimator. Thereforewe installed the monitor holders near the bismuth block in order to set three gold wiremonitors. The factors were determined in the calibration experiments that measurethe thermal neutron flux in the phantom and reaction rate of the gold wire monitors.It was confirmed that the linearity of the reaction rate of the gold wire monitor for themaximum thermal neutron flux in the phantom was clear correlation. The C/E scalingfactor required from the experiment at 3.5MW as the standard irradiation was 1.041.We always can perform faithfully the irradiation planning in order to be realized virtualirradiation plan with the scaling factors. The procedure of dosimetry in medical irradiationmust be little changed. However we are able to know many information of dosimetryfor the epithermal neutron beam from tiny information of the reaction rate of thegold wire monitors. The medical irradiation using the epithermal neutron was carriedout using this procedure on October 21, 2003. This study had significance that it wasconnected for the reliability of measuring technique that it should become a standard,when priority is given to certainty, stability and safety further than what to introduce theepithermal neutron beam newly. In addition, we will install the on-line monitor for timefluctuation correction in the near future.Commonly used NFAPs for NCT include those developed by Harvard/MIT group(Zamenhof et al., 1990): the Advantage Depth (AD), Advantage Ratio (AR), GlobalAdvantage Ratio (GAR), and the Advantage Depth Dose-Rate (ADDR). At OSU, theNFAPs of the treatment time and tumor dose, which is developed by Woollard et al.(1996) have been used for moderator assembly design evaluation for ABNS for BNCTby now. Compared with these NFAPs mentioned above, the newly designed NFAPshave two advantages. First, the new NFAPs take into account the dose distributionthroughout the head phantom rather than only a few points of interest at the centerlineof the beam. Second, the new NFAPs evaluate the capability of the neutron beam ofboth maximizing tumor dose and minimizing normal tissue dose.To fully use the features that the new NFAP provides, we use the modified Zubalphantom, a high-resolution voxel-based head phantom which delineates the insidestructures of the brain. Previous studies have been made to demonstrate the viabilityof the Monte-Carlo N-Particle (MCNP) calculation using the Zubal phantom with a 1/Eneutron beam (Evans et al., 2001). And one purpose of this paper is to demonstratethe viability of incorporating a structure as complex as the moderator assembly intothe calculation.As the component that is mostly close to the patient, the shape and material of thedelimiter is important to the quality of the neutron beam and should be given adequateconsideration. In this paper, we will describe the calculations we made to study theeffect of the size of the delimiter’s beam port on BNCT treatments.18 - Calibration of Epithermal Neutron Beam Intensity forDosimetry at JRR-4. K. Yamamoto a, *, H. Kumada a , T. Kishi a , Y. Torii a ,Y. Sakurai b , T. Kobayashi ba Japan Atomic Energy Research Institute, JAERI, Shirakatashirane 2-4, Tokai-mura,Naka-gun, Ibaraki, 319-1195 Japan, b Kyoto University Research Institute, KURRI,Asashironishi 2-1010, Kumatori-cho, Sennan-gun, Osaka, 590-0494 Japan25

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - P hy s i c s2 - Application of Invasion Mathematical Model in Dosimetryfor Boron Neutron Capture Therapy for MalignantGlioma. K. Yamamoto a, *, H. Kumada a,b , K. Nakai b , K. Endo b , T. Yamamotoc , A. Matsumura ba Japan Atomic Energy Research Institute, Shirakata-shirane 2-4, Tokai-mura, Ibaraki,319-1195,Japan, b University of Tsukuba, Tenoudai 1-1-1 , Tsukuba, Ibaraki, 305-8577, Japan, c Ibaraki Prefectural Central Hospital, Koibuchi 6528, Tomobe, Ibaraki,309-1793, JapanA dose distribution considered the tumor cell density distribution is required on thegeneral principle of radiation therapy. We propose a novel method of determining targetregion considering the tumor cell concentration as a new function for the nextgeneration Boron Neutron Capture Therapy (BNCT) dosimetry system. It has not beenable to sufficiently define the degree of microscopic diffuse invasion of the tumor cellsperipheral to a tumor bulk in malignant glioma using current medical imaging. Referringto treatment protocol of BNCT, the target region surrounding the tumor bulk hasbeen set as the region which expands at the optional distance with usual 2cm marginfrom the region enhanced on T1 weighted gadolinium Magnetic Resonance Imaging(MRI). The malignant glioma is characterized by their aggressive diffuse invasion tothe surrounding normal tissue. Recently, several researchers tried predictions in thesurvival time, etc. by the application of the mathematical model on the tumor cell invasionprocess, which introduced the infiltration characteristic from the research fieldof physic and mathematic. The current stream of mathematical research will providegreat insights into important problem if the mathematical analysis will be able to combinewith laboratory simulation and clinical treatment planning system. Especially, thedynamic spatial diffusion model is effective for developing the dose planning system.In this research, we calculated the cell density of the region boundary of the targetusing tumor cell diffusion model. Further, survival tumor cell density distribution afterthe BNCT irradiation will be predicted by two matters region diffusion model. Basedon these studies, a new concept for BNCT dosimetry system is proposed. The targetboundary cells concentration at 2cm margin is calculated from the diffusion model sothat this value is 1.22x10^6 cells/cm^3. The distribution of post-irradiation has onepeak of the cell concentration at 5.6cm radius. The distribution on frontal lobe gliomacells in the digital brain simulated. This result means a high possibility of new selectingmethod of the target region including the undetectable tumor by using the analyticalmodel that corresponds to the cell concentration distribution. In order to select thetarget region, we should know the initial distribution of the tumor cell from the medicalimages. In the future, much information for medical and biological data, which includethe survival tumor cell distribution, the survival time, the date of recurrence, the locationof recurrence, the information for combined other therapy after that and etc., couldbe provided, when the diffusion model could be linked to the dose planning system. Byusing this model, establishment of an intelligent treatment planning system for radiationtherapy including BNCT may become feasible.4 - Calculations of cellular microdosimetry parametersfor alpha particles and electrons. C.J. Tung a, *, C.S. Liu a,b , J.P.Wang c , S.L. Chang aa Department of Nuclear Science, National Tsing Hua University, Hsinchu 300, Taiwan,b Department of Nuclear Medicine, Taipei Veterans General Hospital, Taipei112, Taiwan, c National Synchrotron Radiation Research Center, Hsinchu SciencePark, Hsinchu 300, TaiwanThe dosimetry in boron neutron capture therapy involves both the high and low linearenergy transfer particles. The relative biological effectiveness of these particles is determinedby the microdosimetric distribution of the energy imparted to the matter incellular volumes. Stochastic quantities, including the lineal energy and the specific energy,and nonstochastic parameters, such as the cellular S-value, are used in cellularmicrodosimetry. Their values depend on the source and target regions in the cell. Theradiation source is usually assumed to be uniformly distributed in one of the regionsof the cell: throughout the cell, cytoplasm, cell surface, or cell nucleus. The biologicaltarget is generally assumed as the cell nucleus or the entire cell.In the present work, the cellular microdosimetry parameters including the cellular S-valueand the single-event specific energy distribution were calculated for alpha particlesand electrons and different source to target region combinations. Calculations weremade using a semi-analytical model that simulated the emission of alpha particles orelectrons by the Monte Carlo method and calculated the energy imparted to the targetvolume by the analytical method. Delta particle equilibrium and partial delta particleequilibrium were applied to alpha particles and electrons, respectively, to calculate theenergy imparted to a cellular target volume. Range-energy relations were employed todetermine the incident and emerging energies of the primary particles crossing the targetvolume. For electrons, the fraction in the energy loss resulting from the generationof bremsstrahlung and high energy secondary electrons was estimated using data ofthe restricted collisional stopping power and the radiative stopping power. The energyloss straggling of electrons entering and leaving a target volume was evaluated by applyinga Gaussian distribution with the mean square energy loss data.Cellular S-values for alpha particles calculated in the present work were in excellentagreement with data of the Medical Internal Radiation Dose (MIRD) Committee ofthe Society of Nuclear Medicine. Cellular S-values for electrons of the present workwere also in very good agreement with results of the MIRD. In addition, the presentwork evaluated the uncertainties associated with the cellular S-values that were notconsidered by the MIRD. The energy loss straggling, responsible for such uncertainties,was also applied to determine the single-event specific energy distribution forelectrons. The energy loss straggling was found to make significant contribution tothe cellular S-values for electrons at, especially, high energies. It revealed that the S-values for electrons were two to three orders of magnitude smaller than the S-valuesfor alpha particles. This indicated that alpha particles were more effective than electronsin imparting energies to the target volume of micrometer size and thus resultinghigher RBEs than electrons. A comparison of the single-event distribution calculated inthe present work and using the Monte Carlo Penelope code showed that the presentmodel was feasible in microdosimetry calculations.6 - BINP accelerator based neutron source. B. Bayanov, Yu.Belchenko, V. Belov, V. Davydenko, A. Donin, A. Dranichnikov, A. Ivanov,I. Kandaurov, G. Kraynov, A. Krivenko, A. Kudryavtsev, N. Kuksanov, R.Salimov, V. Savkin, V.Shirokov, S. Taskaev*, M. TiunovBudker Institute of Nuclear Physics, 11 Lavrentiev ave., 630090 Novosibirsk, RussiaThe Budker Institute of Nuclear Physics (Novosibirsk) and the Institute of Physics andPower Engineering (Obninsk) have proposed an accelerator based neutron source forneutron capture and fast neutron therapy for hospital. Innovative approach is basedupon vacuum insulation tandem accelerator (VITA) and near threshold 7Li(p,n)7Beneutron generation. Negative hydrogen ion beam is injected into VITA. After chargeexchangeof negative hydrogen ion into proton inside the charge-exchange tube in thecenter of high-voltage electrode, the proton beam is formed at the outlet of the tandem.It is accelerated up to double voltage of high-voltage electrode. Neutron generation isproposed to be carried out by protons bombarding a lithium target using 7Li(p,n)7Bethreshold reaction. In ordinary mode, at proton energy of 2.5 MeV, the neutron sourceproduces neutron beam with maximum energy of 790 keV appropriate directly for fastneutron therapy and for neutron-capture therapy after moderation. The most efficientoperating mode of facility is at proton energy of 1.915 MeV that is 34 keV higher thanthe threshold of the 7Li(p,n)7Be reaction. In this mode, neutron beam is generated kinematicallycollimated in forward direction and its average energy of 30 keV, is directlyapplicable for boron neutron-capture therapy.Pilot accelerator based neutron source for neutron capture therapy is under constructionnow at the Budker Institute of Nuclear Physics, Novosibirsk, Russia. Surfaceplasmasource with Penning geometry of electrodes is to be used for obtaining a dc25 keV 10 mA beam of hydrogen negative ions. Two magnetic lenses will be used forlow energy beam transporting. Then, the beam will be accelerated in 33 kV cm–1 electricfield. Striping of negative ion beam is provided by gas target inside high voltageelectrode of tandem accelerator with vacuum insulation. Neutrons are generated byprotons bombarding the target covered with thin solid lithium layer. Complete experimentaltests are planned by the end of the year 2000.8 - Neutron Dosimetry on Phantom Model of PancreaticCancer Patient for Intraoperative Boron Neutron CaptureTherapy. Hironobu Yanagie a , Yosiyuki Sakurai b , Koichi Ogura c , TooruKobayashi b , Yoshitaka Furuya d , Hirotaka Sugiyama e , Tomoyuki Hisa a , MasayoriIshikawa f , Kuniaki Amemiya g , Hiroyuki Takahashi h , Hisao Kobayashi i ,Masao Tago j , Shin-ichiro Masunaga b , Koji Ono b , Keiichi Nakagawa j , MasaharuNakazawa g and Masazumi Eriguchi aa Research Center for Advanced Science & Technology, The University of Tokyo, Tokyo,Japan, b Reseach Reactor Institute, Kyoto University, Osaka, Japan, c Collegeof Industrial Technology, Nihon University, Chiba, Japan, d Department of Surgery,Teikyo University Ichihara Hospital, Chiba, Japan, e Department of Microbiology, NihonMedical University, Tokyo, Japan, f Reseach Center for Atomic Energy, The Universityof Tokyo, Tokyo, Japan, g Department of Quantum Engineering and SystemsScience, The University of Tokyo, Tokyo, Japan, h Research into Artifacts Center forEngineering , The University of Tokyo, Tokyo, Japan, i Institue for Atomic Energy,Rikkyo University, Kanagawa, Japan, j Department of Radiology, University of TokyoHospital, Tokyo, JapanPancreatic cancer is one of the most difficult curative cancer, so it is need new combinationaltherapy. If sufficient boron compound can be accurate to tumor, Boron Neutroncapture Therapy(BNCT) will be apply to pancreatic cancer treatment. We preparethe BNCT to pancreatic cancer patient treatment by intraoperative irradiation. In thisstudy, we performed preliminary dosimetory of the phantom model of abdominal cavity.26

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - P hy s i c sThe fluence of 8 x 107 n/cm2/sec (0.1 ratio) was 4.5 cm depth from the surface in thecase of simple irradiation, and the field of thermal neutron is spred as 13 cm and 11.5cm were usage of Void and Void with LiF collimation, respectively in thermal(OO-0011)mode. In the case of epithermal(CO-0000) mode, epithermal and fast components arefour times higher at the surface level. In the case of mixed beam(OO-0000) mode,thermal neutron flux was same as thermal neutron mode at 10 cm depth, but the gamma-raycomponent was two times higher than thermal neutron mode. When the usageof Void and LiF collimation, the thermal neutron was selectively irradiated to the regionof tumor in combined to the CT scan imaging of pancreatic cancer patient. It mean thatwe could irradiate the tumor selectively and safely as possible as with reducing theaffect to the neighbor normal organs. High resolutional whole body dosimetry systemwill be necessary to expand the application of BNCT to pancreatic cancer.Keywords : Boron Neutron Capture Therapy (BNCT), Pancreatic Cancer, IntraoperativeBNCT, Thermal Neutron, Void, LiF Collimation10 - Dosimetric comparison at FiR 1 using microdosimetry,ionisation chambers and computer simulation. J.Uusi-Simola a,b, *, T. Serén c , T. Seppälä d , A. Kosunen e , I. Auterinen c , S.Savolainen a,ba Department of Physical Sciences, University of Helsinki, PO BOX 64, FI-00014Helsinki, Finland, b Medical Imaging Center, University of Helsinki, PO BOX 340,FI-00029 HUS Helsinki, Finland, c VTT Processes, Technical Research Centre ofFinland, FI-02044 VTT, Finland, d Boneca Ltd., FI-00029 HUS Helsinki, Finland,e STUK, Radiation and Nuclear Safety Authority, Finland, PO BOX 14, FI-00081 Helsinki,FinlandIntroduction: The most commonly used method in BNCT for the experimental determinationof absorbed dose is the dual ionisation chamber technique. The estimateduncertainties associated with the determination of dose components with this methodare 2.5-9 % and 17-24 % for gamma ray and (fast) neutron dose, respectively. Tissueequivalent proportional counter (TEPC) microdosimetry has been applied previously inthe dosimetry of epithermal neutron beams and the method they can provide an independentand accurate method to determine gamma ray and neutron absorbed doses.Dosimetric comparison has been performed using a TEPC, dual ionisation chambersand DORT computer code at FiR 1 BNCT facility in Espoo, Finland. The three methodswere applied to determine neutron and gamma ray absorbed doses at 25, 40, 60 and120 mm depths in a water phantom. The determined absorbed doses were found toagree within the limits of the estimated uncertainties.Materials & Methods: The dosimeter was a commercially available 1.27 cm inner diameterspherical proportional counter with walls made from A-150 tissue equivalentplastic. The chamber was filled to 4.44 kPa pressure with propane based tissue-equivalentgas mixture to simulate a tissue volume with a diameter of 1 µm. Pulses fromthe detector were fed to preamplifier and multichannel analyzer. Measurements wereperformed in a 51x51x47 cm3 PMMA-walled water phantom. The measured spectrawere processed with an in-house written MatLab script to determine the gamma rayand neutron absorbed dose to brain tissue. Measurements were made at 1-10 % ofthe full power of the reactor due to the high sensitivity of the detector, and were scaledusing the beam monitoring system as reference. Manganese and gold foil measurementswere performed to calibrate the beam monitors for low reactor power. Resultsobtained using microdosimetry were compared with results from a DORT simulationand from dual ionisation chamber measurements.Results and Discussion: The gamma dose rates measured with ionisation chamberand calculated with DORT were in good agreement. The gamma ray doses measuredwith TEPC were systematically approximately 10 % lower, which is roughly within theestimated uncertainties. The neutron dose rates measured with TEPC agreed withDORT simulation within 10 % limits. Ionisation chamber technique gave approximately10-30 % lower neutron dose rates when compared to the other two methods. At theusual clinical beam intensity the chamber used for monitoring the beam intensity issomewhat saturated. Thus, at the intensities used in these measurements, the monitorchamber is more sensitive, and it has to be calibrated using gold and manganeseactivation foils. It was assumed that gamma dose rate follows the same pattern asneutron dose rate as related to the monitor count rate.Conclusions: The results from a dosimetric comparison using (1) tissue equivalent proportionalcounter microdosimetry, (2) dual ionisation chamber technique and (3) DORTcomputer simulation to determine gamma ray and neutron absorbed doses in waterfilled PMMA phantom are presented. Differences in the absorbed doses determinedwith the three methods were within the limits of the stated uncertainties. These resultssupport the reliability of the methods.12 - Enhanced blood boron concentration estimation forBPA-F mediated BNCT. M. Kortesniemi a,b, *, T. Seppälä b , I.Auterinen c , S. Savolainen a,b,da HUS Helsinki Medical Imaging Center, University of Helsinki, POB 340, FIN-00029HUS, Finland, b Boneca Corporation, POB 700, FIN-00029 HUS, Finland, c VTTProcesses, POB 1608, FIN-02044 VTT, Finland, d Department of Physical Sciences,POB 64, FIN-00014 University of Helsinki, FinlandThe first Finnish clinical trial in BNCT using BPA-F (boronophenylalanine fructose) asB-10 carrier was started in 1999. By the end of the year 2003, 28 patients have beentreated at the FiR 1 BNCT facility in Otaniemi, Espoo, Finland. The blood boron concentrationregulates directly the BNCT irradiation time in which the prescribed dose tothe patient is delivered. Therefore a proper estimation of the blood boron concentrationfor the treatment field based on the measured blood samples before irradiation is required.The motivation of this study was to enhance the existing bi-exponential methodand apply it directly to the treatment data flow.Prior to the BNCT irradiation the patients were i.v. infused with BPA-F for approximatelytwo hours. The whole blood boron concentration was measured using ICP-AES during the treatment procedure with about 20 minutes intervals. However, bloodsamples could not be taken during the irradiations. The bi-exponential function wasused to model the blood boron kinetics. The minimum of Chi-square value was determinedin an iteration sequence applying Levenberg-Marquardt algorithm. The infusionphase data points were used in an optional fit producing a bi-exponential saturationcurve, which could be used to estimate the infusion end value for the clearance phasefit if the infusion end data was lacking. The uncertainty of the average blood boronconcentration was determined according to covariance matrix and analytical partialderivatives of the model function.The harmonic mean bi-exponential decay half-lives of the studied patient data (n=28patients) were 15+/-8 and 320+/-70 minutes for the faster and slower half-life. Themaximum and mean differences of the estimated mean boron concentrations duringthe irradiation field between the initial estimate (before the field) and the final estimate(using all data points) were 1.1 µg/g and 0.2 µg/g for both fields, respectively. Theuncertainty of the average blood boron concentration during the irradiations was 0.7+/-0.1 µg/g (ppm), being 5% or less of the corresponding blood boron concentration.The mean and maximum differences between the measured and estimated boronconcentrations (n=28) were 15 and 18 times lower, respectively, compared to thepreviously published data (n=10) for the first irradiation field estimate (Ryynänen etal. 2002). Utilization of the infusion phase data also in the bi-exponential algorithmimproved the initial clearance phase fit if the infusion end data was lacking or the measureddata had other additional uncertainties. Otherwise the infusion phase data hadless than 0.1 µg/g effect on the blood boron estimates during the irradiation.The implemented modeling algorithm provides a robust method for temporal bloodboron concentration estimation for BPA-F mediated BNCT. Utilization of the blood borondata from the infusion phase further improves the reliability of the estimate. Theuncertainties of the estimated blood boron concentrations during the treatment irradiationsbased on the parameter uncertainties are at acceptably low level, about 5%. Theoverall data flow during the treatment fulfills the practical requirements concerning theBNCT procedure.14 - The TL analysis methods used to determine absorbedgamma doses in vivo for the BNCT patients treated atFiR 1. J. Karila a,* , T. Seppälä b , T. Serén c , P. Kotiluoto c , I. Auterinen c , C.Aschand, L. Kankaanranta e , S. Savolainen a,fa Department of Physical Sciences, FIN-00014 University of Helsinki, Helsinki, Finland,b Boneca Corporation, FIN-00029 HUS, Helsinki, Finland, c VTT Processes,Technical Research Centre of Finland, FIN-02044 VTT, Espoo, Finland, d Departmentof Physics, Finnish Institute of Occupational Health, FIN-00250 Helsinki, Finland,e Department of Oncology, Helsinki University Central Hospital, FIN-00029 HUS, Helsinki,Finland, f Medical Imaging Center, University of Helsinki, FIN-00029 HUS, Helsinki,FinlandThe gamma dose determination in vivo for the patients treated with BNCT at the Finnishresearch reactor (FiR 1) has been performed with thermoluminescent (TL) dosimeters.The gamma dose determination is complicated by the thermal neutron sensitivityof the TLDs, especially when TL equipment capable for glow curve analysis is notavailable. Still, the methods used have undergone some development.The aim of the study was to evaluate and compare the TL analysis methods used tocorrect the measured TL signal to obtain absorbed gamma dose. An enhanced TLmethod is presented as well. The results are compared with those calculated by treatmentplanning system (TPS).27

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - P hy s i c sThe measurements were performed with MCP-7s TLDs (TLD Niewiadomski & Co.,Krakow, Poland). Mn-Al activation detectors were used to measure 55Mn(n,g) activationreaction rate needed for the neutron correction of the TLDs. The measurementswere carried out during irradiations of 21 glioblastoma patients treated with BNCTunder the clinical research protocol (P01). The measurement points were located inthe ipsilateral auditory canal, at the transaxial tattoo point on opposite side of the head,and in between the eyes.The absorbed gamma dose can be get from the TL signal by utilizing several conversionand correction factors. The correction due to thermal neutrons can be made bysubtracting their contribution from the total TL signal. In the first TL analysis method,this was done by multiplying the thermal neutron sensitivity of the TLDs, for which anapproximated value was available, by thermal neutron fluence corrected by the ratio ofmeasured and calculated 55Mn(n,g) activation reaction rates. In the second method,the thermal neutron correction was made using an experimentally determined factorwith e.g. 6Li kerma dose rate. In these methods, the total dose absorbed to the TLDand the dose component caused by thermal neutrons were actually expressed in differentunits (absorbed dose to ICRU adult brain and LiF kerma, respectively) becauseof the lack of conversion factor to equalize the two. Therefore, an enhanced methodwas introduced in which this deficiency was eliminated.The dose calculations were made with the same TPS (primarily BNCT_Rtpe) withwhich the treatment plans had been created prior the BNCT irradiations.The results obtained with modelling agreed mainly fairly well with the doses measuredin vivo determined according to the enhanced TL analysis method, although high differences,especially in the auditory canal, were also observed. Such differences werepartly caused by e.g. the uncertainty of dose calculations on skin due to the largetallies.The differences between the three TL methods were surprisingly small. The enhancedTL analysis method presented is the most accurate of the methods discussed in determininggamma doses in vivo in BNCT. In future, this method will be used until acalculation-independent method is available.16 - Combined utilization of 10B and 157Gd in NCT. Physicalmeasurements. G. Gambarini a,b *, V. Colli a,b , M. Cortesi a,b , U.Danesi a,d , S. Gay a,b , R. Rosa c , G. Rosi ca Department of Physics of the University, Milan, Italy, b INFN, National Institute ofNuclear Physics, Italy, c FIS-ION, FIS-NUC, ENEA, Italy, d National Neurologic Institute“C. Besta” - Milan, ItalyThe employment of gadolinium in NCT is receiving more and more interest owing tothe efficient killing effect of the Auger electrons, which have demonstrated higher celllethality than alpha particles. Nevertheless, the effect of gamma-radiation emitted inthe reaction of thermal neutrons with 157Gd has to be seriously considered. The useof a combination of boron and gadolinium in NCT has shown promising potentialities.In this work, dosimetry measurements have been performed in a phantom with dimensionnear a human head, containing large regions with accumulations of 10B or/and157Gd. Dose profiles have been drawn utilising gel dosimeters and the various dosecomponents have been separated. Moreover, a few measurements have been carriedout for inquiring the feasibility of utilising neutron radiography for in-real-time control ofthe decrease of the total concentration of boron during radiotherapy treatment, if boronand gadolinium are bound to the same carrier. Small tissue-equivalent samples (water,gel, plastic), with inside a vial containing a concentration of 10B or 157Gd, have beenutilised to this purpose. Finally, the MR images of water phantoms containing variousconcentrations of gadolinium have been detected and studied, to investigate if quantitativedetermination of boron concentration could be possible.18 - Determination of Gadolinium in Biological Materialby Neutron Activation Analysis. V.I.Kvasov a , V.N.Kulakov b ,V.F.Khokhlov b , K.N.Zaitsev a , K.P.Alekseev ba Moscow Engineering Physics Institute, Moscow, Russia, b SRC–Institute of Biophysics,Moscow, RussiaCompounds with gadolinium are applied in neutron capture therapy (NCT) due tothe large neutron capture cross-section of the 157Gd nucleus. The drugs Dipentastand Magnevist are intercellular agents; they are excreted from the organism mainlyunchanged [1]. Therefore, it is sufficient to measure the Gd concentration to obtainprecise information on their content in biological material.We used nuclear characteristics of the gadolinium nuclides most suitable for the analyticalpurposes [2,3] of the neutron activation analysis method (NAA).NAA is possible on three radioactive products of activation of the nuclides 157Gd,153Gd, 159Gd, 161Gd. The detection limit is a value from 0.01 µg for 159Gd to 0.2 µgfor 161Gd [4]. We designed two versions of NAA techniques with the analytical radionuclide161Gd (g-lines: 102.3 and 360.9 keV) at the MEPhI reactor. NAA technique onthe horizontal channel. Samples of Dipentast (2 mL) dissolved in ethanol with a knownGd content in sealed polyethylene vials were irradiated for 10 minutes at the outlet ofthe tangential channel HEC-4.Thermal neutron flux was 6.4(2) 108 n/cm2s, and 198Au cadmium ratio – 4.0(2). Eachsample was irradiated together with a neutron flux monitor (NFM) from the certified setof activation detectors AKN-T. Samples of biomaterials (kidney, brain, liver, blood, andmuscles of experimental mice) containing Dipentast were exposed to the same procedure.The samples were measured 3 minutes after irradiation, NFM - in 2-3 hours.The measurements continued 600 s on a -spectrometer with a resolution of 3.5 keV forthe 60 Co 1332.5 keV line. The geometry of measurements was selected so that thesource could be conditionally considered as a point source. Gd detection limit: 20 µg.NAA technique on the vertical channel in the thermal column of the reactor. Thermalneutron flux in the irradiation position was 2.2(2)°—1011 n/cm2°—s; 198Au cadmiumratio - 400(20). Each sample was also irradiated with a NFM for 5minutes. After 5minutes the samples were measured on a -spectrometerfor 600 s. Gd detection limitin this technique: 1 µg.The studies were carried on non-enriched Gd with pure materials, practically withoutnoise and interference. The sensitivity of the analysis may be enhanced by optimizationof the technique and modernization of the -spectrometer.The work is carried out with financial support from the Russian Fund for TechnologicalDevelopment and ISTC, Project # 1951.References* Magnevist, by Ed. R.Felix, A.Heshiki, N.Hosten, H.Hricak, Oxford, Blackwell Scientific Publication,1994, 192 p.* Mednis I.V. Cross-sections of nuclear reactions used in neutron activation analysis: Referencebook. – Riga: Zinatne,1991. – 119 p.* Groshev L.V., Demidov A.M., Lutsenko V.N., Pelekhov V.I. Atlas of gamma-ray spectra from radiativecapture of thermal neutrons. – M: Atomizdat,1958. - 11 p.* Pelekis L.L., Pelekis Z.E., Taure Ya. J. Radioanal. Chem.,1973. - Vol.15, P.497- 507.20 - BNCT of an Explanted Liver: Dose Calculations withSERA. Gabriele Hampel a , Arturo Lizon Aguilar b , Rüdiger Behrendt b , WolfgangBernnat c , Klaus Eberhardt a , David Nigg d , Charles A. Wemple da Institut für Kernchemie, Universität Mainz, D-55099 Mainz, Germany, b Departmentof Nuclear Medicine, Medical University of Hannover, Carl-Neuberg-Str. 1,D-30625Hannover, Germany, c Institute for Nuclear Energetics and Energy Systems (IKE),D-70550 Stuttgart, Germany, d Idaho National Engineering and Environmental Laboratory(INEEL), Idaho Falls, Idaho, USABoron neutron capture therapy (BNCT) shall be applied for an explanted organ withmetastases at the TRIGA reactor at Mainz, Germany. After the treatment of the patientwith 10B pharmaceutical the liver will be explanted, irradiated in the thermal column,and then reimplanted (autotransplantation). This was first successfully done in December2001 for a 48 year old male patient at the TRIGA reactor in Pavia, Italy. Fora treatment with BNCT it is necessary to determine the total and partial dose components,e.g., boron and gamma dose. Since some parts of the dose can not be readilymeasured, calculations are necessary. For this procedure, the program SimulationEnvironment for Radiotherapy Applications (SERA) developed by INEEL, can be applied,which allows an individual dosimetry calculation for the patient.The SERA program consists of a manual and semi-automated geometric modelingof the treatment objects derived from MRI, CT, or other medical imaging modalities.The dose for these geometric models is calculated with the INEEL radiation transportcomputer code and isodose contour data was analyzed.The SERA program was adapted at MHH to calculate the radiation dose in the case ofliver autotransplantation. Also, the technical characteristics and the physical environmentof the thermal column were modeled to provide the input source term for SERA.The object was to determine the optimal geometry of the thermal column in order toobtain the best results for the application of BNCT on the liver.Special files were generally constructed for a specific neutron and coincident gammasource. For this project the determination of the source characteristics has been developedby the IKE using the MCNP code. Models were developed by IKE experts for theMHH TRIGA reactor core and the thermal column of the Mainz TRIGA reactor.The center point of the planar source has been situated at the beginning of the thermalcolumn, next to the graphite reflector, because a simple model can be generated inthis way. The source was modeled as a disk source with 1/E energy interpolation and10 concentric annular regions. Since SERA calculates the different dose componentsfor the irradiated organ, the total dose, boron dose, and gamma dose was used tooptimize the beam for the liver with metastases. The optimal configuration of the ther-28

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - P hy s i c smal column produces a low total dose in the normal tissue, a low gamma dose in thenormal and tumor tissue and a high boron dose in the tumor tissue.Several thermal column geometries were evaluated during the optimization studies.Bismuth, lead, and lead/bismuth moderator regions of varying thickness were examinedto minimize gamma dose, and a bismuth collimator was also included.22 - High resolution NMR spectroscopy and relaxationstudies of 19F-BPA. R. Campanella a , S. Capuani b , P. Porcari b , L.Menichetti c , and B. Maraviglia ba Dipartimento di Fisica and INFM, Università di Perugia, 06123 Perugia (Italy),b Dipartimento di Fisica and INFM, Università “La Sapienza”, 00185 Roma (Italy),c CNR - Institute of Clinical Physiology, 56124 Pisa (Italy)Introduction: Biodistribution and pharmacokinetics studies of boron compounds are ofoverwhelming importance for the definition of treatment planning in BNCT. Althoughseveral techniques have been proposed for this, imaging methods like Magnetic ResonanceImaging (MRI) and Positron Emission Tomography (PET) are superior for thispurpose, due to their inherent ability to map the distribution of boron compounds intissues with a very high spatial resolution.The aim of our research is to measure the spatial distribution of borono-phenylalanine(BPA) in tissues by means of MRI. MRI of boron compounds is not easily feasible,mainly on account of the very low concentration of the compounds in therapeutic situation,and moreover of the low gyromagnetic ratio and the high quadrupole momentof boron. NMR localized spectroscopy of BPA has been previously realized,[1] but themethod is not applicable for imaging. BPA distribution can instead be imaged usingits fluoridated analogue 19F-BPA. This choice is due to the relevant characteristics of19F, namely its high sensitivity and the absence of quadrupole moment. The presentwork is a preliminary study which aims to give information about the NMR parametersof 19F-BPA, in order to establish the best imaging procedure.Experimental and results: The measurements have been performed on an Avance 400NMR spectrometer.1) 19F-BPA characterization- The compound has been characterized by 19F and 1Hhigh resolution spectra. The 19F spectrum exhibits a single line, while the 1H spectrais similar to the hydrogen spectra of BPA, that has been already reported.2) 19F-BPA-fructose complex- The compound has been characterized by 19F, 13Cand 1H HR spectra. In the 13C spectrum the fructose C(2) resonance at 101 ppmshifts at 110 ppm, showing that the 19F-BPA is completely complexed with fructose[2].The 19F spectrum exhibits three different peaks, at -10.4, -8.0 and -6.4 ppm, whichseem to be originated by three different isomeric forms[2]; the peak at 10 ppm is muchmore intense than the other.The fluorine relaxation times, T1 and T2, have been measured for the various peaks,giving values in the ranges (600-1000) ms for T1, and (18-60)ms for T2.3) 19F-BPA-fructose complex in blood. - To determine the complete characteristics ofthe compound, it has then been dissolved in rat blood. In this case the 19F HR spectrumand the fluorine relaxation times have been measured. The results are similar tothe previous case.When a TLD is exposed to NCT neutron beams, the energy in it stored is substantiallydue to both gamma-radiation (from 1H(n,gamma)2H reactions in phantom orfrom reactor background) and charged particles (from thermal neutron interactionsinto the dosimeter itself). The relative dose contributions in TLDs depend on their isotopiccomposition. Consequently, a couple of TL materials with appropriate isotopiccomposition permit to achieve reliable information about gamma-dose and thermalneutron fluence.The scarce sensitivity of CaF2:Tm detector (TLD-300) to low energy neutrons hassuggested using this phosphor as gamma-dosimeter in thermal or epithermal neutronbeams. The glow curve shapes after gamma irradiation or thermal/epithermal neutronexposures has been studied: this comparison has shown that, in NCT neutron fields,slow neutrons produce negligible effects in TLD-300 in comparison with to the gammacomponentof the field. The calculation of the energy released by thermal neutrons inTLD-300 has confirmed that in NCT beams the TLD-300 signal due to thermal neutronsis lower than the experimental error. Also the effect of internal irradiation hasbeen studied, in order to verify that it produces a negligible increase of the dosimeterresponse. The dependence of TLD-300 sensitivity on photon energy has been investigatedtoo: a noticeable increase of TL emission for energies lower than 150 keV hasbeen evinced; nevertheless, the gamma-dose measurements performed by TLD-300in the BNCT columns of TAPIRO reactor are consistent with those obtained by otherkinds of dosimeters.It is known that LiF:Mg,Ti dosimeters are advantageous for measuring thermal neutronfluences, owing to the high cross section of the reaction 6Li(n,alpha)3H (sigma=945b). If these dosimeters are properly calibrated, and the possible gamma-contributionto their response is correctly subtracted, the thermal neutron fluence can be reliablyevaluated. In the case of high thermal neutron fluxes, radiation damage effects appear,which are not repaired by annealing procedures. The different 6Li percentage ofTLD-700 (about 0.01 %), TLD-100 (7.5 %) and TLD-600 (95.6 %) causes very differentresponses to thermal neutrons of LiF:Mg,Ti detectors. For the various kinds of LiF:Mg,Ti phosphors, the response versus thermal fluence has been studied in order todetermine the maximum admitted fluence to avoid radiation damage effect. Choosingsuitable reactor power and exposure time for each kind of TLD in order to remain belowthe fluence limits, LiF:Mg,Ti dosimeters have given results in good agreement withfluence values obtained by means of activation techniques or deduced by the chargedparticle doses measured with gel dosimeters.29Conclusion: This work demonstrates that 19F-BPA can be imaged by means of 19FMRI; the imaging strategy must take into account the short value of the spin-spin relaxationtime, which can be estimated in the order of a few tens of milliseconds.Acknowledgments: We gratefully acknowledge Prof. R. Marini Bettolo, Dr. L.M. Mignecoand Dr. A. La Bella of the Department of Chemistry, University “La Sapienza”,Rome for the preparation of the boron compounds. The work is supported by ItalianMinistry of Research, PRIN 2001.References1) Zuo et al, MedPhys 26, 1230 (1999)2) Shull et al, JPS, 89, 215 (2000)24 - In-phantom dosimetry in BNCT by means of TLDs.G. Gambarini a , S. Gay a , G. Rosi b , L. Scolari aa Department of Physics of University and INFN, Milan, Italy, b FIS-ION, ENEA, Casaccia(Rome), ItalyA dosimetric method based on thermoluminescent detectors (TLDs) was investigated,in order to answer to the requirements of BNCT concerning experimental determinationof absorbed dose in tissue-equivalent phantoms, irradiated in thermal or epithermalneutron beams. The proposed method takes advantage of the combination oftwo types of TL materials, calcium and lithium fluorides, which give the possibility ofmapping the gamma-dose and the thermal neutron fluence respectively. From suchfluence, knowing the KERMA factors, the in-phantom doses due to all charged particlescreated by thermal neutron interactions can be simply calculated.

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - 2:00 PMPosters Session - C l i n i ca l A p p l i ca t i o n s2 - Quality control of BPA-f for Harvard-MIT BNCT ClinicalTrial. Y. Shibata*, R.G. Zamenhof, P. Busse, H. PatelDepartment of Radiology, Radiation Oncology, Beth Israel Deaconess Medical Center,Harvard Medical School, 330 Brookline Ave. Boston MA, 02215 USABoronophenylalanine fructose (BPA-f) is not a commercially available manufactureddrug and has been synthesized in our laboratory using the method originally developedin Japan and Brookhaven National Laboratory. The quality of synthesized BPA-fshould be controlled by strict tests in order to meet Food and Drug Administration(FDA) and Internal Review Board (IRB) mandated guidelines. In this report we describeour method of synthesis and quality control of BPA-f solution prepared for Harvard-Massachusetts Institute of Technology (MIT) Boron Neutron Capture Therapy(BNCT) Clinical trial.BPA was purchased from Ryscor science Inc. and stored in secure locked shelf. Frombody surface area, total amount of BPA, fructose and infusion volume were calculated.All re-usable equipment such as glassware and instruments were autoclaved. Itemssuch as magnetic stirrers, stands, peristaltic pumps were placed in the clean biohazardhood and sterilized with ultraviolet lamp overnight. Working area in the hood wascleaned with 70% isopropyl alcohol. The hands were washed with 4% chlorhexidinegluconate soap and ware gloves.Measured PBA powder was poured into a sterile receiver and sterile water was added.The pH of the solution was measured using electrical pH meter and 10 N NaOH solutionwas added until pH reach 11.5. Measured fructose powder was added and thesolution was stirred using magnetic stirrer. The solution was filtrated with 0.45 umfiltration unit with a pre-filter. HCl solution was added to bring pH to 7.4. The BPA-fsolution was filtrated again using a 0.45 um filtration unit. Sterile water was addedto adjust final volume. The solution was filtrated using pyrogen filter and peristalticpump. Filtrated solution was refrigerated over-night. On the next day, BPA-f solutionwas filtrated through 0.2 um filtration unit and transferred into sterile infusion bag usinga vacuum filler assembly. All bags and vials were labeled by patient name, ID No.,infusion date, etc.Two batches of BPA-f solution were synthesized for each patient. Small test sampleswere taken from each batch and were send to the test facility for pyrogen and sterilitytest. Pyrogen test results were reported on same day and sterility test results werereported on day 3, 7 and 14. Only the product that passed both of pyrogen and sterilitytest is infused to the patient.From Dec 2002 to May 2003, 9 patients with glioblastoma or cutaneous melanomawere included in Harvard-MIT BNCT Clinical trial. Eighteen batches of BPA-f solutionwere made and only one batch fail to pass pyrogen test. This BPA-f solution wasfiltrated with another pyrogen filter and passed the test. No complication related withBPA-f infusion was found in any patients.Our method of synthesis of BPA-f solution was safe. Contamination during synthesisor at test facility should be controlled. The present methods of quality control madeenough quality assurance for human clinical trials of BNCT.10-min transmission scan of the head was performed before injection of the tracer usingrobotically operated 68Ge rods.Results: All tumors accumulated [18F]FBPA. The average standardised uptake value(SUVave) was 3.8 (range, 2.4 to 8.2). The SUVave was 3.5 (range, 3.2 to 5.0), 3.2(range, 3.1 to 7.3), and 4.1 (range, 2.4 to 8.2) in head and neck carcinomas, glioblastomas,and anaplastic astrocytomas, respectively. Of the normal tissues the skin andthe mucous membranes showed marked tracer accumulation. Patients treated withBNCT had decreased tumor uptake of [18F]FBPA in PET studies following BPA-basedBNCT.Conclusions: The results suggest that most recurred head and neck carcinomas, glioblastomasand anaplastic astrocytomas take up [18F]FBPA, and that tumor [18F]FBPAuptake may decrease following BPA-based BNCT. [18F]FBPA-PET may be a usefulmethod to estimate tumor and normal tissue BPA uptake, and deserves further evaluationin selection of patients for BPA-based BNCT.KEY WORDS: Positron Emission Tomography, Boron Neutron Capture Therapy, Glioma,Head and Neck Cancer, Boronophenylalanine304 - L-[18F]F-BPA PET of gliomas and head and neck cancers.L. Kankaanranta a , K. Havu b , J. Collan a , J. Vähätalo c , M. Kouri a , H.Minn b , H. Joensuu aa Department of Oncology, HUCH, Helsinki, Finland, b Turku Pet Centre, Turku, Finland,c Clinical Research Institute HUCH, Helsinki, FinlandBackground: Patients diagnosed with recurred high-grade glioma or head and neckcarcinoma are potential candidates for boronophenylalanine (BPA)-mediated BNCTprovided that tumor uptake of BPA exceeds that of the normal tissues. We havemade an attempt to estimate tumor BPA uptake with PET using 18F-labeled BPA (L-[18F]BPA) as the tracer. One ongoing Finnish phase I protocol, which evaluates BNCTin the treatment of inoperable recurred head and neck cancer, requires that whenever[18F]BPA-PET has been done prior to BNCT, at least 2.5 times more of [18F]BPAneeds to accumulate in the tumor than in the corresponding contralateral normal tissue.Patients and methods: Five patients with recurred head and neck carcinoma and 10with glioma were studied with PET using [18F]BPA as the tracer between Oct 2001and Jan 2004. Three of the glioma patients had glioblastoma and 7 anaplastic astrocytoma,of which 1 had oligodendral and 1 oligoastrocytic features. The median agewas 53 years (range, 28 to 63), and 9 were male. Ten patients had received prior radiotherapyand 5 prior chemotherapy. All patients had gadolinium contrast MRI performedwithin 4 weeks prior to PET. Four patients were recruited to the ongoing BNCT protocolsafter performing [18F]FBPA PET. 18F was produced using an 103 cm isochronousMGC-20 cyclotron from the 18O(p,n)18F reaction at the Accelerator Laboratory, TurkuPET Centre, Finland. [18F]FBPA was subsequently synthesised from 4-borono-L-phenylalanineas the precursor. Imaging was performed with a 35-slice GE-Advance PETscanner (GE Milwaukee, WI). The scanner consists of 18 rings of bismuth germanatedetectors yielding 35 transverse slices spaced by 4.25 mm. The imaging field of viewis 55 cm in diameter and 15.2 cm in axial length. To correct for photon attenuation, a

31Session Chairs: George Kabalka, Shin-ichiro Masunaga3:00 PM -The syntheses and in vivo biodistribution of novel boronatedunnatural amino acids. G.W. Kabalka*, Z.Z. Wu, M-L.Yao, N. NatarajanDepartments of Chemistry and Radiology, The University of Tennessee, Knoxville, TN37996-1600, USAThe clinical success of boron neutron capture therapy (BNCT) is dependent on the selectivedeposition of 10B in tumor cells. To date, a variety of molecules have been usedto deliver boron to tumors. These include carbohydrates, polyamines, amino acids,nucleosides, antisense agents, porphyrins, and peptides. In recent years, encouragingclinical results have been obtained using 4-dihydroxyborylphenylalanine (BPA) as thetumor specific boronated agent. It is believed that the amino acids are preferentiallytaken up by growing tumor cells. Positron emission tomography (PET) investigationsusing carbon-11 labeled 1-aminocyclobutane-1-carboxylic acid (ACBC) demonstratedthat this unnatural amino acid localizes in tumors more avidly than BPA. For this reasonwe have focused our efforts on unnatural amino acids as boron carriers. Severalboronated amino acids were prepared and their in vivo biodistribution determined.The unnatural amino acids were prepared utilizing borylation reactions developed inour laboratory. The majority of the syntheses utilize metal catalyzed additions of diboronagents to unsaturated carbonyl compounds. The remaining boronated amino acidswere synthesized using classic transmetallation procedures.Biodistribution results in mice bearing melanoma tumors indicated that all compoundswere taken up by the tumor. The data for the cyclic five membered ring analogue,1-amino-3-boronocyclopentanecarboxylic acid, was most striking, exhibiting a nearly22:1 ratio of boron concentration for tumor to brain at the two hour time point, droppingto 7.3 after six hours. The tumor to blood and tumor to skin ratios were also quitehigh. It is important to note that all of the amino acids were synthesized as racemicand diastereomeric mixtures. Thus there is a high probability that a single enantiomerof 1-amino-3-boronocyclopentanecarboxylic acid might exhibit a selectivity approaching80:1!3:20 PM -The usefulness of 2-nitroimidazole-sodium borocaptate-10Bconjugates as 10B-carriers in boron neutroncapture therapy. S. Masunaga a, *, H. Nagasawa b , M. Hiraoka b , Y.Sakurai a , Y. Uto b , H. Hori b , K. Nagata a , M. Suzuki a , A. Maruhashi a , Y. Kinashia , K. Ono aa Radiation Oncology Research Laboratory, Research Reactor Institute, Kyoto University,2-1010, Asashiro-nishi, Kumatori-cho, Sennan-gun, Osaka 590-0494, Japan, bFaculty of Engineering, University of Tokushima, Tokushima 770-8506, JapanPurpose: We evaluated the usefulness of 5 new 10B-compounds (TX-2016, TX-2017,TX-2018, TX-2041, and TX-2042) as 10B-carriers in boron neutron capture therapy(BNCT).Materials and Methods: These 5 compounds are 2-nitroimidazole-sodium borocaptate-10B(BSH) conjugates, that is, hybrid compounds that have both gamma-raysensitizingunit, 2-nitroimidazoles and thermal neutron-sensitizing unit, BSH. The newcompounds and BSH dissolved in physiological saline were administered to tumorbearingmice intraperitoneally. At various time points after administration, the 10Bconcentrations in tumors and blood were measured by prompt gamma-ray spectrometry.For a subsequent tumor-irradiation study, other tumor-bearing mice received 5-bromo-2’-deoxyuridine (BrdU) continuously via implanted mini-osmotic pumps to labelall proliferating (P) cells in the tumors. Based on the data of 10B distribution studies,TX-2041 was chosen. To obtain similar intratumor 10B concentrations during neutronor gamma-ray exposure, irradiation was started from the time point of 20 min afterinjection of BSH at a dose of 125 mg/kg, or 75 min after injection of TX-2041 at a doseof 400 mg/kg. The tumors were then excised, minced and trypsinized. The tumor cellsuspensions thus obtained were incubated with cytochalasin-B (a cytokinesis blocker),and the micronucleus (MN) frequency in cells without BrdU labeling ( = quiescent (Q)cells) was determined using immunofluorescence staining for BrdU. Meanwhile, theMN frequency in total (P + Q) tumor cells was determined from the tumors in mice thatwere not pretreated with BrdU. The cell survival assay was also performed in micegiven no BrdU.Results: TX-2041 kept the 10B concentration in solid tumors sufficiently high evenafter the 10B concentrations went down when other compounds including BSH wereadministered. Further, it did not show so high a 10B concentration in the blood. Thus,TX-2041 was thought to be most suitable for the subsequent neutron beam irradiationstudy because it took at least 45-60 min to deliver sufficient radiation doses with ourresearch reactor. The enhancement effects on both total and Q cells with neutron irradiationin combination with TX-2041 were significantly larger than those with BSH evenEleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - PMParallel Session 3 - B i o l o g ywhen the 10B concentrations in tumors were kept at similar level during irradiation.However, the tendency to enhance the sensitivity of total cells more markedly thanthat of Q cells was more clearly observed for TX-2041 than for BSH. This means thatTX-2041 widened the sensitivity difference between total and Q cells more remarkablythan BSH, although the enhancement effect by TX-2041 was larger than that by BSH.On the other hand, TX-2041 had enhancement effects in both cell populations withgamma-ray irradiation, which BSH could never bring about.Conclusion: From the viewpoint of the characteristics required to localize a sufficientamount of 10B into tumors and to keep the 10B concentration high during neutronbeam irradiation, TX-2041 was the most advantageous compound of the 5 new 10Bcarriersin BNCT. Moreover, TX-2041, with a reliable gamma-ray-sensitizing effect ontumors, is a promising candidate as a 10B-carrier in BNCT, considering that currentlyavailable research reactor thermal neutron beams inevitably include gamma-ray contamination.3:55 PM -Nuclear Targeting in NCT: Synthesis and ComparativeEvaluation of Nuclear Localization Sequences Bound toBoronated Porphyrins and Carborane Oligomers. StephenB. Kahl a , Paola Dozzo a , Myoung-Seo Koo a , Trudy Forte b , StephanieM. Berger b , Kathy Bjornstad b , and Eleanor Blakelya Department of Pharmaceutical Chemistry, University of California, San Francisco, bLife Sciences Division, Lawrence Berkeley National Laboratory, BerkeleyThe nucleus represents the “Holy Grail” for targeting binary sensitizers, including thosefor NCT and PDT. Numerous studies, both theoretical and empirical, have demonstratedthat delivery of such sensitizers to the nuclear can provide a substantial increase inlethal radiation damage resulting from both induction of apoptosis and necrosis. Thenuclear targeting of proteins and other macromolecules has been extensively studiedand relies on the presence of short peptide sequences known collectively as nuclearlocalization sequences (NLS).We have previously reported preliminary studies on the synthesis and uptake in SF-767 human glioma cells of a conjugate prepared from the meta-carborane analog ofBOPP (m-BOPP) and the SV40 T antigen NLS. The present paper will describe thesynthesis and comparative uptake properties of a series of four new peptide NLS-m-BOPP conjugates. Also described will be the synthesis and uptake characteristics oftwo examples of p-carborane amino acid oligomers bearing the SV40 NLS sequenceand a reporter probe. Finally, we will also present evidence of the association of theseamphiphilic complexes with low density lipoproteins.4:15 PM -Biodistribution of p-borophenylalanine (BPA) in dogswith spontaneous undifferentiated thyroid carcinoma(UTC). M.A. Dagrosa a, *, M. Viaggi a , R. Jimenez Rebagliati b , V.A. Castillod , D. Batistoni b , R.L. Cabrini a , S. Castiglia c , G.J. Juvenal a , M.A. Pisarev a,ea Department of Radiobiology, University of Buenos Aires, 1429 Buenos Aires, Argentina,b Department of Chemistry, University of Buenos Aires, 1429 Buenos Aires,Argentina, c Division of Radiopharmacy, National Atomic Energy Commission, Universityof Buenos Aires, 1429 Buenos Aires, Argentina, d Veterinary School of Medicine,University of Buenos Aires, 1429 Buenos Aires, Argentina, e Department of Biochemistry,University of Buenos Aires, 1429 Buenos Aires, ArgentinaIntroduction: Human undifferentiated thyroid carcinoma (UTC) is a very aggressivetumor, which lacks an adequate treatment. The mean survival time is of around 12months. In previous studies we demonstrated that the UTC cell line ARO has a selectiveuptake of p-borophenylalanine (BPA), both in vitro and after transplanting theminto nude mice. More recently we applied the complete boron neutron capture therapy(BNCT) to mice bearing this tumor. A 100 % control of tumor growth and a 50% histologicalcure of tumors with an initial volume of 50 mm3 or less were observed. As afurther step towards the potential application in humans we have performed the presentstudies with larger animals. Therefore we performed biodistribution studies of BPAin dogs with spontaneously UTC, and the present results confirm that spontaneousUTC has a selective uptake of this boron compound.Materials and Methods: Four dogs with diagnosis of spontaneous UTC were studied. ABPA-fructose solution (BPA-F) was infused during 60 min and dogs were submitted tothyroidectomy. Samples of blood were obtained throughout and samples from differentareas of the tumors (and in one dog from normal thyroid) were obtained at 130-150min. Boron concentration in the samples was determined by ICP-OES.Results: A peak of blood boron concentration was observed immediately at the endof the infusion.

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - PMParallel Session 3 - B i o l o g y4:35 PM -The effect of dexamethasone on the uptake of p-boronophenylalaninein the rat brain and intracranial 9L gliosarcoma.G.M. Morris a,b , P.L. Micca b , J.A. Coderre b,c, *a Normal Tissue Radiobiological Research Group, Research Institute, Churchill Hospital,Oxford OX3 7LJ, UK, b Medical Department, Brookhaven National Laboratory,Upton, NY 11790, USA, c Nuclear Engineering Department, Massachusetts Instituteof Technology, 150 Albany Street, Cambridge, MA 02139, USAThe steroid dexamethasone (DEX) is routinely used to treat edema in brain tumorpatients. The objective of the present study was to evaluate the effects of DEX on theuptake of boronophenylalanine (BPA) using the rat 9L gliosarcoma tumor model andsurrounding brain tissue. Two steroid dosage protocols were used. The high dose DEXprotocol involved five 3 mg/kg intraperitoneal injections at 47, 35, 23, 11 and 1 hourprior to the administration of the BPA for a total dose of 15mg DEX/kg rat. The lowdose DEX administration protocol involved two doses of 1.5 mg/kg at 17 hours and1 hour prior to BPA injection for a total dose of 3 mg DEX/kg rat. The control animalsreceived no pretreatment, prior to the administration of BPA. Seventeen days aftertumor implantation, rats were injected i.p. with 0.014 ml/g body weight BPA solution(1200mg BPA/kg; ~59 mg 10B/kg). In all groups, rats were euthanized at 3 h after BPAinjection. Tumor:blood boron partition ratios for the control, low and high dose DEXgroups were 2.3, 2.3 and 2.5, respectively. Boron concentrations were also measuredin the normal brain and in the zone of brain adjacent to the tumor exhibiting edema.Although treatment with DEX had no appreciable effect on boron uptake in the normalbrain of the rat, after the administration of BPA, it did impact on the boron levels in thezone of peritumoral edema. After the high dose DEX administration protocol, boronlevels in the zone of edema were reduced by ~14% (p

33Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - PMParallel Session 4 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sSession Chairs: Kenneth Burn, Chuan-Jong Tung3:00 PM -An epithermal facility for treating brain gliomas at theTAPIRO reactor. K.W. Burn a, *, L. Casalini a , S. Martini b , M. Mazzini b , E.Nava a , C. Petrovich a , G. Rosi c , M. Sarotto a , R. Tinti aa ENEA (FIS-NUC), Via Martiri di Monte Sole 4, 40129 Bologna, Italy, b Dipartimentodi Ingegneria Meccanica, Nucleare e della Produzione (DIMNP), Universita` di Pisa,Via Diotisalvi 2, 56126 Pisa, Italy, c ENEA (FIS-ION), Via Anguillarese 301, 00060 S.Maria di Galeria (Rome), ItalyIntroduction: TAPIRO at ENEA (Casaccia) is a 5kW fast reactor with a small, highly enriched,core currently operating an experimental epithermal BNCT facility for phantomirradiations within the biological shield. To treat patients, the neutron beam had to beconducted outside the biological shield, involving a total redesign of the column.Materials and Methods: Monte Carlo was used to model the neutron and photon transportin the core, epithermal column and phantom. We employed an in-house variancereduction optimiser which currently is patched to MCNP4B and MCNPX2.1.5. Thefeasibility of treating patients was firstly established by calculating the free beam parametersthat provided an immediate comparison with beams of other facilities. Subsequentlyan anthropomorphic phantom, “ADAM”, was employed to optimise the doseprofiles and in-phantom treatment-planning figures-of-merit. We adopted a reasonablywide collimator aperture (10¥14 cm2) and placed ADAM in the side-of-cranium irradiationposition. The material compositions, 10B concentrations and dosimetric conversionfactors followed accepted standards for BPA and brain gliomas. The usualmaterials employed for epithermal neutron beams were adopted: AlF3 (1.85 g/cm3)moderator, nickel reflector, lead collimator, lithiated polyethylene absorber.Results: The final epithermal column configuration is characteristic of a fast reactor inthat no thermal neutron or g shield was necessary. Furthermore some innovative solutionswere adopted to maintain the neutron inventory. Results are presented for thefree beam and for dose profiles and treatment-planning figures-of-merit in ADAM. bothfor a single beam and for a bilateral parallel-opposed irradiation.Discussion: For a single beam irradiation, the maximum dose rate to healthy tissue is0.252 Gy Eq /min. With a maximum 12.6 GyEq to healthy tissue this implies a treatmenttime of 50 min. For the two beam irradiation the time is 45 min per beam. A relativecomparison of the various dose contributions indicates that these figures-of-meritare very satisfactory but they are strongly dependent on the hypotheses in the section“Materials and Methods”. At TAPIRO because of the low source strength, filters maynot be employed as they unacceptably degrade the treatment time. The only handleavailable to vary the beam is through the thickness of the AlF3 moderator. Increasingthe AlF3 thickness lowers the dose rate, increases the treatment time but raisesslightly the therapeutic ratio and vice versa.Conclusions: TAPIRO can provide a high quality epithermal beam of sufficient intensityto treat patients with brain gliomas. The epithermal column is currently under construction.All the results presented here are from calculations. Extensive measurements willbe required to characterise the facility.3:20 PM -Renovation of epithermal neutron beam for BNCT atTHOR. Y.-W.H. Liu a, *, T.T. Huang a , S.H. Jiang a , H.M. Liu ba Department of Engineering and System Science, National Tsing Hua University,Hsinchu, Taiwan, ROC, b Nuclear Science and Technology Development Center,National Tsing Hua University, Hsinchu, Taiwan, ROCHeading for possible use for clinical trial, THOR (Tsing Hua Open-pool Reactor) at Taiwanwas shutdown for renovation of a new epithermal neutron beam in January 2003.In November 2003, concrete cutting was finished for closer distance from core andlarger treatment room. The resulting distance from core boundary to irradiation pointis ~180cm (original: ~280cm). The beam cross section is 65cm x 60cm (original: 35cmx 33cm). The irradiation/treatment room size is 380cm wide, 240cm high, and 390cmdeep (original : 200cm wide, ~180 cm high). This article presents the design base thatthe construction of the new beam is based on.The filter/moderator design along the beam is : Cd(0.1cm) + Al (10cm) +FLUENTAL(16cm) + Al (10cm) + FLUENTAL(24cm) + Void(18cm) + Cd(0.1cm) +Bi(10cm) with 6cm Pb as reflector. Following the filter/moderator is a 88 cm long 6cmthick Bi-lined collimator with Li2CO3 +PE at the end. The collimator is surrounded byLi2CO3+PE and Pb. The reactor power can be brought up to 2MW. The calculatedepthermal neutron flux under 1MW at the beam exit is 1.7 E9 n/cm2/s. The thermalneutron flux is 0.2 E9 n/cm2/s. The fast neutron dose is 2.8 E-11 cGy cm2 /n, gammaray dose is 1.3 E-11 cGy cm2 /n. The beam is very forward with current to flux ratio =0.8. The epithermal neutron flux is more than 3 times higher than the old beam andthe contamination of fast neutron and gamma ray are much less. The slightly higherthermal neutron flux is due to the use of natural Li near the beam exit. If Li-6 can beobtained and used instead, it can be reduced by a factor of 3. Due to the long collimator,the beam is very forward. At 10cm away from the beam exit, the epithermal neutronflux is still closed to 1E9 n/cm2/s. The thermal-to-epithermal neutron flux ratio at thispoint is close to the value (0.05) recommended by IAEA report.A phantom dose calculation was also performed to give more insight of the beam characteristic.This is done by placing a 18cm x 18 cm x 20cm brain-equivalent phantomat 10 cm from the beam exit. The simulation starts from the core through the filter/collimatorto the phantom to avoid appoximation that would have to be made to form asurface source for the phantom-only calculation. Assuming the boron concentration intumor is 65ppm, in normal brain is 18ppm, MCNP calculation shows that the advantagedepth is 8.9cm, advantage ratio is 5.6. The maximum dose rate for normal tissueis 50 cGy/min. It requires 25 minutes for normal tissue to reach 12.5Gy. The maximumtherapeutic ratio is 6.Both the in-air and in-phantom parameters of the present new beam design at THORare comparable with those of the current BNCT facilities and are reasonably good forBNCT purpose. The construction of the beam is scheduled to be finished by the end ofApril 2004. Measurements of beam characteristics will soon follow.3:55 PM -Flux and instrumentation upgrade for the epithermalneutron beam facility at Washington State University.D.W. Nigg a, *, J.R. Venhuizen a , C.A. Wemple a , G.E. Tripard b , S. Sharp b , K.Fox ba Idaho National Engineering and Environmental Laboratory, 2525 North FremontStreet, PO Box 1625, MS 3860, Idaho Falls, ID 83415-3855, USA, b WashingtonState University, Nuclear Radiation Center, Pullman, WA 99164, USAThe Idaho National Engineering and Environmental Laboratory (INEEL) and WashingtonState University (WSU) have recently constructed a new epithermal-neutronbeam for collaborative Boron Neutron Capture Therapy (BNCT) preclinical researchat the WSU TRIGATM research reactor facility. Some initial measurements for the unoptimizedbeam were performed to verify the basic design. More recently, the reactorcore fuel loading pattern has been reconfigured to maximize the intensity of the beam.This article summarizes the results of some neutronic performance measurements forthe upgraded system, based on activation techniques as well as on results from newon-line beam instrumentation.The new epithermal-neutron beam extraction components are located in the thermalcolumnregion of the WSU reactor-shielding monolith. The original graphite has beenremoved from this region and replaced with a new epithermal-neutron filtering, moderating,and collimating assembly, as shown. Downstream of the filtering and moderatingregion is a bismuth and lead gamma shield, followed by a conical neutron collimatorcomposed of bismuth surrounded by borated polyethylene. The WSU facility incorporatesa high-efficiency neutron moderating and filtering material (FLUENTALTM)developed by the Technical Research Centre of Finland (VTT) into the design.The original WSU reactor core had a flux distribution that peaked on the oppositeside of the core from the beamline entrance. In late 2003, the fuel loading pattern wasextensively reconfigured to force a flux peak as near as possible to the entrance ofthe filter assembly.Foil activation measurements based on a simplified version of the protocol that wasused for earlier INEEL measurements at the FiR 1 epithermal-neutron facility operatedby VTT were made to characterize the free-beam neutron spectrum in the transverseplane at the exit of the conical bismuth collimator. Additional confirmatory measurementsof the total flux level relative to the original unoptimized beam were performedusing cadmium-covered GE-Reuter-Stokes fission chambers placed within a bismuthshield located on the bottom inside surface of the collimator.The measured spectrum was unfolded from a set of gold, indium, tungsten, manganese,copper, and scandium activation foil measurements using a direct least-squaresfitting procedure. Integrating the measured spectrum over the epithermal energy rangeproduces a total epithermal neutron flux of 1.66 x 109 n/cm2-sec at 1 MW with a propagateduncertainty of approximately 5% (1_). This is almost exactly three times the fluxproduced with the original reactor fuel loading, as expected. The factor of 3 improvementwas confirmed by the on-line fission chamber measurements.The free-field neutron KERMA rate of the WSU beam was estimated from the measurementsto be 2.53x10-11 cGy total neutron KERMA in standard tissue per unituseful epithermal-neutron flux, with an estimated uncertainty of about 10%. This is inexcellent agreement with the anticipated design value of 2.75x10-11 cGy-cm2.

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - PMParallel Session 4 - N u c l ea r E n g i n e e r i n g a n d P hy s i c s4:15 PM -Characteristics of the new THOR epithermal neutronbeam for BNCT. C.J. Tung a, *, Y.L. Wang a , F.Y. Hsu b , S.L. Chang a ,Y-W.H. Liu ca Department of Nuclear Science, National Tsing Hua University, Hsinchu 300, Taiwan,b Department of Radiological Technology, Yuanpei University of Science andTechnology, Hsinchu, Taiwan, c Department of Engineering and System Science,National Tsing Hua University, Hsinchu, TaiwanA characterization of the new Tsing Hua Open-pool Reactor (THOR) epithermal neutronbeam and an evaluation of its boron neutron capture therapy (BNCT) performancewere studied using Monte Carlo calculations and the treatment planning system (TPS).The beam facility is currently under construction and expected in completion in March2004. It will be used for a full spectrum study in BNCT leading to possible clinical trials.This report discusses the characteristics of the new THOR beam in air, a referencehead phantom, and a patient treated in the Harvard-MIT clinical trial.The epithermal neutron flux for 1 MW THOR power calculated using TORT and MCNP4C codes is 1.7 x 109 n cm-2 s-1 in air at the beam exit, accompanied by photon andfast neutron absorbed dose rates of 0.21 and 0.47 mGy s-1, respectively. The depthprofiles for the individual dose components in a modified Synder head phantom werecalculated using the MCNP 4C code for clinical boron concentrations. These profileswere then used to determine the figures of merit, or the advantage parameters. With10B concentrations in normal tissue and tumor of 11.4 and 40 ppm, the calculatedadvantage depth dose rate is 0.53 RBE-Gy min-1 at the advantage depth of 85 mm,giving an advantage ratio of 4.8. Applying the same irradiation geometry and prescriptiondose as those adopted in the Harvard-MIT clinical trial, i.e. three fields with a totalof 12.9 RBE-Gy maximum biologically weighted normal tissue dose, the dose patternsfor a patient were derived from the NCTPlan TPS using the new THOR beam andcompared with results of the MITR-II M67 beam. The THOR beam yielded a tumordose ranging from 33.4 to 72.7 RBE-Gy (or 0.91 to 1.98 RBE-Gy min-1 in dose rate)as compared to 16.5 to 38.0 RBE-Gy (or 0.047 to 0.109 RBE-Gy min-1) for the MITR-IIM67 beam. Corresponding brain tissue dose extends from 0.39 to 12.9 RBE-Gy for theTHOR beam as compared to 0.39 to 10.8 RBE-Gy for the MITR-II M67 beam. The totalirradiation time of all three fields for the THOR beam is 36.7 minutes.The in air characteristics of the new THOR beam meet most of the criteria set by theInternational Atomic Energy Agency for desired neutron beam parameters for BNCT.The phantom characteristics of this beam provide advantage parameters showing thateven the deepest seated brain tumors can be treated with the required therapeuticdoses to the tumor and normal tissues. Further, the application of the NCTPlan TPSrevealed that the new THOR beam yielded a much larger dose to the tumor and somewhatlarger dose to the tissue than the MITR-II M67 beam. The present study confirmsthe suitability of the new THOR beam for possible BNCT clinical trials.4:35 PM -Depth-dose evaluation for osteosarcoma and prostatecancer treatments using an epithermal neutron beam.Tetsuo Matsumoto, Hiroshi Nakata and Kazuo AkutsuAtomic Energy Research Lab., Musashi Institute of Technology, Ozenji 971, Asao-ku,Kawasaki-shi, 2150013 JapanFor boron neutron capture therapy (BNCT) to enjoy the same recognition as conventionalradiotherapy, it should be applied to treat not only brain tumors but also variouscancers. An epithermal neutron beam could be now expected to apply for a deep andwidespread tumor. We reported the depth-dose distributions for liver, lung and pancreascancer treatments by BNCT at last symposiums1, 2. The present communicationcovers calculation of depth-dose distributions using an epithermal neutron beamfor possible treatment of both osteosarcoma and prostatic cancers. The osteosarcomais generally attacked in the growth, and is a metastasis cancer having strong radiationresidence. The prostatic cancer is attacked in the older, but is recently increasing inJapan.The osteosarcoma (3x3x3 cm3) is modeled around the knee (8x8x15.5 cm3) includingFemur, Patella and Medical meniscus. The prostatic cancer (1x1x1 cm3) is modeledinside the prostate (3x3x3 cm3) around the generative organs. Rectum (6x4x8 cm3),Bladder (8x10x5 cm3) and Testicle (1x1.5x5 cm3) are also included in the model.The calculations were performed by the three-dimensional continuous-energy MonteCarlo code MCNP. The epithermal neutron beam from a TRIGA-II reactor was usedas a primary neutron source for the depth-dose calculation. The 10B concentrationsin the tumor and normal tissue were assumed at 30 ppm and 3 ppm, respectively. Thephysical dose without considering RBE value was calculated for the estimation of thedepth-dose distributions by MCNP.The calculations showed that the average physical dose in the osteosarcoma was 16Gy/h and the dose ratio of tumor to normal tissue was 4.8 when attaining an epithermalneutron flux of 1.5x109 ncm-2s-1. The average physical dose in the prostatic cancerwas 14 Gy/h and the dose ratio of tumor to normal tissue was about 2.2. The doseevaluation suggests that BNCT could be applied for both osteosarcoma and prostaticcancer treatments.1. Tetsuo Matsumoto: 2001. Depth-dose evaluation for liver cancer treatment by BNCTand GdNCT using an epithermal neutron beam. Frontiers in Neutron Capture Therapy,edited by Hawthorne, pp.1351-1356, New York2. Tetsuo Matsumoto and Fuji Fukushima: 2000. Depth-dose evaluation for lung andpancreas cancer treatment by BNCT using an epithermal neutron beam. Program &Abstracts in 9th international Symposium on Neutron Capture Therapy for Cancer,pp.133-134, Osaka, Japan4:55 PM -In-phantom imaging of all dose components in boronneutron capture therapy by means of gel dosimeters.G. Gambarini a,b, *, V. Colli a,b , S. Gay a,b , C. Petrovich c , L. Pirola a , G. Rosi da Dipartimento di Fisica, Universita degli studi di Milano, via Celoria 16, 20133 Milano,Italy, b INFN, National Institute of Nuclear Physics, Bologna, Italy, c FIS-NUC, ENEA,Bologna, Italy, d FIS-ION, ENEA, Casaccia, Rome, ItalyThe experimental method for in-phantom imaging and profiling the absorbed dose inNCT, separating the contributions of the various secondary radiation components, hasbeen improved. The method is based on suitably designed gel dosimeters in form oflayers (thickness 1-3 mm) and the discrimination of dose components is achieved bymeans of pixel-to-pixel manipulations of images obtained with gel-dosimeters havingdifferent isotopic composition. Gel dosimeters are aqueous systems and their compositionscan be conveniently adjusted so that in neutron fields the absorbed dose dueto each secondary radiation is really almost the same as that absorbed by tissue. Inthe studied dosimeter (Fricke-XylenolOrange infused gel), ionising radiation causeschanges of the optical absorption of visible light. From the optical transmittance images,detected using a CCD camera, dose images are obtained. The method for separatingdose contributions is that of introducing, in the phantom of interest, coupledlayers of gel dosimeters both having good tissue-equivalence for neutron secondariesbut with a composition differing for the content of one isotope whose reactions withneutrons give production of charged particles.In gel with standard composition, the absorbed dose in neutron fields is the sum ofthe gamma dose due to background and to the reactions 1H(n,gamma)2H and thefast neutron dose, mainly due to recoil protons. In a gel with identical composition butadded with 10B, the absorbed dose is the sum of the mentioned contributions plusthe therapy dose. If nitrogen is added, in the same way, the dose due to the protonsemitted in the reaction 14N(n,p)14C contributes to the total. By means of suitablemanipulations of images, the various dose contributions can be obtained. Gammaand recoil-proton contributions can be separated too, by elaborating the absorbeddoses in standard gel and in gel made with heavy water. Proper software has beendeveloped for the pixel-to-pixel manipulation of images. The software has been implemented,in order to operate all the algorithms that are necessary for obtaining, fromthe GL images of couples of gel dosimeters having different compositions, the imagesof the various dose components. Manipulation algorithms include parameters like thecalibration coefficients of the specific gel dosimeters and the dependence of their sensitivityon the radiation LET.Large dose images are obtainable with this method, because the layer geometry ofdosimeters avoids sensible variation of neutron transport due to the isotopic compositionof gel. It is important to observe that the peculiarity of the method is not in the kindof used gel dosimeter or of imaging technique (that is optical or magnetic resonanceanalysis). Fricke-XylenolOrange-infused gels have been chosen because it is particularlysimple to prepare gel matrixes with different isotopic composition and moreoverthe used chemical compounds and the resulting gel are not toxic for the operator. However,whichever kind of gel dosimeter and analysis would be adequate, provided thatthe dosimeters do not change neutron transport, and this is true for layer geometry.5:15 PM -Experimental and computational validation of BDTPSusing a heterogeneous boron phantom. G.G. Daquino a, *, N.Cerullo b , M. Mazzini b , R.L. Moss c , L. Muzi ba CERN, PH/SFT, J00200, CH-1211, Geneva 23, Switzerland, b DIMNP, Universityof Pisa, Via Diotisalvi 2, 56126 Pisa, Italy, c JRC, European Commission, PO Box 2,Westerduinweg 3, Petten, The NetherlandsIntroduction: The idea to couple the Treatment Planning System (TPS) to the informationon the real boron distribution in the patient acquired by Positron Emission Tomography(PET) is the main added value of the new methodology set-up at DIMNP of Universityof Pisa, in collaboration with the JRC of Petten (NL). Building on the CARONTEexperience (Cerullo et al., 1998), this methodology has been implemented in a new34

Eleventh World Congress on Neutron Capture TherapyTuesday, October 12, 2004 - PMParallel Session 4 - N u c l ea r E n g i n e e r i n g a n d P hy s i c sTPS named BDTPS (Boron Distribution Treatment Planning System).Materials and methods: BDTPS is an integrated system, which makes use of theMCNP code (Briesmeister, 2000). Besides the “standard” features included in all treatmentplanning systems, BDTPS implements a software architecture based on threestrictly dependent models, named the 3D, the Monte Carlo (MC) and the Boron (B)models. The 3D model is constructed through a stack of CT images of the patient’sorgan and serves as reference for the automatic reconstruction of the MC model. TheB model is built on the basis of a co-registered stack of PET images.The test and validation of BDTPS required the use of an ad hoc phantom namedHEBOM (HEterogeneous BOron phantoM). HEBOM’s structure, comprising severalPMMA slabs, was conceived to host both the instrumentation for neutron fluence anddose measurement and 64 vials, which were filled with solutions containing boron indifferent concentrations to mimic a highly heterogeneous distribution of this element.P/N diodes were used for measuring the thermal neutron fluence, while paired Ar/Mgand TE/TE ionization chambers measured the fast neutron fluence and the gammadose. The experimental validation was integrated with computational tests made usingboth SERA and MCNP-4C3.Discussion: Some discrepancies appeared in the comparison between the BDTPS andSERA results, but reasonable agreement was found between the results of BDTPSand the analytical model in all HEBOM layers and for all the measured parameters (borondose, thermal neutron flux, proton recoil dose, total gamma dose). This highlightedin particular the difficulty for a “standard” TPS to follow the boron dose distribution witha heterogeneous boron distribution in the tissues.Conclusions: This work demonstrated the correct operation of BDTPS by means ofmeasurements and reference calculations. BDTPS version 1.0 is currently the onlyintegrated TPS package able to evaluate the boron dose on the basis of the realmacroscopic boron distribution acquired through PET scanning. Hopefully, this newtool will be applied and further validated in the Italian BNCT research project (Cerulloet al., 2002).35

36Session Chairs: Dave Nigg, Yoshio Imahori9:15 AM -Preliminary Treatment Planning and Dosimetry for aClinical Trial of Neutron Capture Therapy using a FissionConverter Epithermal Neutron Beam. W.S. Kiger III a, *,X.Q. Lu a , O.K. Harling b , K.J. Riley c , P.J. Binns c , J. Kaplan a , H. Patel d , R.G.Zamenhof d , Y. Shibata d , I.D. Kaplan a , P.M. Busse a , M.R. Palmer da Department of Radiation Oncology, Beth Israel Deaconess Medical Center, HarvardMedical School, 330 Brookline Avenue, Boston, MA 02215, USA, b NuclearEngineering Department, Massachusetts Institute of Technology, 128 Albany Street,Cambridge, MA 02139, USA, c Nuclear Reactor Laboratory, Massachusetts Instituteof Technology, 128 Albany Street, Cambridge, MA 02139, USA, d Departmentof Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330Brookline Avenue, Boston, MA 02215, USAA Phase I/II clinical trial of Neutron Capture Therapy (NCT) was conducted at Harvard-MITusing a fission converter epithermal neutron beam. This epithermal neutronbeam has nearly ideal performance characteristics (high intensity and purity) and iswell-suited for clinical use. Six glioblastoma multiforme (GBM) patients were treatedwith NCT by infusion of the tumor-selective amino acid boronophenylalanine-fructose(BPA-F) at a dose of 7.0 g/m2 body surface area over 90 min followed by irradiationwith epithermal neutrons. Treatments were planned using NCTPlan and an acceleratedversion of the Monte Carlo radiation transport code MCNP 4B. Treatments weredelivered in two fractions with 2 or 3 fields. Field order was reversed between fractionsto equalize the average blood boron concentration between fields. The initial dosein the dose escalation study was 7.0 RBE Gy, prescribed as the mean dose to thewhole brain volume. This prescription dose was increased by 10% to 7.7 RBE Gy inthe second cohort of patients. A pharmacokinetic model was used to predict the bloodboron concentration for determination of the required beam monitor units with goodaccuracy; differences between prescribed and delivered doses were 1.5% or less.Estimates of average tumor doses ranged from 33.7 to 83.4 RBE Gy (median 57.8RBE Gy), a substantial improvement over our previous trial where the median value ofthe average tumor dose was 25.8 RBE Gy.9:55 AM -MINERVA—a multi-modal radiation treatment planningsystem. C.A. Wemple a, *, D.E. Wessol a , D.W. Nigg a , J.J. Cogliati b , M.L.Milvich b , C. Frederickson b , M. Perkins b , G.J. Harkin ba Idaho National Engineering and Environmental Laboratory, P.O. Box 1625, IdahoFalls, ID 83415-3885, b Department of Computer Science, Montana State University,Bozeman, MT 59717 USAResearchers at the Idaho National Engineering and Environmental Laboratory andMontana State University have undertaken development of MINERVA, a patient-centric,multi-modal, radiation treatment planning system. This system can be used forplanning and analyzing several radiotherapy modalities, either singly or combined,using common modality independent image and geometry construction and dose reportingand guiding. It employs an integrated, lightweight plugin architecture to accommodatemulti-modal treatment planning using standard interface components. The MI-NERVA design also facilitates the future integration of improved planning technologies.The code is being developed with the Java Virtual Machine for interoperability.The MINERVA design includes the three common software modules - image processing,model definition, and data analysis- used by any radiation treatment planningsystem, along with a central patient module to coordinate communication and datatransfer. The source and transport plugin modules communicate with these moduleseither directly through the patient database or through MINERVA’s openly published,extensible markup language (XML)-based application programmer’s interface (API).All internal data is managed by a database management system and can be exportedto other applications or new installations through the API data formats.The lightweight common radiation treatment planning system modules (Patient, Image,Model, and Analyze) are functionally complete. A full computation path has beenestablished for molecular targeted radiotherapy treatment planning, with the associatedtransport plugin developed by researchers at the Lawrence Livermore NationalLaboratory. Development of the neutron transport plugin module is proceeding rapidly,with completion expected later this year. Future development efforts will include developmentof deformable registration methods, improved segmentation methods forpatient model definition, and three-dimensional visualization of the patient images,geometry, and dose data. Transport and source plugins will be created for additionaltreatment modalities, including brachytherapy, external beam proton radiotherapy, andthe EGSnrc/BEAMnrc codes for external beam photon and electron radiotherapy.This works was sponsored by the United States Department of Energy, Office of Science,under DOE Idaho Operations Office Contract DE-AC07-99ID13727.Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - AMTreatment Planning10:35 AM -Intercomparison of Neutron Capture Therapy TreatmentPlanning Systems. J.R. Albritton a , W.S. Kiger, III ba Nuclear Engineering Department, Massachusetts Institute of Technology, 77 MassachusettsAvenue, Cambridge, MA 02139, USA, b Department of Radiation Oncology,Beth Israel Deaconess Medical Center, Harvard Medical School, 330 BrooklineAvenue, Boston, MA 02215, USAClinical trials of Neutron Capture Therapy (NCT) are extremely expensive and timeconsuming. Thus, it is very desirable to pool patient data from different sites for analysisof the clinical results. However, both the complexity and the lack of standardizationof both the physical and computational dosimetry of NCT significantly impedethe direct comparison of patient data from different sites. Currently, 5 different treatmentplanning systems (TPSs) are or have been used clinically for NCT: MacNCT-Plan, NCTPlan, BNCT_rtpe, SERA, and JDCS. This paper outlines studies currentlyunderway to comprehensively test and compare 4 of these NCT treatment planningsystems in order to facilitate the pooling of patient data from the different clinical sitesfor analysis of the clinical results as well as to provide an important quality assurancetool for existing and future TPSs.A pre-existing suite of computational test problems will be extended from one-dimensionaldepth-dose profiles to include the multidimensional data used in patient treatmentplanning, i.e., isodose contours and dose volume histograms. The referencedata from this extended test suite will be the common basis for comparison of the differentNCT treatment planning systems. Two different phantoms will be used to evaluatethe planning systems: the modified Snyder head phantom and a large water-filledbox, similar to that used in the International Dosimetry Exchange for NCT. The largewater box is a test problem that is geometrically uninteresting; the phantom itself hasno heterogeneity and is geometrically very simple. This problem is designed to primarilyevaluate the transport physics, etc., of the planning codes and avoid significantlytesting the fidelity of the planning codes geometric representation of the phantom. Themodified Snyder head phantom problem, on the other hand, is designed to test allimportant aspects of the planning codes. An analytical representation of the modifiedSnyder head phantom will be used for reference calculations in the well-validated,state of the art Monte Carlo radiation transport code MCNP5. Volume image datawill be used to define and construct the phantom geometries in each of the treatmentplanning systems. Various monodirectional and monoenergetic neutron and photonbeams will be modeled along with more realistic beams, such as a broad spectrumepithermal neutron beam. Using kerma factors native to each TPS, boron-10 doserate, thermal and fast neutron dose rate, induced and incident photon dose rate, andbiologically weighted dose rates for normal tissue and tumor will be calculated. Depthdose(and thermal neutron flux) profiles will be calculated on the central (beam) axis ofthe phantom for all neutron and photon beams and compared to the reference profiles,as will isodose contours and dose volumes histograms. The test suite will further beextended to include clinical test cases using patient image data relevant to currentclinical trials.Preliminary results comparing depth-dose profiles along the central axes of the twophantoms indicate reasonable, but certainly not excellent agreement between the 4planning codes and the reference problems.10:55 AM -Improvement of dose calculation accuracy for BNCT dosimetryby the multi-voxel method in JCDS. H. Kumada a,b, *,K. Yamamoto a , T. Yamamoto b , K. Nakai b , Y. Nakagawa c , T. Kageji d , A. Matsumuraba Department of Research Reactor, Tokai Research Establishment, Japan AtomicEnergy Research Institute, Tokai, Ibaraki, Japan, b Department of Neurosurgery, Instituteof Clinical Medicine, University of Tsukuba, Tsukuba City, Ibaraki, Japan, cDepartment of Neurosurgery, National Kagawa Children’s Hospital, Kagawa, Japan,d Department of Neurosurgery, School of Medicine, University of Tokushima, Tokushima,JapanIntroduction: To carry out the boron neutron capture therapy(BNCT) clinical trialsbased on accurate dosimetry of several absorbed doses given to a patient, the JAERIComputational Dosimetry System(JCDS) which can determine the absorbed dosesby numerical simulation with Mont Carlo calculation, have been developed by JapanAtomic Energy Research Institute(JAERI). By applying JCDS, the first BNCT clinicaltrial with epithermal neutron beam has been currently carried out at JRR-4 in JAERIin October, 2003.To estimate the absorbed doses and its three-dimensional distributions, JCDS hademployed a voxel calculation method as dividing uniformly the patient’s head model by10x10x10 mm3 mesh. The calculation accuracy of JCDS had been verified based oncomparison with experimental data with water phantom irradiations. The verificationresults proved that JCDS has enough performance for BNCT dosimetry, except forthe estimations near the phantom’s surface. However, the verification also indicated

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - AMTreatment Planningthat JCDS causes discrepancy between the JCDS data and the phantom study nearthe surface, when JCDS estimates the sharp change arising in a distance shorter thana unit 1cm3 voxel cell. The aim of this study is to improve the accuracy of the BNCTdosimetry with efficient calculation work. In the dose calculation method, we havedeveloped the multi-voxel calculation method reconstructing the original voxel modelby combining of several voxel cell sizes such as in 5x5x5 mm3 (0.125cm3 voxel cell),1cm3 and 20x20x20mm3. The region around the surface requiring improvement ofcalculation accuracy especially can be set by a 0.125cm3 voxel cell. The multi-voxelcalculation method has been installed to JCDS.Material and Methods: In order to verify the accuracy of the multi-voxel calculationmethod, the calculation results were compared with the results of conventional voxelmethod and the experimental data obtained from the phantom experiments. Furthermore,to verify the practicality of the multi-voxel method, retrospective evaluation of anactual BNCT which had been carried out in JRR-4 was performed using the multi-voxelmethod.Results and Discussions: For the comparison with the phantom experiments usingepithermal neutron beam, the calculated distribution of the thermal neutron flux wasimproved to correspond throughout the radiation field with the error of less than 5%.The computing time for the multi-voxel method increased only approximately in 30%for the uniformed 1cm3 voxel calculations. The results of the retrospective evaluationfor actual BNCT using the multi-voxel method confirmed that its calculation methodcould be utilized to actual BNCT dosimetry. These results proved that the multi-voxelcalculation enables JCDS to more accurately estimate the absorbed doses to a patientby efficient calculations. Currently, JCDS is being improved more to combine the“Mesh Tally” function of MCNP5 as a useful function for efficiently computing the spatialdistributions. Furthermore a dedicated PC cluster system for BNCT dosimetry asan environment of parallel computing has been constructed. We expect that the newversion of JCDS and the PC cluster system enable to carry out a more precise and optimumdose planning with a short calculation time, which is a benefit for clinical use.11:15 AM -Comparison of the performance of two NCTtreatmentplanning systems using the therapeutic beam of the RA-6 reactor. M.R. Casal a,b, *, S.J. González b , H.R. Blaumann c , J. Longhino c ,O.A. Calzetta Larrieu c , C.A. Wemple da Instituto de Oncología Ángel H. Roffo, Facultad de Medicina, Universidad de BuenosAires, Av. San Martín 5481, Buenos Aires, Argentina, b Comisión Nacional de EnergíaAtómica, Centro Atómico Constituyentes, Av. del Libertador 8250 (1429), Buenos Aires,Argentina, c Comisión Nacional de Energía Atómica, Centro Atómico Bariloche,Av. Bustillo Km.9.5, Bariloche, Río Negro, Argentina, d Idaho National Engineeringand Environmental Laboratory,Bechtel BWXT Idaho, LLC, Idaho Falls, Idaho, USAPurpose: We inspected the performance of two NCT treatment planning systems,NCTPlan and SERA, in some simple geometries for the hyperthermal neutron beamof the RA-6 reactor. Measurements and calculations were performed in a referencerectangular phantom at the central beam axis and in a peripherical axis, and in a cylindricalphantom at the central beam axis. The two codes give accurate results in bothaxis, showing good agreement with experimental results.Introduction: The CNEA-BNCT group has recently started clinical applications withthe neutron beam at the RA-6 reactor;therefore there is great interest to evaluate theperformance of both of the available treatment planning systems (TPSs), SERA v.1C0and NCTPlan v.1.3, with the broad spectrum and multidirectional beam that is usedfor the treatments. Also, it is of great importance to fully understand the differencebetween the calculation methods and the data used to perform them, to evaluate theorigin of any discrepancy that may appear.mean values) at depths greater than 3 cm. The gamma dose rate calculated withSERA differs from the experimental results, but again the difference (5.5%) is acceptablefor clinical use. The fast neutron dose rate calculated with both TPSs agrees withmeasurements within experimental uncertainties.Conclusions: Calculated components with both TPSs show good agreement with experimentalresults, either at the central beam axis or at a peripherical axis in the referencephantom. As boron and nitrogen doses are calculated using the thermal neutronfluence, we can also asume a good agreement in those components. The agreementbetween the calculations and measurements in the cylindrical phantom is also good.The minor differences that have been founded can be explained in terms of the differentapproaches that each TPS uses to model the geometry and to assign kermafactors to a given incident neutron energy.11:35 AM -Simulation Studies of BNCT for Lung Cancer. YoshioImahori a , Katsuyoshi Mineura a , Ryou Fujii b , Tatsuo Ido c , Koji Ono da Department of Neurosurgery, Kyoto Prefectural University of Medicine, b Division ofPET development projects, Ichigaya TRS, Inc.(Tokyo)., c Cyclotron and RadioisotopeCenter, Tohoku University, d Department of Radiation Oncology, Kyoto University,Research Reactor InstituteWe simulated an expanded indication of BNCT for lung cancer based on the fluoroboronophenylalanine(18FBPA)-positron emission tomography (PET) results. The environmentof lung is conducible to an indication for BNCT because the air fraction is verylarge; consequently neutrons can reach to a deep focus. The amino acid derivativeboronophenylalanine (BPA) has been used in BNCT, which plays a role as a boron-10carrier by approaching vicinity of the cell nucleus. It has been shown that pharmacokineticsof positron marker 18FBPA was similar to those of BPA, and intracellular accumulationof BPA depends on primarily amino acid transporter based on the kinetic PETanalysis. The increase in transporter accompanying angiogenesis of a tumor worsensthe prognosis of the patient, and the accumulation of 18FBPA can be visualized in atumor by whole-body PET, creating an image of the degree of malignancy. Therefore,in the present study, we have simulated the expansion of the indication of BNCT forlung cancer based on clinical 18FBPA-PET data.The clinical usefulness of 18FBPA was studied by whole-body PET (Siemens ECATACCEL). The assessment criteria were as follows: (1) uptake value into normal majororgans, (2) degree of retention in the liver, and (3) whether or not emission count of thesame degree as that during 18FDG testing was obtained. After studying in the normalsubjects, the tracer distribution was investigated from the images of lung cancer cases.The cases include 3 with primary lung cancer and 3 with metastatic lung cancer.A comparison with the 18FDG normal images revealed that the whole body basal imagewas almost identical to that for 18FDG, and there was a little 18FBPA uptake intonormal organs. The finding that was significantly different to FDG was that there wasless uptake of 18FBPA into the normal brain. The 18FBPA-PET images were at a similarlevel to the normal 18FDG setting of SUV 4.0. We studied a potential indication ofBNCT for lung cancer. Lung cancers were well visualized by whole-body 18FBPA-PET.The air fraction ratio in the lung is 0.48 based on a calculation from normal lung histology.Consequently, the range of fluctuation in the air fraction ratio due to inspiration is0.48 to 0.67. The mean value is 0.58 and the tissue fraction is calculated as 0.42. Theapparent T/N ratio was 7.6 from patients with metastatic lung cancer in case 1. If thisis corrected by the tissue fraction value of 0.42, the corrected T/N ratio is 3.2. Theseresults suggest a feasible application to BNCT for lung cancer.Materials and Methods: For all the calculations and measurements, we used the hyperthermalneutron beam of the RA-6 reactor. Measurements were performed in arectangular (reference) and a cylindrical phantoms, both filled with water. Measurementsand calculations were performed at the beam central axis on both phantoms,and on an peripherical axis, 4cm apart from the main axis beam, in the referencephantom.Results and Discusion: The thermal neutron flux and fast dose rate calculated withboth TPSs on the reference phantom agree with experimental results within experimentaluncertainties. The gamma dose rate calculated with SERA differs from themean values of experimental results, but the difference is on average , 5%.The calculations and measurements on the peripherical axis show that both TPSscalculate the different dose components outside the central beam axis with the necessaryaccuracy for clinical use.In the cylindrical phantom, SERA calculations for the thermal neutron flux agree withthe experimental results, while NCTPlan overestimates it by 11% (compared with the37

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - B i o l o g y1 - Effect of BNCT on proliferation in normal and precanceroustissues. Elisa M. Heber a , Verónica A. Trivillin a , María E. Itoiz a,b ,David Nigg c , Erica L. Kreimann d , Amanda E. Schwint a, *a Department of Radiobiology, National Atomic Energy Commission, Avenida GeneralPaz 1499, (1650) San Martín, Prov. Buenos Aires, Argentina, b Department ofOral Pathology, Faculty of Dentistry, University of Buenos Aires, Argentina, c IdahoNational Engineering and Environmental Laboratory, Idaho Falls, USA, d Currently atGenitourinary Medical Oncology MD Anderson Cancer Center, USAIntroduction: Within the context of exploring new applications of Boron Neutron CaptureTherapy (BNCT) and contributing to the study of BNCT biology and radiobiologywe previously proposed and validated the use of the hamster cheek pouch oral cancermodel for BNCT studies. The model allows for the study of precancerous tissue aroundthe tumor, an issue of clinical relevance given the phenomenon of field cancerization.We previously reported the first evidence of the usefulness of BPA-BNCT for the treatmentof oral cancer in an experimental model with no damage to normal oral tissue.The aim of the present study was to assess the effect of BPA-BNCT on proliferativeactivity of normal oral tissue and precancerous tissue in the hamster cheek pouch oralcancer model employing incorporation of 5-bromo-2’-deoxyuridine (BrdU), a pyrimidineanalogue of thymidine, as an end-point. Proliferative activity is relevant to normaltissue tolerance and to potential inhibition of tumor growth from precancerous tissue.Materials and Methods: The everted pouches of tumor-bearing and normal hamsterswere irradiated either with beam alone or 3.5 hours after intraperitoneal administrationof 300 mg BPA/kg b.w. at an average fluence of thermal neutrons of 1.1 (S.D.: 0.1)x 10Exp12 neutrons/sq.cm with the thermalized epithermal beam of the RA-6 Reactorat the Bariloche Atomic Center. The total dose was estimated at 14.9 Gy-eq. fortumor, 10.7 Gy-eq. for precancerous tissue and 7.5 Gy-eq. for normal pouch. Detailsof the irradiations and macroscopic and histological follow-up have been previouslyreported. Thirty minutes prior to sacrifice, we administered 2 ml of a 1% solution ofBrdU in distilled water (16g BrdU/kg b.w.) i.p. to each hamster. One to thirty dayspost-irradiation the animals were euthanized and samples of precancerous tissue andnormal pouch tissue were processed for immunohistochemical demonstration of BrdUemploying the peroxidase-antiperoxidase technique. Labeled nuclei were counted inall the epithelial strata above a fixed layer of basal layer, spanning the whole pouchsection for each animal. The following histological categories were discriminated forprecancerous tissue: epithelium with no unusual microscopic features (NUMF), hyperplasiaand dysplasia.Results: BNCT exerted a statistically significant (nested ANOVA and planned t-tests)inhibitory effect on precancerous tissue. At 1 day post-treatment the reduction in proliferationwas 83% for NUMF areas, 78% for areas of hyperplasia and 81% for areas ofdysplasia and at 30 days post-treatment the reduction was 84% for NUMF areas, 87%for areas of hyperplasia and 79% for areas of dysplasia. BNCT inhibited or delayed thedevelopment of new tumors from precancerous tissue in 67% and 33% of the casesrespectively. BNCT elicited a 22% reduction in proliferation in normal tissue 14 and 30days post-treatment that was not associated to observable histological alterations.Discussion and conclusions: BNCT induced a dramatic reduction in the proliferativecapacity of precancerous tissue that would be associated to the inhibition/lag in thedevelopment of new tumors from precancerous tissue reported in the present study.BNCT may thus be of potential use in the control of field cancerized areas aroundtumors.3 - Convection Enhanced Delivery of Boronated Bioconjugatesto Epidermal Growth Factor Receptor PositiveGliomas for NCT. Weilian Yang a , Rolf F. Barth a, *, J. Gong Wu a , WernerTjarks b , Michael J. Ciesielski b , Robert A. Fenstermaker b , Carol J. Wikstrandca Department of Pathology, The Ohio State University, Columbus, OH, 43210, b Departmentof Neurosurgery, Roswell Park Cancer Institute, Buffalo, NY, 14263, c Departmentof Pathology, Duke University, Durham, NC, 27712Convection enhanced delivery (CED) is a potentially powerful method to improve thetargeting of low and high molecular weight agents to the central nervous system byapplying a pressure gradient to establish bulk flow through the brain interstitium duringinfusion. The purpose of the present study was to evaluate CED as a means toimprove the intracerebral (i.c.) and intratumoral (i.t.) uptake of a heavily boronatedmacromolecule (BD) linked to either EGF or an anti-EGFRvIII monoclonal antibody(MoAbs) L8A4 and to increase the therapeutic efficacy following NCT of rats bearinga syngeneic EGFR (+) glioma. The BD was linked to either EGF or L8A4 usingheterobifunctional reagents. BD-EGF and BD-L8A4 were radiolabeled with 125I andadministered by CED at a rate of 0.33 ml/min for 15, 30 and 60 min with correspondingvolumes of infusion [Vi] of 5, 10 and 20 ml, respectively. Animals were euthanized at0, 6, 12, and 24 h after infusion. The uptake and biodistribution of the 125I-BD-bioconjugatesin tumor or normal tissues were studied by means of quantitative autoradiography(QAR) and g-scintillation counting. The volume of distribution (Vd) in brain wasassessed using a computer interfaced image analysis system. Following CED, the Vdincreased from 34.4 to 123.5 ml with corresponding Vi ranging from 5 to 20 ml. TheVd of BD-EGF and BD-L8A4 in the brain was 64.8 ml and 59.8 µl, respectively, withCED (Vi 10 ml) and the Vd:Vi ratio was 6.1-7.0 compared to a Vd of 9.4-11.2 ml anda Vd:Vi ratio of 0.9-1.2 after direct i.c. injection. As determined by QAR and g-scintillationcounting at 24 h following CED, 47.4% of BD-EGF and 60.1% ID/g of BD-L8A4were localized in F98EGFR and F98EGFRvIII gliomas compared to 33.2% of ID/g and43.7% after direct i.t. injection and 12.3-15.2% ID/g in F98WT gliomas. Following NCT,the animals that received CED BD-EGF had a mean survival time (MST) of >50±7days compared to 42±6 days in animals received i.t. BD-EGF (p57±23 days compared to 40±5 days withoutCED BD-EGF (p

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - B i o l o g y39BPA alone or in combination with other boron drugs used in BNCT for hepatoma.Keywords: Hepatoma, Boron neutron capture therapy, Boronophenylalanine7 - A new approach to determine tumor-to-blood 10B concentrationratios from the clinical outcome of a BNCTtreatment. S.J. González a, *, D.G. Carando b , M.R. Bonomi ca UARyCN, Comisión Nacional de Energía Atómica, Centro Atómico Constituyentes,Av. Del Libertador 8250, 1429 Buenos Aires, Argentina, b Dpto. de Matemática, Universidadde San Andrés, Vito Dumas 284, 1644 Victoria, Buenos Aires, Argentina,c Dpto. de Terapia Radiante, Instituto Oncología Ángel H. Roffo, Av. San Martín 5481,1417, Buenos Aires, ArgentinaIntroduction: In Boron Neutron Capture Therapy (BNCT), the absorbed dose in tumortissues strongly depends on the densely ionizing particles that result from the10B(n,alpha)7Li reaction. The determination of boron concentration in tumor duringirradiation is based on the 10B concentration in blood. A good estimate of the tumor-to-blood10B concentration ratio is therefore crucial. Up to now, all methods fordetermining this ratio relied on surgery and biodistribution analysis. We propose a newapproach to determine the tumor-to-blood ratio: a maximum likelihood estimate fromthe clinical outcome of a BNCT treatment. This approach does not involve surgery andcan be performed after any BNCT treatment. Although the present method does notmean to replace experimental measurements, it can play an important role in treatmentplanning.Materials and methods: Our starting point is the clinical response of thirty nine malignantmelanoma nodules treated with BNCT. We find the value of the tumor-to-blood10B concentration ratio r that maximizes the likelihood of the clinical outcome. Toevaluate this likelihood for each possible value of r, we need to estimate de total dosereceived by each nodule for every r. First, we computationally reconstruct the patientanatomy from CT scans. Then, we evaluate the dosimetry data output from NCTPlanon the surface of the reconstructed anatomy for different values of r. Finally, we registerthe dosimetry data with a picture of the patient anatomy to obtain the dose in eachnodule. The tumor control probability of each nodule (an every r) is then computedwith the model introduced by Overgaard et al.. With these probabilities we evaluatethe likelihood function. We then estimate the tumor-to-blood ratio maximizing the likelihoodfunction. Finally, we obtain a confidence interval for the ratio with a parametricbootstrap.Results and discussion: The maximum likelihood estimate for the tumor-to-blood 10Bconcentration ratio is r = 3.05 ± 0.46 with 95% confidence. The statistical deviationof the estimated ratio is 0.24. This estimated ratio is consistent with results found inliterature. We also compare our results with ratios experimentally obtained for theclinical case under review. These experimental results are roughly consistent with ourstatistical estimation, considering the 95%-confidence interval.Conclusion: We present a statistical method to determine the tumor-to-blood 10B concentrationratio in BNCT. We show its performance in a clinical case of cutaneousmultiple nodular melanomas. We obtain a ratio of r = 3.05 ± 0.46, consistent with thosefound in literature. One of the advantages of the proposed method is that it does not involvesurgery. Also, a single patient with multiple nodules is enough to give statisticallyrelevant results, as the present case shows. Another important feature is that afterevery BNCT treatment, data can be collected to adjust the estimated tumor-to-bloodratio. The proposed method can be applied to adjust some other parameters involvedin the model, such as the CBE and RBE factors.9 - Biodistribution of 10 B in a rat liver tumor model followingintra-arterial administration of sodium borocaptate(BSH)/ degradable starch microspheres (DSM) emulsion.Minoru Suzuki a, *, Kenji Nagata a , Shinichiro Masunaga a , YukoKinashi a , Yoshinori Sakurai b , Akira Maruhashi b , Koji Ono aa Radiation Oncology Research Laboratory, Research Reactor Institute, Kyoto University,Noda, Kumatori-cho, Sennan-gun,Osaka 590-0494, Japan, b Division of RadiationLife Science, Research Reactor Institute, Kyoto University, Noda, Kumatori-cho,Sennan-gun,Osaka 590-0494, JapanBackground and Purpose: We reported that intra-arterial administration of borocaptatesodium (BSH)/lipiodol emulsion provided selectively high 10B concentrations (approximately200 ppm at 6 h after administration) in experimental liver tumors. In the presentstudy, we investigated intra-arterial administration of BSH with another embolizingagent, degradable starch microspheres (DSM). The purpose of the present study wasto evaluate the biodistribution of 10B after intra-arterial administration of BSH/DSMemulsion using a rat liver tumor model, and to compare the results with our previouslyreported results for the administration of BSH/lipiodol emulsion or BSH alone.Materials and methods: Female Wistar rats (200-220 g) were used. Rat liver tumorswere developed by direct injection of Walker 256 cells, rat breast cancer cell line, intothe liver parenchyma. BSH (75 mg/kg)/DSM (30 mg/kg) emulsion was administeredvia the hepatic artery. 10B concentrations in the tumors, liver and blood were measuredat 1 and 6 h after administration.Results: The median 10B concentration in the tumor at 1 h after administration ofBSH/DSM emulsion was 231 ppm and 11.6 times higher than that treated with BSHalone. This value was lower than that following administration of BSH/lipiodol emulsionwithout significance (p = 0.070). At 6 h, the median 10B concentration in the tumor in“BSH+DSM” group was 81.5 ppm and significantly lower (p = 0.0018) than that in the“BSH+lipiodol” group. The 10B concentration in the liver at 1 h after administration ofBSH/DSM emulsion was the highest among the three treatment groups. At 6 h, the10B concentration in the liver in the “BSH+DSM” group was significantly higher (p =0.0047) than that in the “BSH+lipiodol” group. The tumor/liver 10B concentration ratios(T/L ratios) were significantly smaller than those in the “BSH+lipiodol” group at 1 h (1.4vs. 3.6, p = 0.0071) and 6 h (1.1 vs. 14.9, p < 0.0001).Conclusions: Boron neutron capture therapy (BNCT) using BSH/DSM emulsion (BSH/DSM-BNCT) is not suitable for treatment of multiple liver tumors due to the low T/L 10Bconcentration ratio. However, the high 10B accumulation in the liver tumor followingintra-arterial administration of BSH/DSM emulsion suggests that BSH/DSM-BNCT hasthe potential for application to malignant tumors in other sites.11 - Experimental pharmacokinetic studies of Gd and 10Bcontainingcompounds at the MEPhI Reactor. V.N.Kulakov a ,I.N.Sheino a , V.F.Khokhlov a , A.A.Portnov b , K.N.Zaitsev b , V.I.Kvasov b ,K.P.Alekseev a , V.V.Stepanov a , T.A.Nasonova a , O.A.Dobrynina aa SRC – Institute of Biophysics, Moscow, Russia, b Moscow Engineering PhysicsInstitute, Moscow, RussiaIn the preclinical NCT studies, the compounds with Gd - Dipentast [1] and [10B] - BPAare used. The experimental pharmacokinetic studies are based on the determinationof the concentration of the administered drug in the organs and tissues of laboratoryanimals. The content of Dipentast in biological tissues was determined using the methodof neutron activation analysis developed by us at the MEPhI reactor. Dipentast andsimilar pharmaceuticals are intercellular drugs [2] and are excreted from the organismmainly unchanged; therefore, the Gd contents correlates with the drug content. Thetracking method was used to determine 10B in biological tissues. The samples wereirradiated in the neutron beam of the MEPhI reactor. A film of cellulose acetobutyrateof 10 to 100 µm thick was used as a detector.The studies involved rats and mice, both intact and inoculated with tumors (sarcomaS-45 in rats, melanoma B-16 in mice). The intratumoral route of administration is mostacceptable for Dipentast. BPA, on the contrary, is capable of penetrating into tumorcells. Both systemic and regional routes of administration are used for BPA. The drugswere administered intravenously and intratumorally. The Gd content was determinedin blood, brain, lungs, liver, kidneys, muscles, and tumor tissue of the animals sacrificedat different terms.For pharmacokinetic modeling, combined differential equations were developed, reflectingthedynamics of the drug’s mass balance in the organism (multicompartmentmodel). The coefficients of the equations were determined by the least-squares methodwith random search in the space of parameters being determined.The Dipentast’s, pharmacokinetic model for its intratumoral administration was basedon Nicholson’s diffusion model [3]. The diffusion coefficient of the drug in tissue wasobtained empirically from the data of radioisotope studies of the drug’s half-life. Theobtained diffusion coefficient D=(0.58±0.14) 10-7 cm2/s appeared to be rather closeto its values for complex polymers in brain tissue [4]. The developed models allowdetermining quantitatively the dynamics of the Dipentast-type drug in the organism ofexperimental animals. The main direction of our further research is development of ageneral mathematical model describing the drug migration in an organism taking intoaccount the intensity of blood-stream, parameters of blood and development of thecapillary network in the pharmacological target for various routes of drug administrationto the organism.The work is carried out with financial support ISTC, Project # 1951.References1. V.N.Kulakov, V.F.Khokhlov, Yu.V.Goltyapin et al., Patent RU 2150961.2. Magnevist, by Ed. R.Felix,.A.Heshiki, N.Hosten, H.Hricak, Oxford, Blackwell Scientific Publication,1994, 192 p.

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - B i o l o g y403. Nicholson C. Diffusion from an injected volume of substance in brain tissue with arbitrary volumefraction and tortuosity. Brain Res. 333:325-329, 19853. Prokopova-Kubinova S., Vargova L., et al., Poly[N-(2-hydroxypropyl)methacrylamide] PolymersDiffuse in Brain Extracellular Space with Same Tortuosity as Small Molecules. Biophys. J., 2001, v.80 (January), p. 542-548.13 - A Biological Comparison of Neutron Beams for BNCT.A. J. Mason a , D. T. Beynon a , J. W. Hopewell b , S. Green c , C. N. Culbertson a ,J. Capala d , P. Munck af Rosenschöld ea School of Physics and Astronomy, University of Birmingham, Edgbaston, Birmingham,UK, b Deptartment Of Clinical Oncology, Churchill Hospital, Oxford, UK,c Department of Medical Physics, University Hospital Birmingham, Edgbaston, Birmingham,B15 2TH, UK, d Studsvik Medical AB, Nyköping, Sweden, e Department ofMedical Radiation Physics, Lund University Hospital, Lund, SwedenIntroduction: The biological properties of a reactor-based and an accelerator-generatedclinical epithermal neutron beam are compared using an in vitro V79 cell culturetechnique.Materials and Methods: Vials of V79 cells were irradiated with epithermal neutrons, atvarying depths in a water-filled polyethylene phantom, using the R2-0 reactor beamat Studsvik, Sweden, and the accelerator beam at the University of Birmingham, UK.Irradiations were at 4 C to minimise any dose-rate effects. Following irradiation cellsurvival analysis was carried out. A comparison was made with cells exposed to 250kVp X-rays.Results: The cell survival curves for cells irradiated with neutrons and X-rays werefitted to a linear quadratic function. The overall RBE of both neutron beams was determinedat three levels of cell survival and by taking the ratio of the _ values in thelinear quadratic model. The RBE of the Studsvik beam varied between 4.22 ± 0.97and 1.81 ± 0.49 and the Birmingham beam from 1.94 ± 0.40 to 1.27 ± 0.81, as a functionof depth in the phantom. Based on the assumption that the DMF of the photoncomponent was 1.0, the RBE of the high LET components (nitrogen capture and fastneutrons) was calculated. In the Studsvik beam this varied from 27.8 ± 5.8 to 8.52 ±3.1, compared with between 10.4 ± 2.93 and 6.33 ± 1.84 for the Birmingham beam.Neutron-RBE increased significantly with depth in the Studsvik beam, but did not varyin the Birmingham beam. The nitrogen dose was estimated in these calculations, asthe nitrogen content of cells in suspension was unknown.Discussion: Dosimetry measurements and MCNP simulations showed that the fastneutron component of the Studsvik beam was considerably higher than that of theBirmingham beam. Furthermore, the maximum neutron energy in the accelerator-generatedbeam at Birmingham was 1.16 MeV, much lower than for the reactor beam. Thedifferent energy spectra may explain the higher RBE values in the Studsvik beam. Themajority of high LET dose in the Birmingham beam was due to nitrogen capture, andthe relative contribution of this component to the total dose does not vary greatly withdepth. Consequently the high LET RBE of the Birmingham beam was fairly constantwith depth in the phantom. It was possible to attempt to resolve the difference in RBEof the nitrogen and fast neutron components in the Birmingham beam. One conclusionas a result of this analysis is that the RBE of the nitrogen component was slightlylower than the RBE of fast neutrons in the accelerator based beam. However, the largeuncertainties in the nitrogen dose make the result uncertain.Conclusions: This study indicates that the lower fast neutron component in the accelerator-basedbeam translates to measurable differences in the overall biologicaleffectiveness when compared with a reactor-based beam. The present results suggestthat use of the Studsvik dose prescription on the Birmingham beam would provide aconservative starting point for a dose escalation study.15 - Pre-irradiation of BPA-BNCT increases the accumulationof BSH into the tumor: Implications for fractionatedBNCT. Kenji Nagata, Minoru Suzuki, Yoshinori Sakurai, Yuko Kinashi,Shin-ichiro Masunaga, Koji OnoResearch Reactor Institute, Kyoto UniversityPurpose: It is a purpose of this study to examine whether pre-irradiation of boronneutron capture therapy (BNCT) with boronophenylalanine (BPA) increases sodiumborocaptate (BSH) concentration in tumors.Methods: SCC VII cells were collected from exponentially growing cultures, and cellswere inoculated subcutaneously into both the hind legs of 8- to 9- week-old femalemice. A dose of 500 mg/ kg body weight d,l-BPA was administrated to mice intraperitoneally.Neutron irradiation (BPA-BNCT) was carried out 90 min following administrationof BPA at Kyoto University Reactor. The thermal neutron fluence was 1.07 ± 0.10 x1012 n/ cm2 or 4.20 ± 0.66 x 1012 n/ cm2. All irradiated mice were anesthetized withNembutal. BSH (75 mg/ kg body weight) was intraperitoneally administered 3 hoursthrough 381 hours after BPA-BNCT. The mice were then killed 30 min after the administrationof BSH. The tumors or the surrounding normal tissues were excised andplaced in 2 ml volume Teflon tubes with tight cap to prevent drying. The B-10 concentrationswere measured by prompt gamma ray spectrometry using a thermal neutron.The investigation of boron bio-distribution by neutron induced autoradiography wasperformed using a CR-39 etched track detector. 90 min after the administration ofBPA, or 30 min after BSH administration, the mice were killed, and the dried-frozentumorss were cut using a cryotome. The slices mounted on the CR-39 plastic platesand were irradiated with thermal neutrons. Using a microscope, macroscopic imagesof the distribution of the a/ Li tracker on the plates were evaluated.Results: The mean boron concentration of tumors was 15.6 ± 0.7 _g/ g at 90 min,after BPA (500 mg/ kg) was administered to mice by intraperitoneal injection. Tumorboron concentrations expressed a maximum at 96 hours after BPA-BNCT for 10 min,and the concentrations was 16.5 ± 0.5 _g/ g. Tumor boron concentrations at 72 hoursafter BPA-BNCT for 10 min and 30 min were 16.1 ± 0.4 _g/ g and 19.3 ± 0.6 _g/ g,respectively. They were dependent on the fluence of thermal neutron. At 168 and 336hours after BPA-BNCT for 10 min, the concentrations were 15.2 ± 0.7 _g/ g and 12.7 ±1.7 _g/ g. At 381 hours after BPA-BNCT for 30 min, the concentration was 18.1 ± 0.4_g/ g. The extension of a period of the boron high concentration in tumors dependedon the fluence of thermal neutron. In an autoradiography after BSH administration, thea/ Li tracks in tumor area were observed as sparse dots on CR-39 plates, and densedots were depicted in the blood vessels and adjacent normal tissues. In contrast, inthe autoradiography after BPA administration, the tracks in tumors were homogeneousdots. After BPA-BNCT, the tracks in necrotic tissues were observed as dense dotsConclusion: Precedence of BPA-BNCT raises the boron concentration in tumors andthis treatment may make us improve treatment effects of BNCT. We, however, have toelucidate detailed BSH distribution in tumors after BPA-BNCT more.17 - Uptake mechanism of p-BPA-ol into tumor cells. YokoNik a , Hirohumi Kondoh a,b,c , Tomoyuki Asano b,d,c , Mitsunori Kirihata b,ca Bio-Research, Inc. Kobe Japan, b Division of Applied Biological Chemistry, GraduateSchool of Agriculture, c Life Sciences, Osaka Perfecture University, Osaka, Japanp-Boronophenylalanine (BPA) is a tumor-seeking amino acid and also one of the mosteffective boron carrier for boron neutron capture therapy (BNCT). On the other hand,p-boronophenylalaninol (BPA-ol) is designed and synthesized by us as a water solubleanalogue of BPA. BPA-ol has secondary amino and primary hydroxyl groups on theadjacent carbon atoms. The killing effect of DL-BPA-ol was reported as about 4.3-foldthat of DL-BPA against C6 gliosarcoma cells on neutron irradiation experiments. However,there has been no report on the uptake mechanism BPA-ol by tumor cells.In this study, we investigated that the uptake mechanism of BPA-ol into tumor cells. It isassumed that BPA would be transported by carrier protein located in amino acid transportsystem, so called L-system. This system transports neutral and aromatic aminoacid such as tyrosine and phenylalanine into the cells. It is guessed that BPA is takenup into cells because of structural similarity to tyrosine. Moreover, it is known thatthis transport system serves the penetration pathway of amino acid-like substances.Therefore, we guessed BPA-ol, the analogue of BPA, would be taken up into cancercells by this transport system, and carried out the following experiments.First, uptake of D-BPA-ol or L-BPA into B16F10 mouse melanoma cells was examinedwhether what it would be influenced in the state of co-incubation with L-tyrosine orD-tyrosine. Second, in order to elucidate the relationships between differentiation ofcancer and uptake of BPA-ol, we examined that the boron uptake by D-BPA-ol or L-BPA-ol or L-BPA into various human cancer cells, HeLa (epithelioma), KB (epidermalcarcinoma), HEp2 (epidermoid carcinoma), A-549 (lung cancer), and two melanomacell lines. One is G-361 which is moderate melanogenic, the other is Ihara cell whichhas high melanogenic ability. After these cells were exposed to the boron compoundsfor 24 hours, the boron concentration in the cells was determined using inductivelycoupled plasma atomic emission spectroscopy (ICP-AES).The boron uptake of D-BPA-ol into B16F10 cells was inhibited by the coexistenceincubation with L-tyrosine below at the half, and also the boron uptake of L-BPA wasinhibited by coexistence of tyrosine to about 70 percent. Therefore, the relationshipbetween a tyrosine transport system and the uptake mechanism of D-BPA-ol and L-BPA-ol was suggested.In the uptake experiments using various cancer cells, for all cell kinds, the uptake ofD-BPA-ol was higher than that of L-BPA-ol. Compared with the uptake of D-BPA-ol andthat of L-BPA, the uptake of D-BPA-ol was higher than L-BPA in the two melanomacells and HEp2 cells. On the other hand, in HeLa and A-549 cells, the uptake of L-BPAwas higher, and almost equivalent in KB cells. Likely to L-BPA, the each uptake ofboth D- and L-BPA-ol into high melanogenic Ihara cells was very high. This result isassumed because the tyrosine demand nature of the Ihara malignant melanoma cellwith high melanin generation ability is high.

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - B i o l o g y19 - Gadolinium Neutron Capture Therapy - ExperimentalGdNCT Effect on Tumor and Vessel. Masao Takagaki a,d, *,Takanobu Tomaru b , Yoshinori Sakurai c , and Narayan S. Hosmane da Dept Neurosurg, Aino University Hospital, Osaka Japan, b Dept Cardiology,SchMed, Toho University, c Life Science, Research Reactor Institute of Kyoto University,d Dept Chem Biochem, Northern Illinois University157Gd is an fascinating atom for NCT, since it has a large thermal neutron cross sectionof 255,000 barn which is 65 times as that of 10B and releases Auger electrons,internal conversion electrons, g rays and X rays by a single thermal neutron capturereaction sharing among them the total kinetic energy of 7.7 MeV that is more than 2times as that of 10B(n, a)7Li reaction. In BNCT, high LET particles of a and its recoil7Li particle release 3.3 MeV only within their total trajectory of less than 14mm. Suchlimited energy transfer within short trajectory in tissue can save serious radiation injuryonto the normal brain surrounding the tumor. On the other hand, however, dosedistribution in the tumor is sharply dependent on the microdistribution of 10B in tumorthat revealed to be heterogeneous carried non-uniform dose distribution in the tumoreven aside from a variety of proliferation condition of the tumor cells. Furthermore,the important point of this therapy is selective destroy of the tumor cells invading peritumoralparenchyma shown as abnormal high density area in the T2-weighted MRIimage since 80-90% recurrence occur in this area even after multimodal treatment ofthe tumor. Time delay of boron uptake in these invading tumor cells and leads to lossof BNCT effect. Dose distribution of Gd-based NCT is more uniform in tumor accompaniedwith peritumoral harrow than that of boron-based NCT and might be suitable forpathological heterogeneity of malignant tumors and even on the tumor cells invadinginto peritumoral lesion. In this study, GdNCT effect on brain tumor bearing cats arepresented. We also evaluated preventive effect of Gd-NCT on intimal proliferation followingvascular injury in animals as an effect of GdNCT on vessel.In recent studies we have analyzed the biodistribution of BOPP (tetrakis-carboranecarboxylate ester of 2,4-bis-(a,b-dihydroxyethyl)-deutero-porphyrin IX) and showedthat when BOPP was injected 5-7 days before BPA, and the animals were sacrificed60 min after the i.p. injection of BPA, a significant increase in boron uptake by the tumorwas found (38-45 ppm with both compounds vs 20 ppm with BPA alone). On day5 the ratios were: tumor/blood 3.75; tumor/ distal skin 2. Other important ratios weretumor/thyroid 6.65 and tumor/lung 3.85 (Dagrosa et al, submitted).The present studies were performed in order to evaluate the outcome of the BNCTprocedure utilizing the combined injection of BOPP and BPA.21 - Efficacy of Porphyrin-Mediated BNCT in Mice. MichikoMiura a , Gerard M. Morris a , Peggy L. Micca a , Marta M. Nawrocky a , Michael S. Makar a ,Kent J. Riley b , Peter J. Binns b , Otto K. Harlin g , Jeffrey A. Coderre ba Medical Department, Brookhaven National Laboratory, b Nuclear Engineering Department,Massachussetts Institute of TechnologyFive groups of BALB/c mice (12 mice/group) bearing EMT-6 mammary carcinomas implantedsubcutaneously on the leg were given 285 mg/kg CuTCPBr, a copper octabromotetracarboranylphenylporphyrin,by serial intraperitoneal injections (4 injections/2days). The porphyrin was emulsified in a 9% Cremophor EL and 18% propylene glycolformulation. Four of those groups were irradiated at various reactor exposures at theMITR thermal neutron facility for physical doses of 70, 95, 105, and 115 Gy. The fifthgroup was euthanized on the day of irradiation for biodistribution data. Two days afterthe last injection a mean (+ SE) of 190 + 26 mg/g B was found in tumor and 36 + 12mg/g B was found in blood giving a tumor-to-blood ratio of 5.4. Long-term survivors(after 195 days) in the following groups were: 3/10 at 70 Gy, 6/12 at 95 Gy, 6/12 at 105Gy, and 7/12 at 115 Gy. In no case was a mouse euthanized due to radiation damageeven at the highest doses. CuTCPBr appears to be as efficacious as its analog,CuTCPH, in controlling EMT-6 tumors by BNCT.23 - Boron neutron capture therapy for undifferentiatedthyroid carcinoma:preliminary results with the combineduse of BPA and BOPP. M. Viaggi a, *, M.A. Dagrosa a , J.Longhino b , H. Blaumann b , O. Calzetta b , S.B. Kahl c , G.J. Juvenal a , M.A.Pisarev a,da Department of Radiobiology, Constituyentes Atomic Center, Av del Libertador 8250,1429 Buenos Aires, Argentina, b RA-6, Bariloche Atomic Center, National AtomicEnergy Commission, Av Bustillo s/n, San Carlos de Bariloche, Rio Negro, Argentina,c Department of Pharmaceutical Chemistry, University of California, San Francisco,CA, USA, d Department of Biochemistry, University of Buenos Aires, School of Medicine,Buenos Aires, ArgentinaUndifferentiated thyroid carcinoma (UTC) is a malignant tumor of rapid growth andgreat invasiveness. It does not respond to treatment with radioiodine, chemo- or radiotherapy,and has a very poor prognosis (mean survival of around 12 months).Human undifferentiated thyroid cancer cells (UTC) have a selective uptake of borophenylalanine(BPA) both in vitro and in vivo. We found a 50% histologic cure of micewhen the tumors were 50 mm3 or less in volume, when treated with BPA and a neutronbeam and a 100 % of tumor growth control in all animals. However, it is desirable tofurther improve the efficacy of this form of therapy.41

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - C h e m i s t r y1 - Synthesis of boron-containing heterocyclic compounds.Yuri Azev a,b , Irina Slepukhina a , Detlef Gabel a, *a Department of Chemistry, University of Bremen, P.O. Box 330440, D-28334 Bremen,Germany, b Ural Scientific Research Institute of Technology of Medicinal Preparations,620219 Ekaterinburg, Russian FederationIntroduction: The transport and accumulation of boron-containing compounds intotumor tissue is a necessary prerequisite of BNCT (boron neutron capture therapy).Here, we describe the preparation of heterocyclic compounds which might be of usein BNCT. A few examples of heterocyclic compounds for BNCT such as nucleic acidcomponent, DNA binders and triazines have been described in the literature. Theseheterocycles contain the o-carborane cluster. Recently, we have described the interactionof the RNH2-B8H11NH-R cluster with N-heterocycles and found that pyridinecan substitute easily the exo-amino group of the azanonaborane. We noticed that thereaction occurs readily with aliphatic amines, and with pyridine, but not with aromaticamines. Pyrrol does not react, but imidazole does. We therefore wanted to explore thepossibilities to react other nitrogen-containing heterocycles with larger steric demandwith the azanonaborane; this seemed appropriate because many of the DNA ligandscontain a benzimidazol moiety. s-Triazines as heterocyclic carriers of boron-containingfragments appear of interest. These compounds may act as antimetabolites of pyrimidinebases and can accumulate in cancer cells. s-Triazines are also being investigatedas anticancer drugs. In the present work we describe the synthesis of carborane-containing1,3,5-triazines, the synthesis of 1,2,4-triazines containing the B12H122—clusterand attempts to attach the RNH2-B8H11NH-R cluster to DNA binders.Results and Discussion: The preparation of carborane derivates was achieved withoutproblems. The resulting compounds have limited water solubility, as was expected.The use of piperazine substituents increases, however, the water solubility at physiologicalpH. Further tests of the compounds concerning their cellular toxicity and theirbiodistribution are called for. The dodecaborate cluster derivatives of the 1,2,4-triazineopen up new possibilities for attaching the cluster to heterocycles. The easewith which the sulfur can substitute the sulfone offers a good possibility to structurallyvaried compounds of this and similar heterocycles. Of special interest is the fact thatthe borate-substituted 1,2,4-triazine becomes a rather strong base, and is isolated inits N-protonated form. The strong electron-donating power of the cluster had beennoticed before. By alkylation of the sulfur with groups other than the one used here,neutral yet hydrophilic compounds can be expected. The attempts to prepare an adductof the furanylbenzimidazol and the azanonaborane cluster gave no results. Thismight be due to steric hindrance of the molecule, where access of the nitrogen to therather bulky cluster might be impeded. We cannot, however, rule out the possibility thatthe lack of reaction was caused by the low solubility of the benzimidazol in benzene;also with the pyridylbenzimidazol, which possesses similar solubility in benzene, butoffers an additional, stericly less hindered nitrogen atom for reaction, the desired reactiondid not occur.3 - Pt-carborane complexes as new BNCT agents. L. M.Rendina a , D. Caiazza b , E. L. Crossley a , W. A. Pearce b , J. A. Todd b , S. L.Woodhouse b , E. J. Ziolkowski,a The University of Sydney, Sydney NSW 2006, Australia, b The University of Adelaide,Adelaide SA 5005, AustraliaDespite the successful use of platinum drugs such as cisplatin (Platinol), carboplatin(Paraplatin) and oxaliplatin (Eloxatin) in combination therapies for numerous solidtumors including ovarian, lung, testicular, bladder, colorectal, gastric, head and neckcancers, there exists a clear need to expand the clinical efficacy of this importantclass of anti-cancer agents. Many human tumors are responsive but not yet curable byplatinum chemotherapy as a result of serious toxic side-effects and, moreover, manysolid tumors that initially respond to platinum-based therapy acquire resistance in thelonger term. The great majority of human tumors are poorly responsive or refractoryto all forms of chemotherapy and, indeed, 90% of all drug cures occur in only 10% ofcancer types [1].It is well established that in Boron Neutron Capture Therapy (BNCT) the localization of10B-containing agents near chromosomal DNA causes maximum cell damage uponirradiation with thermal neutrons [2, 3]. By combining a multinuclear platinum entityand multiple boron atoms within the same molecule, tumor cell destruction may beenhanced by coupling the neutron capture reactions of the 10B nucleus with the avidand persistent DNA-binding properties of the platinum centres. Indeed, an additive orperhaps synergistic therapy of certain tumors is also feasible here, particularly whenone considers the additional NC contribution of the platinum centre.The goal of our work is to expand the clinical efficacy of platinum chemotherapy bycoupling the potent DNA-binding characteristics of certain platinum compounds withthe remarkable neutron capture properties of the 10B nucleus that have been exploitedin BNCT. Preliminary results from our laboratory regarding the synthesis, DNA-binding,and tumor cell uptake of archetypal Pt-B complexes have been most encouraging[4, 5]. The complexes are the first examples of transition metal complexes with boron-containingligands that have been shown to target DNA and possess anti-canceractivity against a variety of tumor cell lines including those resistant to cisplatin. Thekey results of this work will be presented.References[1] Chu, E.; DeVita, V. T. in Cancer: Principles & Practice of Oncology, 6th Ed.; DeVita, V. T.; Hellman,S; Rosenberg, S. A., Eds., Lippincott Williams & Wilkins: Philadelphia, 2001, p. 289.[2] Hawthorne, M. F.; Lee, M. W. J. Neuro-Oncol., 2003, 62, 33.[3] Soloway, A. H.; Tjarks, W.; Barnum, B. A.; Rong, F. -G.; Barth, R. F.; Codogni, I. M.; Wilson, J. G.Chem. Rev., 1998, 98, 1515.[4] Todd, J. A.; Rendina, L. M. Inorg. Chem., 2002, 41, 3331.[5] Woodhouse, S. L.; Rendina, L. M. Chem. Commun., 2001, 2464.5 - Boron Containing Nitroaniline Mustards – Synthesisof a potentially Hypoxia Selective Compound for BNCT.P. Lemmen*Technical University Munich, Organic Chemistry I, Lichtenbergstr. 4, D-85747 Garching,GermanyIntroduction: Angioneogenesis is a crucial factor of tissue growth. It enables its supplywith oxygen and nutrients. Frequently angioneogenesis is imperfectly regulatedin solid tumors, leading to poorly vascularized regions. These experience permanentshortage in oxygen &Mac246; i.e. hypoxia. Whereas in tissue normally supplied withblood the partial pressure of oxygen is about 25-30 Torr, it is frequently reduced intumor tissue (E.g. from 25-40 to 2-12 Torr in a rat model (Cerniglia et al., 1997)).Because of the tendency of various oxygen demanding enzymes to utilize other appropriateelectron sinks instead, this depletion in oxygen in principle opens the opportunityto devise BNCT prodrugs. These should be freely diffusible, but be boundand retained locally after reduction in hypoxic tissue. Nitroaniline mustards have beenshown to be reduced in hypoxic tissue, leading to inceased mustard reactivity towardsnucleophiles (Palmer et al., 1996). The expected increase in reactivity is illustrated bythe successive increase of pKa by 12 units during the reduction of 2,4-dinitroaniline to1,2,4-triaminobenzene. We therefore undertook the synthesis of the boron containingnitroaniline mustard 3-(bis-2- chloroethyl)amino-2,4-dinitro benzoic acid 2- carboranylethylamide(1) for evaluation in BNCT.Materials & Methods: 3-(Bis-2-chloroethyl)amino-2,4-dinitro benzoic acid (Palmer etal., 1994) and 2-carboranyl- ethylazide (Wilson et al., 1992) were synthesized followingmodified literature procedures.Results: 2-Carboranyl-ethylazide was smoothly hydrogenolyzed by hydrogen /10% palladium on charcoal in ethanol containing 1% chloroform (Secrist & Logue,1972) to yield 2-carboranyl- ethylamine-hydrochloride. This was coupled to 3- (bis-2-chloroethyl)amino-2,4-dinitro benzoic acid by DCC / pyridine in DMF. The title compound1 was isolated in good yield by flash chromatography.Discussion, Suggestions for Future Work: A synthetic pathway leading to boroncontaining nitroaniline mustards was established. This compound will be evaluated,among others, with respect to its redox potential, bioreduction, and biodistribution inan animal model bearing a hypoxic tumor. In case, that the solubility turns out to betoo low, we will increase it by degradation of the o-carborane-moiety to the nido-carborane.The synthesis will be expanded to analogs with a different nitro substitutionpattern and different carborane - aniline distances.7 - Development of Intravenously Injectable ChitosanNanoparticles for Neutron Capture Therapy. Yugo Nakatani,Hideki Ichikawa, and Yoshinobu FukumoriFaculty of Pharmaceutical Sciences and High Technology Research Center, KobeGakuin University, Arise 518, Ikawadani-cho, Nishi-ku, Kobe 651-2180, JapanIn our previous studies, gadolinium-loaded chitosan nanoparticles (Gd-nanoCPs) injectedintratumorally (i.t.) exhibited a significant tumor-growth suppression in gadoliniumneutron capture therapy (Gd-NCT) trials. However, it was difficult to achieve acomplete cure, since Gd-nanoCPs administrated via i.t. route might not be distributedevenly in the entire tumor. This implied that delivering Gd through the systemic circulationto the tumor site, for instance, by intravenous (i.v.) injection, might be essential fordistributing Gd over the whole tumor and obtaining higher therapeutic efficiency in Gd-NCT. However, there seemed to be two major problems in developing i.v. injectablechitosan nanoparticles. The first was the rapid release of Gd-DTPA from Gd-nanoCPsin plasma: more than 50% of Gd-DTPA was released from Gd-nanoCPs in humanplasma for 3 h, making it difficult to accumulate Gd in tumor at effective concentrationlevel. The second was such a large particle size of Gd-nanoCPs as about 450 nm. It iswell known that particles larger than 300 nm tend to accumulate in spleen by filtrationand to be taken up by spleen macrophages after i.v. injection. Thus, size-reduction of42

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - C h e m i s t r yGd-nanoCPs was firstly tried by using chitosan of a lower molecular weight. Secondly,an attempt was made to modify the surface of Gd-nanoCPs with colloidal magnetitesto introduce the Gd-nanoCPs to tumor site by applying magnetic field before Gd-DTPAbeing released in blood.Preparation of small-sized Gd-nanoCPs was carried out based on the w/o emulsion-dropletcoalescence technique with a slight modification. Chitosan with differentmolecular weight (950, 50, 10 kDa) was applied at various concentrations (0.5, 1.5,2.5%). As the concentration and the molecular weight of chitosan in the inner aqueousphase decreased, the mean particle sizes gradually decreased; however, the zetapotential and Gd content of Gd-nanoCPs was not significantly affected. The smallestparticle size became 155 nm, when 1.5% Daichitosan 100D (EL)® (Mw=10,000) wasused. This was the size that would make it possible to utilize the enhanced penetrationand retention (EPR) effect for accumulating Gd-nanCPs in tumorGd-nanoCPs surface-modified with colloidal magnetites (Gd-nanoCP-Ms) were preparedas follows. Ten milliliters of Gd-nanoCPs suspension were mixed with 100 ml ofcolloidal magnetite suspension (100 mg Fe /ml), and then stirred by magnetic stirrerfor 1 h at room temperature. The resultant Gd-nanoCP-Ms were washed twice withwater by centrifugation at 1000 rpm for 60 min. The particle size of Gd-nanoCP-Mswas about 194 nm, indicating that non-modified Gd-namoCPs were covered with amonolayer of colloidal magnetites of 20-30 nm. The results gave us a progress to allowi.v. injection of Gd-nanoCPs.43

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - N u c l ea r E n g i n e e r i n g3 - Field Assessment using Structure-segmented Phantom.Chenguang Li*, Thomas E. Blue, Nilendu GuptaNuclear Engineering Program, Department of Mechanical Engineering, The OhioState University, Columbus, OH,43202The purpose of this article is to introduce, describe and demonstrate the use of anew neutron field assessment parameter (NFAP) with which to evaluate complexdose distributions resulting from mixed neutron and gamma-ray fields in Boron NeutronCapture Therapy (BNCT). This new NFAP is based on absorbed dose distributionsfor normal tissues/structures and tumor, and yields a score that accounts for thecompeting goals of sparing normal tissues and maximizing tumor dose. The NFAP isformulated by modifying a previously defined Objective Function, so that it is appropriatefor BNCT. The resulting BNCT Objective Function (BOF) allows for the inclusionof tissue specific Relative Biological Effectiveness (RBE), and tissue specific dosetolerances and weights. Its usefulness is demonstrated in this study for an idealizedneutron beam.5 - Optimizing the OSU-ABNS Base Moderator AssemblyGeometry for BNCT. B. Khorsandi*, T. E. BlueNuclear Engineering Program, The Ohio State University, Columbus, OH 43210,USAIn this paper, the next step in the evolution of the design of the Ohio State University(OSU) Accelerator Based Neutron Source (ABNS) moderator assembly geometry ispresented. MCNP was run to find the most appropriate moderator and reflector thicknessesfor our standard moderator assembly shape, for a revised choice of moderatorassembly materials. The performance of the moderator assemblies was judged byevaluating the intensity and quality of the neutron field using two in-phantom fieldassessment parameters, T and Dtumor|6cm. Then, Fermi Age Theory was used inconjunction with MCNP code runs to refine the shape of the moderator region of themoderator assembly.For our standard moderator assembly shape, the most important dimension of themoderator assembly, with respect to neutron field quality, is the axial thickness of themoderator. The largest part of the moderator assembly is the reflector. Hence, MCNPruns were made to determine the effect of the axial and radial-thicknesses of the reflector,on Dtumor and T, with the axial-thickness of the moderator as a parameter. Inall trials, the outer diameter of the delimiter was set equal to the outer diameter of thereflector.We found that for our standard moderator assembly shape that increasing the radialand/or the axial thickness of the reflector, or decreasing the axial thickness of themoderator, decreases T. The effects of these modifications of the moderator assemblydesign on Dtumor is described in the paper in detail. Generally speaking, we foundthat, for modifications of dimensions of the standardly shaped moderator assembly,those modifications which improve (decrease) T worsen (decrease) Dtumor.Additionally, we found that both T and Dtumor could be simultaneously improved bymodifying the shape of the moderator region of the moderator assembly according tothe predictions of simple Fermi Age Theory calculations. Using this theory to guide ourmodifications of the shape, while using MCNP to determine Dtumor|6cm and T for theresulting moderator shapes; the moderator shape was modified such that shorter Tswere obtained without a corresponding negative impact on Dtumor|6cm. The modifiedshape of the moderator is a cylindrical-truncated cone.The revised base moderator assembly design, which we designate as 25 MgF2/25-56CaF2/CaF2, with a cylindrical-truncated cone moderator, exhibits about a 25% decreasein treatment time, while maintaining adequate neutron field quality comparedwith the previous base moderator assembly design (30 Fluental/33.5-56 PbF2/Li2CO3)with the standard moderator assembly shape.7 - NCT experimental beams of TAPIRO fast reactor. G.Rosi a , S. Agosteo b , G. Curzio c , V. Colli d , A. Del Gratta c , G. Gambarini d , S.Gay d , E. Nava e , C. Petrovich e , L. Scolari da ENEA (FIS-ION), (Rome), Italy, b Nuclear Engineering Department of Polytechnicand INFN, Milan, Italy, c DIMNP, University of Pisa, Italy, d Physic Department of theUniversity and INFN, Milan, Italy, e ENEA (FIS-NUC), Bologna, ItalyTAPIRO is a low-power (5 kW) high-flux (4x10^12 cm^-2 s^-1) fast research reactoroperating at the Casaccia ENEA Centre (Rome). In the main column of this reactor,thermalizing structures can be set up or removed, in order to obtain thermal orepithermal columns, depending on the working opportunity. During last years, thermaland epithermal columns have been designed and built up, aimed at experimentationregarding neutron capture therapy. In particular, the thermal column has been utilizedfor inquiring about cell survival, for validating dosimetry methods and for performingexperimental treatment of small animals with skin melanoma. The epithermal columnhas been utilized for in-phantom dosimetry studies and for treatment of small animalswith glioblastoma. The characteristics of both neutron fields have been enquired, in orderto verify the suitability of a fast reactor for obtaining good quality neutron beams forNCT clinical trials, to have experimental validation of MCNP calculations and, finally,to strengthen the confidence in the good quality of the future beam of the TAPIROnew facility, now in construction, aimed at treating brain glioblastoma in patients. Dosemeasurements have been performed utilizing gel dosimeters and thermoluminescentdosimeters (TLD), flux measurements with activation technique and with TLDs too.9 - “TORT-coupled MCNP4C” technique and its applicationto BNCT related calculations. Y-W H. Liu a, *, U.Y. Chen aand R.J. Sheu ba Department of Engineering and System Science, National Tsing Hua University,Taiwan, ROC, b National Synchrotron Radiation Research Center, Taiwan, ROCConsistent Adjoint Driven Importance Sampling(CADIS)developed by Wagner andHaghighat has been shown to be very useful in accelerating the MCNP calculations. Ituses the adjoint function calculated by TORT to calculate biased sources and weightwindow lower bounds for use in the MCNP calculation for variance reduction. Later,A3MCNP, a code incorporating the above feature into MCNP4A was announced.In this study, based on the CADIS theory, a technique called “TORT-coupled MCNP4C”is established. It is established in a way different from A3MCNP that no change of bothcodes would have to be made. The “TORT-coupled MCNP4C” technique is first testedwith 3 test problems for accuracy and efficiency. It is then applied to the calculation offast neutron flux in a brain-equivalent phantom using an idealized epithermal neutronbeam. Three cases of fast neutron flux tally are analyzed: (1) centerline tally, (2) midplanetally, and (3) volume tally.By using “TORT-coupled MCNP4C”, improvement of FOM’s (figure of merit) by a factorof 9 is observed in the centerline-tally case, and a factor of 3 in the plane-tally case.Even in the volume-tally case, results with 1s 10 keV) dose rate = 3.0_10-3 Gy/s; gamma doserate = 3.0_10-4 Gy/s. According to phantom studies, the maximum dose in tissue isat a depth of a0.5 cm in the irradiated object. This channel is used for preclinical NCTstudies in cell cultures, small laboratory animals and dogs with spontaneous melanomawith use of various drugs.For the treatment of deep-seated tumors it was suggested to implement an irradiationroom for clinical NCT studies on the channel passing through the thermal column. Forthis purpose, it is necessary to redesign the thermal column, including partial replacementof graphite with a combined aluminum-based thermal/epithermal neutron spectrumshaper. The existing shutter will be replaced with a shutter of a new rotary design;also, a new collimating device will be designed. Preliminary calculations have shownthe potential of the redesigned thermal column to provide an epithermal neutron flux of2_109 n/cm2s, with the concomitant total dose of fast neutrons and photons not morethan 5_10-13 Gy per unit flux of thermal or epithermal neutrons. This neutron beam willbe the base of an irradiation clinical NCT facility for the treatment of both surface anddeep-seated tumors. The proportion of thermal and epithermal neutrons in the beamcan be varied using 6Li-containing filters.The thermal column redesign project being developed includes the following issues:- full-scale design studies with the purpose to determine the composition and geometryof the area of formation of epithermal and thermal neutron beam in the thermalcolumn;- development of engineering specifications for the redesign of the thermal columnand new rotary shutter;- production of the shutter and collimator;- dismantlement of the thermal column and installation of the new thermal and epithermalneutron spectrum shaper;- measurements of the neutron beam characteristics, and phantom studies;44

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - N u c l ea r E n g i n e e r i n g- studies on new neutron capture compounds in the new neutron beam;- certification of the implemented clinical irradiation base by the results of physicalmeasurements, and the results of preclinical trials in dogs with spontaneous tumorswith use of the developed compounds.Basing on these trials, recommendations on the clinical application of NCT on theimplemented irradiation base will be set forth. Two channels of the MEPhI reactor willbe the first Russian base for specialized experimental and clinical research on NCTof cancer.References1. Khokhlov, V.N. Kulakov, K.N. Zaitsev, et al., The Russian Project on Neutron Capture Therapy forCancer, Frontiers in Neutron Capture, ed. by Hawthorne et al. Kluwer Academic/Plenum Publishers,N-Y, 2001. p.425-428.13 - Model of focusing capillary neutron optics system(CNOS) for invasive neutron capture therapy (INCT). G.I.Borisov, R.I. Kondratenko, M.A. KumakhovInstitute for Roentgen Optics, RRC “Kurchatov Institute” , Moscow, Russian Federation/ Unisantis SA, Geneva, SwitzerlandA model of focusing assembled CNOS consisting of 85 straight mono capillaries forminga hexagonal conic assembly. Small base of cone is fully covered by mono capillaries,and the space between diverging capillaries inside the CNOS cone is filledwith materials effectively absorbing or scattering neutrons and hard photon radiation.In this instance, each capillary performs the function of a collimator within its angulargeometric aperture and of a neutron guide of thermal neutrons within respective waveaperture. This enables larger partial contribution of thermal neutrons into the total fluenceof the beam formed. Thermal neutrons, in this case, are reflected from the wallson inner holes of mono capillaries not more than once ensuring maximum possibletransmission of CNOS.Mono capillary glass type AB-1; Inner diameter of mono capillary hole, mm 2.0; Monocapillary length, mm 640; Number of reflections of neutrons from mono capillary walls0.64; Mono capillary geometric aperture 7.6◊10 -7; Mono capillary wave aperture1.0◊10 -6; Total area of CNOS holes, cm2 2.67; CNOS focal distance, mm 1000;Diameter of CNOS focal spot, mm 6; CNOS transmission, % 66; Thermal neutronsfluence in CNOS focal spot at 5 MW, s-1 reactor power 1.9◊10 7The obtained transmission is explained by errors in mono capillary manufacture andCNOS assembly. The developed CNOS is used at the horizontal tangent reactorIR-8 RRC “Kurchatov Institute” for INCT investigations with experimental animals. Inthis instance, melanoma B-16 on mouse leg receives dosage of 2.3 Gr for 15 hours,concentration there being B(10) 8ppm.15 - RFI Linac for Accelerator-based Neutrons.W. Joel Starling, Chief EngineerLinac Systems, LLCA new type of radio-frequency (rf) linear accelerator (linac) under development at LinacSystems, LLC in Albuquerque, NM, could serve as the ideal accelerator componentof an accelerator-based neutron source for Neutron Capture Therapy. Supported bya Small Business Innovative Research (SBIR) grant from the US Department of Energy(DOE), a prototype of this patented rf linac is currently under construction. Themechanical design of the linac structures will be shown, the status of the componentswill be reported, the neutron source driven by this rf linac will be described, and thecalculated performance of the entire system will be presented.The Rf Focused Interdigital (RFI) linac is basically an interdigital (or Wideröe) drifttube rf linac structure with rf quadrupole focusing incorporated into each drift tube. The200-MHz RFI prototype comprises a microwave ion source, an articulated Einzel lenslow-energy beam transport (LEBT) section for steering and focusing the proton beamfor injection into a short radio-frequency quadrupole (RFQ) linac, followed by the RFIlinac. The overall length of the RFI prototype is approximately 3 meters. The RFI linacstructure is about 5 times more efficient than the conventional Drift Tube Linac (DTL)structure in this energy range.Due to budgetary constraints, however, a planned limitation in the rf power system willallow testing of the RFI prototype to only 33% of its full capability. The peak linac structurepower of 66 kW and the peak beam power of 50 kW will be supplied by a 144-kW,33% duty rf power system. Each of the linac structures can be tested, individually, incw operation; but they can only be tested with accelerated beam at a still significant33% duty factor - resulting in an average proton beam current on target of 6.6 mA.This prototype would be available for use at its full duty factor given an upgrade in itsrf power system. The RFI prototype is scheduled to come into operation by the endof this year.The solid Lithium target and neutron moderator are areas requiring further development.Designs and plans for the neutron source’s high-energy beam transport (HEBT),solid Lithium target and neutron moderator have been developed, but funding hasnot yet been secured to build and test these components. Linac Systems expects,moreover, that neutron capture therapy groups have their own set of target and neutronmoderator specifications or designs to which our RFI linac technology could betailored.17 - Analysis of Intracellular Distribution of Boron andGadolinium in 9L Sarcoma Cells Using a Single-EndedAccelerator (Micro PIXE). Kiyoshi Endo a , Yasushi Shibata a ,Humiyo Yoshida a , Kei Nakai a , Tetsuya Yamamoto a , Akira Matsumura a , KeizoIshii b , Takuro Sakai c , Takahiro Sato c , Masakazu Oikawa c , Kazuo Arakawa c ,Hiroaki Kumada d , Kazuyoshi Yamamoto da Department of Neurosurgery, Institute of Clinical Medicine, University of Tsukuba,Tsukuba, Japan, b Department of Engineering, University of Tohoku, Sendai, Japan,c Department of Ion-beam-applied Biology, JAERI (Japan Atomic Energy ResearchInstitute), Takasaki, Japan, d Department of Research Reactor, JAERI, Tokai, JapanIn patients with a malignant brain tumor, we usually first excise as much of the tumormass as possible and then prepare the patient for radiation therapy and or chemotherapyfor tumor cells that might remain diffusely distributed in normal brain tissue. Thistherapeutic strategy has not proved completely effective so far, thus we have been investigatingusefulness of Boron Neutron Capture Therapy (BNCT). And we have beeninvestigating in Boron and Gadolinium as elements to capture neutrons.The objective of this study was to investigate the movement and distribution of thesetwo elements in the cells. We also discuss the possibility of using Gadolinium for clinicalmedical therapy.We used the Single-Ended Accelerator (Micro PIXE) at JAERI Takasaki in TakasakiCity, Gunma Prefecture to quantify intracellular boron and gadolinium. This machineproduces a micro beam of 1 micrometer in diameter and allows us to analyze the distributionof these elements directly.As for the procedure, first we fix the mayra membrane with a glass ring and a bite ringof 2cm in diameter. After washing well, we put 9L sarcoma cells on this membrane andculture these cells in minimum essential medium(MEM) at 37°C until they form a monolayer.Next we add the Gd-BOPTA to the culture and incubate the cells for 24~72hrsat 37°C. Then we wash the membrane in THAM liquid and place it immediately aboveliquid nitrogen for few minutes. After that the membrane is vacuum-dried for 24 hoursto fix these cells on the holder. In this phase, the most important thing is to fix the cellsevenly to analyze the distribution of elements in the cells with Micro PIXE.According to the results of the latest experiment, we were able to analyze the distributionof P, S, Gd, etc. And we could see them in the cells themselves. But we observedthe distribution of K and Gd around the cells which are in the cells originally. Thisindicated the cells wall had been destroyed or just damaged, and the things of intracellular were flown out to the extra cells by some reasons. Or gadolinium had not beenuptake by the cells, or that washing was insufficient. Another reason is the way of cellsfixation, we think. Now we are trying to solve some problems regarding analysis, fixationmethod, technical problems, etc.In the future, we hope to be able to analyze the intracellular distribution of boron, andto get information about gadolinium and boron. Then we hope to it will be possible toapply these two elements as a more effective clinical BNCT.45The specifications of the RFI prototype have been influenced by collaborations withNCT researchers, and as such, the prototype is designed to support the P-Li nuclearreaction by supplying an average proton current of 20 mA in continuous-wave (cw)operation on a solid lithium target at a beam energy of 2.5 MeV. It should be notedthat the technology represented in the RFI prototype is not limited to the P-Li reaction– other nuclear reactions could be supported by simply tailoring the accelerator to supplythe appropriate particle beam current and energy.

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - P hy s i c s1 - Monte-Carlo calculations for the development of aBNCT neutron source at the Kyiv Research Reactor.O.O. Gritzay a, *, O.I. Kalchenko a , N.A. Klimova a , V.F. Razbudey a , A.I. Sanzhura , S.E. Binney ba Institute for Nuclear Research, Kyiv, Ukraine, b Oregon State University, Corvallis,OR, USAThe results presented in this paper display our continuing steps toward development ofa neutron source with parameters required by Boron Neutron Capture Therapy(BNCT)at the Kyiv Research Reactor (KRR). The purpose of this work was:1) calculation of the neutron flux which can be achieved at the greatest possible approachof a patient to the reactor core (380 cm from the core center);2) analysis of the influence of a nickel collimator and a nickel-60 filter on the characteristicsof the neutron beam;3) creation and validation of the MCNP calculational pattern for an actual core loadin the KRR.Results of calculations were carried out by means of the MCNP4C code included:1) An epithermal neutron flux of 3e+9 to 5e+9 neutron/cm^2/s with an epithermal-tofastflux ratio of 80 to 230 could be obtained at the KRR, using a natural nickel layeron the interior borated polyethylene collimator wall and a nickel-60 filter.2) Use of the nickel-60 filter may be useful to increase the ratio epithermal-to-fastflux without a substantial decrease in the magnitude of the epithermal neutron flux.3) The MCNP model proposed in this paper could also be useful for reactor safetycalculations.3 - Characteristics of BDE dependent on 10 B concentrationfor accelerator-based BNCT using near-threshold7Li(p,n) 7 Be direct neutrons. K. Tanaka a, *, T. Kobayashi b , G. Benguab , Y. Nakagawa c , S. Endo a , M. Hoshi aa Research Institute for Radiation Biology and Medicine, Hiroshima University, Kasumi1-2-3, Minami-ku, Hiroshima 734-8553, Japan, b Research Reactor Institute,Kyoto University, Asashiro-Nishi 2-1010, Kumatori-cho, Sennann-gun, Osaka 590-0494, Japan, c National Kagawa Children’s Hospital, Zentsuji-cho 2063, Zentsuji,Kagawa 765-8501, JapanIntroduction: Intra-operative boron neutron capture therapy (BNCT) conducted in Japanhas the advantages of (1) high dose delivery to deeper regions and (2) the possibilityto be performed together with the procedure to debulk the tumor in one surgicaloperation if accelerator-based irradiation systems are installed at hospitals. In orderto realize compact irradiation systems with flexible irradiation directions suitable for it,we have revealed the feasibility of intra-operative BNCT using direct neutrons fromthe 7Li(p,n)7Be reaction near its threshold (1.881 MeV) . The judging parameter wasthe size of the region satisfying the dose requirements in protocol (treatable region),for Japanese intra-operative BNCT of brain tumors. Then, the boron-dose enhancer(BDE) has been introduced to increase the contribution of the 10B(n,a)7Li dose. Thetreatable protocol depth (TPD), which is the greatest depth of the treatable region usuallyon its centerline of the irradiation field, was introduced as a possible index for BDEmaterials based on ease and simplicity in comparing physical characteristics of irradiationfields. However, the 10B concentration is different for each irradiation at presentdue to the differences of patient physiology and amount induced, etc. In this study, thecharacteristics of the BDE are identified as to the dependence of polyethylene BDEthickness on the 10B concentration using TPD as an evaluation index.Materials & Method: The production of neutrons and gamma rays in the Li target wassimulated with Lee’s method and their transport was calculated using MCNP-4B.These methods were validated through phantom experiments considering angulardependences. Doses in a cylindrical water phantom of 18 cm in diameter and 20 cmin length were computed for the proton energy of 1.900 MeV. BDE was assumed tobe polyethylene cylinders of 18 cm in diameter. The present dose protocol for intraoperativeBNCT for brain tumors in Japan are specified in terms of the treatable dosefor tumor due to HCP (15Gy), the tolerance dose for normal tissue due to HCP (15Gy)and the tolerance dose for normal tissue due to gamma rays (10Gy). The doses areevaluated as physical absorbed dose. TPD for above mentioned dose protocol wasevaluated for 10BTumor values of 10 ppm to 100 ppm and the 10BTumor to 10BNormalratio (T/N ratio) of 2 to 10.Results & Conclusions: The TPD was increased by increasing T/N ratio, and by increasing10BTumor and 10BNormal for constant T/N ratio. In the case where theachievable range of T/N ratio is the same, an increase in 10BTumor has the advantageover a decrease in 10BNormal that the TPD is increased more effectively. However,the BDE thickness (BDE(TPDmax)) for the maximum TPD (TPDmax) is influencedmore rapidly, which means that the fluctuation of 10BTumor causes more deviationon BDE(TPDmax). As the achievable 10BTumor increases over a certain level,BDE(TPDmax) becomes zero. This demonstrates the effectiveness of the compactirradiation system without BDE and the feasibility of its use in BNCT.5 - Dose-rate scaling factor estimation of THOR BNCTtest beam. F.Y. Hsu a, *, C.J. Tung b , J.C. Chen c , Y.L. Wang b , H.C. Huang b ,R.G. Zamenhof da Yuanpei University of Science and Technology, 306 Yuanpei Street, Hsinchu 300,Taiwan, b National Tsing Hua University, 101 Section 2 Kuang Fu Road, Hsinchu300, Taiwan, c National Yang Ming University, 155 Section 2 Li Nong Street, Peitou,Taipei 112, Taiwan, d Tufts-New England Medical Center, 750 Washington Street,Boston, MA 02111, USAIntroduction: A test beam was build at Tsing Hua Open-pool Reactor (THOR) for BNCTpreliminary experiments since 1998 and a new beam was designed and was underconstruction for further clinical trials. Dose rate scaling factor (DRSF) is defined asthe measured dose rates in a physical phantom divided by the treatment planningcomputed dose rates in a corresponding phantom. Because the differences betweenmeasured and computed dose rate results are always exist, DRSF relates to modifycomputed dose rates to approach measured dose rates. As the value of DRSFs areestimated and input in the BNCT treatment planning: NCTPlan, the computed dosedistributions will represent the irradiated dose rate distributions that a phantom or apatient really received.Materials and Methods: Neutron activation analysis and dual ion chamber techniqueswere choosed to measure and analyze the depth dose rate distributions of THORBNCT test beam. Structured gamma and neutron spectra at the test beam exit wereapplied as the reactor source in the treatment planning code. Depth dose rate distributionsof all dose components, thermal neutron, fast neutron, photon and boron-10,of THOR BNCT test beam were measured and analyzed by means of relevant instrumentsand techniques. A new PC version, Monte-Carlo based, treatment planningcode: NCTPlan, developed by the BNCT medical physics group of Beth IsraelDeaconess Medical Center (BIDMC) and a Snyder head phantom were applied tocompute and assess the dose rate distributions of THOR BNCT test beam in this work.Dosimetric experiments were performed under 1.5MW reactor power condition, andthe Snyder head phantom was located at the front of exit point of THOR BNCT testbeam. Boron-10 concentrations considered in tumor and normal tissue preset as 30ppm and 7.5 ppm respectively.Results and Discussion: Depth dose rate distributions of thermal neutron, fast neutron,photon, boron-10 in tumor, boron ¡V10 in normal tissue, tumor total dose and normaltissue total dose were measured and represented respectively. Maximum tumor totaldose rate occurs at about 2 cm depth. Values of DRSFs were estimated: 0.55 forthermal neutron, 2.88 for fast neutron, 0.86 for photon, 0.57 for boron-10 (normal) and0.56 for boron-10 (tumor). Weighted NCTPlan (NCTPlan*DRSF) computed dose rateswere matched with measured results very well (less than 1 % difference) for thermalneutron, fast neutron and boron both in tumor and normal tissue. For fast neutron, ithas larger difference (about 13%). This may be caused by the uncertainty of the MonteCarlo reactor source file.Conclusions: Monte Carlo based BNCT treatment planning is a convenient tool forcomputing the therapeutic dose distributions. Accuracy of treatment planning is veryimportant for radiotherapy. Estimation of DRSF is a necessary and important work forany reactor beam running BNCT. The techniques and experiences performed in thiswork will continually improved and applied in the next tests and trials of THOR BNCTnew beam in later 2004.7 - Tape high power neutron producing target for NCT. V.Kononov a , G. Smirnov b , S. Taskaev c, *a Institute of Physics and Power Engineering, 1 Bondarenko Sq., 249020 Obninsk,Kaluga Region, Russia, b Russian Federal Nuclear Center – Institute of TechnicalPhysics, m.b. 245, 456770 Snezhinsk, Chelyabinsk region, Russia, c Budker Instituteof Nuclear Physics, 11 Lavrentiev ave., 630090 Novosibirsk, RussiaSolid lithium targets with intense liquid cooling are used now for accelerator based boronneutron capture therapy. Typical power of available proton beam is within 5 kW, theone of the accelerators under construction – 25 kW. New approach in target conceptionis needed for more high power proton beam developed for decrease of treatmenttime. Innovative lithium target for more high power proton beams is proposed. Mainidea is based on energy accumulation at movable tape. The target is made of flexiblecarbon fiber tape. Its moving device relative to charged particle beam directed to thetarget contains tape transport mechanism with magazines for tape introduction andreceiving with winding drums. This tape can be made by coating the layer of neutronproducing matter (hydride, nitride, oxide or fluoride of lithium) on the substrate. Theactive matter layer can be covered by protective film, of lithium oxide, for example toprevent from mechanical damage. To improve the heat removal from the operatingpart the substrate can be made of the substance with high thermal conductivity, ofaluminum, for instance. In case if coefficient of thermal expansion of active materialand substrate matter are different the active material can be put on substrate (tape) in46

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - P hy s i c sfragments along the tape with their length comparable with the tape width. The patentwas received for this target.9 - Development of equipments for determination ofBNCT source spectral parameters. J. Burian a, *, B. Jansky a ,M. Marek a , E. Novak a , L. Viererbl a , A.C. Fernandes b , Yu.A. Kaschuck c , L.A.Trykov d , V.S. Volkov da Nuclear Research Institute Rez, plc, Husinec-Rez, Rez near Prague 25068, CzechRepublic, b Nuclear and Technological Institute, Sacavem, Portugal, c Troitsk Institutefor Innovation and Fusion Research, Troitsk, Russia, d Institute of Physics andPower Engineering, Obninsk, RussiaThe knowledge of neutron and gamma ray energy spectra can strongly influence theBNCT information about delivered dose to target volume as well as to the surfacehealthy tissue region. This region is very often decisive to keep assessed healthy tissuelimit. Modification of neutron Bonner spectrometer to one block i.e. Bonner SpectrometerMonoblock (BSM) and gamma ray Si semiconductor spectrometer are beingdeveloped and verified in real conditions of LVR-15 reactor beam for these purposes.BSM consists of polyethylene (PE) block with Cd and PE with boron shielding. Sevendetectors of thermal neutrons (DTN) are inserted in seven measuring channels withdifferent thickness of PE for on-line measurement in geometry of scattered beam. Aspecial insertion with set of foils is used for irradiation in the direct beam. DesignedBonner Spectrometer Monoblock (BSM) is presented in paper as Fig.1.Gamma radiation spectrometer dosimeter SPEDOG was chosen as a base for developmentof methodology for scattered gamma ray beam measurement. Silicon detectoris sensitive, less then 10+ 4 counts/s is permitted input level. The principles of Comptonscattering can be applied when materials with low Z (atomic number) are used asscattering sample. Thin layer of aluminum can be used for these purposes.The first test measurements for development of spectrometers were implemented inNRI in conditions of known, standard spectra (NRI Laboratory of Spectrometry), aswell as in conditions of reactor beam (reactor LVR-15).Laboratory of Spectrometry: Neutron and gamma spectra escaping the surface of ironspheres of 30, 50 and 100 cm in diameter and water spheres of 30 and 50 cm in diameterwere used for that test measurement. A neutron source Cf-252 with the emissionof 3x10+8 n/s was positioned in the center of a sphere. Typical configuration for measurementin standard fields is shown in Fig.2. The cones, used for neutron and gammashielding enabled to determine the background in real conditions. Different materialsand thickness of cones were used: PE+B (polyethylene with boron) for neutrons, Feironfor gamma rays.Reactor beam: Epithermal neutron beam was constructed on LVR-15 reactor, the facilityhas been used for BNCT clinical trials. Today free beam parameters are: Fepi =7.13 x10+8 /cm2s, Ffast = 5.16x10+7 /cm2s, Dg = 1.98 Gy/h. The first test measurementfor developed equipments was performed in October 2003, development willcontinue during 2004. The Fig.3 is showing the sensitivity of BSM readings in differentspectra. In Fig.4 gamma spectra with and without scattering sample are demonstratedin group structure.11 - The irradiation system and dose estimation joint-systemfor NCT wider application in Kyoto University. Y.Sakurai*, A. Maruhashi, K. OnoKyoto University Research Reactor Institute, KURRI, Asashironishi 2-1010, Kumatoricho,Sennan-gun, Osaka 590-0494, JapanThe research for neutron capture therapy (NCT) at the Kyoto University ResearchReactor (KUR) is remarkably developing for the last few years. However, the mostimportant subject is the preparations for the KUR provisional shutdown coming inMarch 2006. In this paper, our present concept and plan are reported about the novelirradiation system and dose estimation system for wider applications of NCT. For theirradiation field, the target nuclear reaction was selected to 7Li(p,n)7Be and the neutronmoderator was selected to heavy water. The minimum proton current was about13 mA for epi-thermal neutron irradiation, and about 9mA for mix-neutron irradiation.In thermal neutron irradiation, the proton current needed more than 18 mA for 2.5-MeVprotons, but only 4 mA for 5.0-MeV protons. For the dose estimation system, we areaiming at the completion of the “dose estimation joint-system”. The data from the onlinemeasurement systems such as beam monitors and gamma-ray telescopes arefed back to the results for the in-body dose estimation, and then the dose estimationsfor irradiation field and a living body are jointed. For the beam-monitor system, multichambermethod was adopted. The surveys were performed for the wall materials andchamber-gases.13 - Design and construction of shoulder recesses intothe beam aperture shields for improved patient positioningat the FiR 1 BNCT facility. I. Auterinen a, *, P. Kotiluoto a , E. Hippeläinenb , M. Kortesniemi b,e , T. Seppälä b , T. Serén a , V. Mannila c , P.Pöyry c , L.Kankaanranta d , J. Collan d , M. Kouri d , H. Joensuu d , S. Savolainen b,c,ea VTT Processes, POB 1608, FIN-02044 VTT, Finland, b Boneca Corporation, POB700, FIN-00029 HUS, Finland, c Department of Physical Sciences, University ofHelsinki,POB 64, FIN-00014, Finland, d Department of Oncology, Helsinki UniversityCentral Hospital, POB 180, FIN-00029 HUS, Finland, e Helsinki Medical ImagingCenter, University of Helsinki, P.O. Box 340, FIN-00029 HUS, FinlandIntroduction: At the FiR 1 BNCT facility when a patient has been positioned for a lateralirradiation field the supine position has not been applicable because of collisionof shoulders and the beam aperture collar. To facilitate this situation and to improvepatient comfort shoulder recesses were designed horizontally on both sides of thebeam aperture.The increase in the radiation dose to the patient’s body was a concern and this wasstudied both by modeling with the MCNP code and by performing dose measurementsbefore and after the modifications at the irradiation facility.Materials and Methods: The geometrical modeling for the shoulder recesses was doneusing computer modeling, the treatment planning system and plastic models testedin the treatment simulator. Dose simulations were performed with MCNP using ananthropomorphic human model based on the BOMAB phantom. Measurements ofthe effect of the recesses to the dose distribution around the beam aperture wereperformed using the twin ionization chamber technique.Results: MCNP simulations showed that the main contribution to the increase in theeffective dose was from the neutron dose of the arm. Dose measurements using thetwin ionization chamber technique showed that neutron dose increased on the sidesas predicted by the MCNP model but there was no noticeable change in the gammadoses.Discussion: In the design of the facility the whole body dose of the patient was estimatedwith a DORT model to be 0.6 Sv/h. In this work for the worst case with the arminside the shoulder recess the estimate was 0.7 Sv/h. The increase of the effectivedose to the patient’s body (mainly due to the dose increase in arm) was consideredacceptable, and a decision was made to construct recesses for shoulders by modifyingthe beam aperture.When making the recesses into the lithium containing neutron shield material tritiumcontamination had to be taken into account. An underpressurised glove box was constructedand machine tools with local exhaust were used to confine the sawdust.According to the first experiences with head and neck tumor patients the new shoulderrecesses allow for more comfortable treatment position.Conclusions: The shoulder recesses give space to more flexible patient positioningand can be considered as a significant improvement of the Finnish BNCT facility. Bylimiting the depth and height of the recesses the effect on the neutron shielding wasminimized.15 - A dosimetric study on the use of bolus materials fortreatment of superficial tumors with BNCT. Tiina Seppälä a,b, *,Juhani Collan c , Iiro Auterinen d , Tom Serén d , Eero Petri Kotiluoto d , MikaKortesniemi a,e , Koen Van Leemput e , Leena Kankaanranta c , Heikki Joensuuc , Sauli Savolainen ea Boneca Corporation, P.O.B. 700, FIN-00029 HUS, Helsinki, Finland, b Departmentof Physical Sciences, P.O.B. 64, FIN-00014 University of Helsinki, Finland, c Departmentof Oncology, Helsinki University Central Hospital, P.O.B. 180, FIN-00029 HUS,Finland, d VTT Processes, P.O.B. 1608, FIN-02044 VTT, Finland, e HUS, MedicalImaging Center, University of Helsinki, P.O.B. 340, FIN-00029 HUS, FinlandFor treatment of superficially located tumors, such as head and neck cancers thatinvade the skin, the tumor dose may remain low on the skin when such tumors aretreated with epithermal neutrons in boron neutron capture therapy (BNCT). When theFinnish research reactor (FiR 1) epithermal neutron beam is used, the total tumordose on skin is approximately only 35% of the dose maximum. In conventional radiationtherapy a shaped bolus has been used as a scattering material placed on top ofthe target to increase the surface dose. The aim of the present study was to examinethe effects of bolus material for treatment of superficially located tumors with BNCT, toverify the calculated (n,g) reactions rates of Mn-55 and Au-197 and the neutron andthe gamma doses in a phantom with bolus, to measure the neutron activation of thebolus materials after irradiation, and to estimate, using data from the depth dose, whenit might be advantageous to use a bolus in BNCT.47

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - P hy s i c s48A paraffin bolus (thickness 7.6 mm, diameter 200 mm) and a water gel bolus (thickness4.4 mm, diameter 200 mm) were constructed. The calculated neutron and gamma distributionsin the cylindrical PMMA phantom (diameter 200 mm, length 240 mm) with abolus were verified with Au and Mn activation foils and twin ionization chamber measurements.The neutron activation of the bolus materials was measured. The reactionand the dose rates were calculated with SERA treatment planning system (TPS). Theeffect of a 5 mm thick paraffin bolus placed on the skin next to the planning target volume(PTV) of recurrent head and neck carcinoma with skin invasion was investigated.Two dose plans, with and without a bolus, were computed.The measured and the calculated Mn-55(n,g) and Au-197(n,g) activation reactionrates agreed within the measurement uncertainties (3%, 1SD), except at the depthsof 0-1 cm. The measured gamma dose rates were larger than the ones expectedbased on calculations for the deeper parts of the phantom when a bolus was used.The measured and the calculated neutron dose rates agreed within measurementuncertainties (21%, 1SD). The induced neutron activation of the paraffin and water gelmeasured using a standard survey meter immediately after irradiation was less thanbackground reading. A 5 mm thick paraffin bolus placed on the patient skin increasedthe calculated total weighted tumor dose by 50% in the superficial part of the tumor.However, based on the phantom measurements, even 60% increase in the skin dosecan be achieved using the bolus.According to the calculations and the measurements performed, both paraffin andwater gel can be used as a bolus material in BNCT. However, paraffin is recommendedfor clinical practice, as it is durable and can be easily shaped. A 5 mm thick paraffinbolus increases the surface dose by approximately 50%, which may be advantageousin the treatment of superficial tumors where PTV is located in tissue depths less than6 cm as measured from the skin surface.17 - Source Model Development for a Fast NeutronTherapy Planning System. G.N. Malyshkin, E.A.Kashaeva,R.F.Mukhamadiev, V.G.Orlov and S.I.SamarinRussian Federal Nuclear Center - VNIITF named after academician E.I.Zababakhin,456770, P.O.Box 245, Snezhinsk, Chelyabinsk Region, RussiaThe SERA treatment planning system is now being adapted to the Snezhinsk NeutronTherapy Center based on the generator of 14-MeV neutrons. Dosimetric planning isbased on Monte Carlo simulations of dose deposition in the patient’s body under givenirradiation conditions.SERA defines the source of radiation as an energy or angular distribution of neutronsand photons on a plane located between the real source and the exposed objectperpendicular to the radiation beam axis. The adequate simulation of the irradiationprocess requires a lot of preparatory work to determine the source plane position anddevelop a simplified model of the generator.The paper describes this effort implemented with the use of the PRIZMA code - a universal3D Monte-Carlo code designed for linear radiation transport simulations.19 - Dose Calculations in “Dose-Supplementary” Therapyof Cancer. I.N.Sheino, V.F.Khokhlov, V.N.KulakovSSC-Institute of biophysics, Moscow, Russia21 - High-resolution alpha-autoradiography w/ contactUV microscopy technique. K. Amemiya a, *, H. Takahashi b , Y. Kajimotoa , T. Wachi a , M. Nakazawa a , Y. Nakagawa c , H. Yanagie d , T. Hisa d , M.Eriguchi d , T. Majima e , T. Kageji f , S. Miyatake g , S. Kawabata g , Y. Sakurai h , T.Kobayashi h , N. Yasuda i , M. Kagawa j , K. Oguraka Department of Quantum Engineering and Systems Science, The University of Tokyo,Tokyo 113-8656, Japan, b Research into Artifacts, Center of Engineering, TheUniversity of Tokyo, Tokyo, 153-8904, Japan, c Department of Neurosurgery, NationalKagawa Children’s Hospital, Kagawa 765-8501, Japan, d Research Centerfor Advanced Science and Technology, The University of Tokyo, 153-8904, Japan,e Photonics Research Institute, National Institute of Advanced Industrial Science andTechnology, Ibaraki 305-8568, Japan, f Department of Neurological Surgery, The Universityof Tokushima, Tokushima 770-8503, Japan, g Department of NeurosurgeryOsaka Medical College, Osaka 569-8686, Japan, h Research Reactor Institute, KyotoUniversity, Osaka 590-0494, Japan, i Research Center for Radiation Safety, NationalInstitute of Radiological Sciences, Chiba 263-8555, Japan, j Department of Pathology,Osaka Medical College, Osaka 569-8686, Japan, k College of industrial Technology,Nihon University, Chiba 275-8575, JapanThe information on subcellular microdistribution of 10B compounds inside a cell is significantto evaluate the efficacy of boron neutron capture therapy (BNCT) because thedamage brought by the released alpha/lithium particles is highly localized along theirpath, and radiation sensitivity is quite different among each cell organelles. In neutroninducedalpha-autoradiography (NIAR) technique, 10B can be measured as tracks forthe energetic charged particles from 10B(n,alpha)7Li reactions in solid state track detectors.To perform the NIAR at intracellular structure level for research of 10B uptakeand/or microdosimetry in BNCT, we have developed high-resolution NIAR method accompanyingwith contact ultra-violet (UV) microscopy technique. In this technique, wecan observe not only charged particles tracks for alpha and lithium on CR-39 plastictrack detectors, but also transmission UV image of cells on the same CR-39 plasticssimultaneously. In the UV imaging, UV of 254 nm in wavelength from commerciallyavailable low-pressure mercury lamp irradiates biological specimen mounted on a CR-39 plastic. Nucleic acids mainly absorb the UV, and then we can observe transmissionUV image of the biological cells as a relief on the CR-39 with an atomic forcemicroscope (AFM) after etching process. Lateral resolution of the UV imaging is expectedto be < 100 nm, which is sufficient for the imaging of cellular histology. Control,BSH-administrated and BPA-administrated tumor tissues from tumor-transplanted ratswere used in this feasibility research. The tissue samples were fixed and embeddedin epoxy resin. And then the sliced sections of the samples were mounted on CR-39plastics and irradiated by thermal neutrons. The irradiated samples were exposed toUV from germicidal lamp. After that the samples were etched in hot NaOH solution forshort time. The etched surface was observed with an AFM. We achieved clear visualizationof etch pits for alpha/lithium particle tracks and the cellular histology as reliefon the CR-39 surface simultaneously. The track counts in the high-resolution NIARimage show good linear correlation with the 10B concentration. The method describedhere has the potential of the direct observation of the generated energetic chargedparticles, including background protons, in cellular histology in BNCT. The processesare simple, therefore the routine use is expected, for example, to test the microdistributionof newly developed 10B compounds, to perform microdosimetry in BNCT and soon. In further studies a significant effort must be directed toward the development ofa more reliable procedure for sample preparation, such as the use of the sections ofquickly frozen tissues.23 - Perturbation by Ionization Chambers. Hanna Koivunoro,Iiro Auterinen, Antti Kosunen, Petri Kotiluoto, Sauli SavolainenDepartment of Physical Sciences, University of Helsinki, P.O. Box 64, FIN-00014 HelsinkiUniversity, Finland; VTT Processes, VTT Technical Research Centre of Finland,P.O. Box 1608, FIN-02044 VTT, Finland; Radiation Metrology Laboratory, Radiationand Nuclear Safety Authority-STUK, FIN-00881, Helsinki, Finland; Departments ofRadiology and Laboratory Diagnostics, Helsinki University Central Hospital, P.O. Box340, FIN-00029 HUS, FinlandThe neutron and photon dosimetry procedure of an epithermal neutron beam includestwin ionization chamber measurements. Tissue equivalent TE(TE) chambers are usedfor total dose detection and neutron-insensitive Mg(Ar) chambers for gamma dosedetection. Relative gamma and neutron sensitivities for both detectors have been determinedand a measurement procedure has been established for epithermal neutronbeam measurements at the Finnish BNCT facility, FiR 1. However, the uncertainty withthis method remains unsatisfactorily high (5-13%) for both radiation qualities. One reasonfor the uncertainty is the disturbance to the neutron/photon field by the chambermaterials in the phantom. The purpose of this work was to determine the perturbationto the photon and neutron fluences caused by the ionization chamber while thechamber is placed in the phantom. Perturbations of the charged particles were notconsidered. The perturbation to the photon and neutron dose was determined for perpendicularand parallel chamber orientations relative to the beam central axis.The construction and materials of a Mg(Ar) ionization chamber and a TE(TE) ionizationchamber in a 30cmX30cmX30cm water phantom with PMMA walls were modeledwith MCNP. Simulations were performed at the central axis on the 14cm diametercircular beam at the depths of 2.5cm (close to the photon dose maximum in water)and 6cm from the beam aperture plane. To investigate the photon and neutron fluenceperturbations, absorbed doses to water were determined in the gas volume inside thechambers and in the same geometry with the chamber materials replaced with water.The ratio of these two doses is used here as a fluence perturbation effect. The neutrondose was calculated using fluence-to-kerma conversion factors and the photon doseusing energy absorption coefficients taken from ICRU report 46. Separate simulationswere performed with the photon source component and the neutron source component,which in addition to the neutron dose generates the induced photon dose, themajority of the total photon dose.At the 2.5cm depth, the presence of the Mg(Ar) chamber decreased the photon doseby 16±0.3% and at the 6cm depth 8±0.2%, when the chamber was placed perpendicularto the beam axis. When the Mg(Ar) chamber was placed parallel to the beam axis,the photon dose rate at 2.5cm decreased by 17±0.3% and at 6cm, by 10±0.3%. TheTE(TE) chamber perpendicular to the beam axis caused 6±0.5% decrease in photon

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - P hy s i c sdose and 5±1.7% increase in neutron dose at 2.5cm. At 6cm, photon dose decreasedby 3.5±0.2% and neutron dose increased by 3.5±0.8%. At 2.5 cm, the TE(TE) chamberparallel to the beam axis caused 6.9±0.3% decrease in photon dose and 7.5±2%increase in neutron dose. At 6cm, photon dose decreased by 3.8±0.2% and neutrondose increased by 2.8±2.4%. The decrease in photon dose and increase in neutrondose when the chamber is placed in the phantom supports earlier measurement resultsand can partly explain the uncertainty in the measurement method. The effect tothe dose by perturbation of the charged particle field will be the subject of the furtherinvestigation.49

Eleventh World Congress on Neutron Capture TherapyWednesday, October 13, 2004 - 2:00 PMPosters Session - C l i n i ca l A p p l i ca t i o n s1 - Development of neutron capture therapy and otherneutron techniques in Obninsk, Russia. I. Gulidov a , A. Sysoeva , Yu. Mardynsky a , S. Ulianenko a , V. Kononov b , A. Glotov b , O. Kononov ba Medical Radiological Research Center RAMS, Obninsk, Russia;, b Institute forPhysics And Power Engineering, Obninsk, RussiaIntroduction: During long time neutron techniques have developed in Obninsk, Russia.At reactor BR-10(IPPE) patients from MRRC RAMS were treated with FNT and NCTenhanced FNT. Reactor BR-10 was closed in 2002 for dismantling. Now in the processof creation medical blocks at reactor WWRc (ICP) and accelerator KG-2,5 (IPPE).Materials and Methods: From 1985 up to 2002 more than 450 patients with differentmalignancies have treated in MRRC RAMS with fast reactor neutrons. Most of themhad locally advanced or recurrent tumors. The source of fast neutrons is reactor BR-10(IPPE) with mean energy of neutron beam about 1 MeV. Mixed photon-neutron techniqueswith contribution of fast reactor neutrons in total dose 20-40% are developedand are evaluated. Results obtained in patients with head and neck tumors are mostimportant from point of view of future NCT investigations in patients with brain tumors.Long time results are analyzed in 133 patients with squamous cell carcinoma of headand neck up to now.Results and Discussion: Developed techniques of mixed photon-neutron therapy arewell tolerated by surrounding tumor normal tissues. In the same time these techniquesare effective. Only in 27% of patients with laryngeal carcinoma no tumor responsewas registered after first stage of radiation therapy. Five-year actuarial local controlrate in patients with primary laryngeal carcinoma after radical course of mixed photon-neutrontherapy is 63% and 73% after combined therapy (mixed photon-neutrontherapy at first stage and, then, radical surgery). Five-year actuarial local control ratein patients with primary carcinoma of oral cavity and oropharynx is 51%. Completeresponse rate after mixed photon-neutron therapy is 57% in patients with recurrencesof carcinoma of oral cavity and oropharynx. One-year actuarial overall disease-specificsurvival rate in patients with recurrent tumors of oral cavity and oropharynx is 82%.The evaluation of all clinical experience and radiobiology studies of MRRC RAMS,technical data of IPPE and data of literature gives possibility to recommend combinationof NCT with photon therapy. Combination of BNCT with photons and in perspectivewith other kinds of irradiation (for example, protons) is promising and must bethoroughly investigated. Obninsk has good perspectives for creation of medical facilitiesfor NCT and FNT. Project of medical block at reactor WWRc (ICP) was finishedlast year. We hope that obtained in cooperation with IPPE scientists experience givesus the possibility to receive specialized medical facility for NCT and FNT on the basisof accelerator KG-2,5 this year.3 - Development of a Multimodal Monte Carlo RadiotherapyPlanning System. W.S. Kiger, III a A.G. Hochberg a,b , J.R. Albrittonc , and X-Q. Lu aa Department of Radiation Oncology, Beth Israel Deaconess Medical Center, HarvardMedical School, 330 Brookline Avenue, Boston, MA 02215, USA, b Institut Nationaldes Sciences et Techniques Nucléaires (Commissariat à l’Energie Atomique), Paris,France, c Nuclear Engineering Department, Massachusetts Institute of Technology,77 Massachusetts Avenue, Cambridge, MA 02139, USAA new Monte Carlo based treatment planning system is under development at Harvard-MIT.It is envisioned that this planning system, known as the MultiModal MonteCarlo (MiMMC, pronounced as mimic) planning system, will be a research tool forradiotherapy planning capable of simulating multiple particles and modalities, e.g.,external beam photon radiotherapy and Neutron Capture Therapy (NCT). This developmentbuilds on prior experience in NCT treatment planning with MacNCTPlan andNCTPlan as well as experience in clinical radiotherapy planning. MiMMC is writtenusing MATLAB (The Mathworks, Natick, MA) because of its significant advantages asa high level mathematics programming language, its graphics capabilities, and easeof development.MiMMC can import 8 or 16 bit DICOM or TIFF format CT or MR images for definitionof the patient geometry and has an adjustable window and level. MiMMC uses a voxelgeometry to describe the patient anatomy for radiation transport simulations with a uniformbut adjustable voxel size. Voxel models are generated by combining thresholdedimage data and over-riding regions of interest contoured by the user. Selection ofbeam orientations employs a simple and straightforward user interface with provisionsfor dealing with and tracking multiple fields. Initially, the MCNP Monte Carlo radiationtransport code is used for dose calculations because of its many advantages, includingvery flexible general source definition and its capabilities to record and later read andtransport particle tracks, i.e., surface source files. In MiMMC, dose can be prescribedto a fixed point or to a dose-volume parameter, e.g, the maximum, minimum, or meandose within a volume. Dose distributions can be evaluated using simultaneous transverse,sagittal, and coronal views of isodose contours overlaid on the image data anddose volume histograms for multiple regions of interest. Multi-field plan optimizationwith dose constraints and prescriptions based on DVH parameters will be incorporatedinto MiMMC. A preliminary version of this optimization algorithm for selecting the fieldweights has been used in planning one NCT patient treated at Harvard-MIT.Once the initial development is complete, MiMMC will be open source software; weexpect it to be easily extensible and adaptable to other users’ needs and researchinterests.5 - The pattern of tumor recurrence following BPA-mediatedBNCT of patients with glioblastoma. Leena Kankaanrantaa , Tiina Seppälä b , Eero Salli c , Riitta Mäntylä c , Sauli Savolainen c , MauriKouri a , Iiro Auterinen d , Markus Färkkilä e , Heikki Joensuu aa Department of Oncology, HUCH, Finland, b Boneca Corporation, Helsinki, Finland,c Helsinki Medical Imaging Center, University of Helsinki, d VTT Processes, VTT,Finland, e Department of Neurology, HUCH, FinlandTwenty-two previously untreated patients diagnosed with glioblastoma multiformewere entered in the Finnish BNCT trial P-01 between May 1999 and December 2003.Boronophenylalanine (BPA) 290 to 500 mg/kg complexed with fructose was given as a2-hour i.v. infusion as the boron carrier. Twelve first patients received BPA 290mg/kg,and the dose was escalated in 10 subsequent patients from 330 mg/kg to 450 mg/kg.During BNCT the whole blood boron concentration was on average 14.8 ± 2.8 ppm(SD), and the average weighted tumor boron dose was estimated to be 90% of thetotal weighted dose. Epithermal neutrons were delivered via 2 portals using 11 or 14cm circular apertures at the dedicated FiR 1 BNCT facility. Treatment plans were computedusing the BNCT_Rtpe treatment planning system (INEEL/MSU, Idaho/Montana,USA) based on gadolinium enhanced axial T1-weighted MRI scans. The patients wereregularly followed up after BNCT using MRI.An automatic image registration and image resampling were used to transform the T1-weighted MR volumes of the recurrent tumors to the coordinates of the pretreatmentMRIs taken for BNCT treatment planning. The dose given with BNCT to the site oftumor recurrence was estimated using the isodose curves of the treatment plan. Thedistance of the recurrent tumor from the skin and the gross primary tumor were measured.Sixteen of the 22 treatment plans could be evaluated for the present study.The average distance of the center of the recurred tumor from the skin was 4.2 ± 1.7cm (SD), and the distance from the closest edge of the pretreatment macroscopicprimary tumor was only 0.2 ± 1.3 cm (range, -1.7 cm to 3.8 cm). The center of therecurred tumor was located on average 1.5 ± 1.0 cm inside of the outer edge of thePTV, which included the gross tumor and the surrounding edema plus a 1 to 2 cmmargin. The estimated total gross tumor dose delivered to the recurred tumor was48.6 ± 15.7 Gy (W) when 3.5 times more boron-10 was assumed be present at thesite of the tumor recurrence than in the whole blood. However, if the boron concentrationin glioblastoma satellite cells is assumed to be only one half of that found inmacroscopic glioblastoma as suggested in the literature, the dose given to the site ofthe tumor recurrence was 26.7 ± 8.3 Gy (W). The calculated average total weightednormal tissue dose at the site of tumor recurrence was 9.4 ± 2.6 Gy (W), when anequal boron-10 concentration was assumed to be present in the normal brain tissueas in the whole blood.We conclude that the pattern of tumor recurrence following BPA-mediated BNCT isgenerally similar as following conventional external radiation therapy. Most glioblastomasrecur after BNCT at the margin of the gross tumor, where the most viable glioblastomacells are likely to reside. Further developments should include improvementof BPA uptake in the tumor and the satellite tumor cells, and improved methods tomeasure BPA uptake in vivo.50

51Session Chairs: Amanda Schwint, Subash Chandra8:30 AM -Neutron Capture Therapy of Epidermal Growth FactorReceptor Positive Gliomas Using Boronated Cetuximab(IMC-C225) as a Delivery Agent. Rolf F. Barth a, *, Gong Wu a ,Weilian Yang a , Peter J. Binns b , Kent J. Riley b , Hemant Patel d , Jeffrey A.Coderre c , Werner Tjarks f , A.K. Bandyopadhyaya f , B.T.S. Thirumamagal f ,Michael J. Ciesielski e , Robert A. Fenstermaker ea Department of Pathology, The Ohio State University, 165 Hamilton Hall, 1645 NeilAvenue, Columbus, OH 43210, USA, b Nuclear Reactor Laboratory, MassachusettsInstitute of Technology, Cambridge, MA, USA, c Department of Nuclear Engineering,Massachusetts Institute of Technology, Cambridge, MA, USA, d Department ofRadiology, Beth Israel Deaconess Hospital, Boston, MA 02215, USA, e Departmentof Neurosurgery, Roswell Park Cancer Institute, Buffalo, NY 14263, USA, f College ofPharmacy, The Ohio State University, 165 Hamilton Hall, 1645 Neil Avenue, Columbus,OH 43210, USACetuximab (IMC-C225) is a monoclonal antibody directed against both the wildtypeand mutant vIII isoform of the epidermal growth factor receptor (EGFR). The purposeof the present study was to evaluate the boronated monoclonal antibody (MoAb), cetuximab,as a boron delivery agent for neutron capture therapy (NCT) of brain tumors.Twenty-four hours following intratumoral (i.t.) administration of boronated cetuximab(C225-G5-B1100), the mean boron concentration in rats bearing either F98EGFR orF98WT gliomas were 92.3±23.3 µg/g and 36.5±18.8 µg, respectively. In contrast, theuptake of boronated dendrimer (G5-B1000) was 6.7±3.6 µg/g. Based on its favorablein vivo uptake, C225-G5-B1100 was evaluated as a delivery agent for BNCT inF98EGFR glioma bearing rats. The mean survival time (MST) of rats that receivedC225-G5-B1100, administered by convection enhanced delivery (CED), was 45 ± 3d compared to 25 ± 3 d for untreated control animals. A further enhancement in MSTto >59 d was obtained by administering C225-G5-B1100 in combination with i.v. boronophenylalanine(BPA). These data are the first to demonstrate the efficacy of aboronated MoAb for BNCT of an intracerebral (i.c.) glioma and are paradigmatic forfuture studies using a combination of boronated MoAbs and low molecular weightdelivery agents.9:10 AM -Radiobiology of BNCT mediated by GB-10 and GB-10+BPA in experimental oral cancer. Verónica A. Trivillin a ,Elisa M. Heber a , Maria E. Itoiz a,b , David Nigg c , Osvaldo Calzetta d , HermanBlaumann d , Juan Longhino d , Amanda E. Schwint a, *a Department of Radiobiology, Constituyentes Atomic Center, National Atomic EnergyCommission, Avenida General Paz 1499, (1650) San Martin, Provincia de BuenosAires, Argentina, b Department of Oral Pathology, Faculty of Dentistry, University ofBuenos Aires, Argentina, c Idaho National Engineering and Environmental Laboratory,Idaho Falls, USA, d Department of Nuclear Engineering, Bariloche Atomic Center,National Atomic Energy Commission, Provincia de Rio Negro, ArgentinaIntroduction: We previously reported the first evidence of the efficacy of BPA-BNCTfor the treatment of oral cancer in the hamster cheek pouch model with no damage tonormal tissue. We showed that GB-10 is able to deliver therapeutically useful amountsof boron to hamster cheek pouch tumors, albeit not selectively. We provided evidenceof the homogeneous deposition of GB-10 in different tumor areas, an asset in termsof targeting capacity in heterogeneous tumors. In addition, we reported biodistributiondata that would suggest a potential therapeutic advantage for the combined administrationof GB-10 and BPA. The aim of the present study was to assess, for the firsttime, the response of hamster cheek pouch tumors, precancerous tissue and normaltissue to BNCT mediated by GB-10 and BNCT mediated by GB-10 and BPA administeredjointly using the thermalized epithermal beam of the RA-6 Reactor at the BarilocheAtomic Center.Materials & Methods: Fifteen animals bearing a total of 57 tumors and ten normalhamsters were irradiated for 49.5 minutes, 3 hours after administration of GB-10 (50mg 10B/kg b.w.). Fifteen animals, bearing a total of 54 tumors and 10 normal animalswere irradiated for 17.3 minutes after the combined administration of GB-10 (34.5mg 10B/kg b.w.) and BPA (infusion: total dose 31 mg 10B/kg b.w.). The animals wereirradiated 3 hours post-administration of GB-10 and 1.5 hours after the last injectionof BPA. Five tumor-bearing hamsters and 5 normal hamsters were irradiated with thebeam alone. The average flux of thermal neutrons was 3.4 ± 0.3 x 10Exp08 neutrons/sq.cm.sec. The estimated total physical doses to tumor were 5.28 (S.D.: 0.41) Gy forGB-10-BNCT and 4.26 (S.D.: 0.39) Gy for (GB-10 + BPA)-BNCT. Tumor, precanceroustissue and normal tissue response was assessed by visual inspection, a tumor volumeassay and histological analysis at 1, 7, 14, 21 and 30 days post-irradiation.Results: GB-10-BNCT exerted 75.5% tumor control (partial + complete remission) withno damage to precancerous tissue around tumor or to normal tissue. (GB-10 + BPA)-Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - AMPre-Clinical & TranslationalBNCT elicited 64.8% tumor control at a lower physical dose to tumor with no radiotoxiceffects. BNCT mediated by (GB-10 + BPA) resulted in a marked reduction in the doseto normal tissue and would thus allow for significant dose escalation. Light microscopyevaluation revealed that GB-10-BNCT induced blood vessel damage and areas ofnecrosis in tumor tissue but failed to induce alterations in precancerous or normalepithelia or in the underlying blood vessels.Discussion and Conclusions: GB-10 is a therapeutically useful “stand-alone” boronagent for BNCT of oral cancer. Uptake in precancerous tissue may allow for the treatmentof field cancerized areas around tumor. Although GB-10 does not target tumorcells selectively it leads to selective tumor lethality while sparing normal tissue, possiblyvia a selective effect on tumor blood vessels. (GB-10 + BPA)-BNCT may providea therapeutic advantage if dose is prescribed to normal tissue tolerance.9:30 AM -The use of the Biological Equivalent Dose (BED) conceptto assess mixed Low and High LET Radiations withParticular Reference to BNCT. B.Jones a , J Townley a , R Dale b ,J Hopewell c , and S Green aa Queen Elizabeth University Hospital, Birmingham, B15 2TH, UK, b RadiotherapyPhysics & Radiobiology, Charing Cross Hospital, London W6 8RF, c Department ofClinical Oncology, Churchill Hospital, Oxford, OX3 7LJThe radiobiology associated with BNCT is more complex than for low LET X-rays(XRT). Most experimental and clinical BNCT studies have used only single exposures;Coderre and Morris [1] have warned that fractionated BNCT will require special radiobiologicalconsiderations. To develop this suggestion further, BED equations appropriatefor high LET radiations have been applied to BNCT. The RBEmax concept (theratio of high to low LET a values) is used [2] to estimate a the high LET BED. In thisway, the low LET(a/b) value can be used [3]. The BED equations can be extended toinclude multiple high LET components (e.g. in BNCT), on the assumption that thereare no high-low LET interactions. If Compound Biological Effect (CBE) is known, amore complex equation can be obtained which replaces the X-ray total dose in isoeffectiveconditions where the overall BED of BNCT is the same as that for megavoltageX-rays. The advantages of this approach include:1. Established low LET a/b ratios (e.g. a/b=2 Gy for brain necrosis) for early and latereacting tissues; tumour repopulation dose-equivalent values can also be used [3].2. Treatment fractionation effects can be assessed.3. For protracted exposures, low dose-rate corrections can also be incorporated tothe low LET part of the BNCT treatment.For example, for a total physical dose of 15 Gy given in one, two and three fractionson the Birmingham epithermal neutron beam (depth = 5.5 cm), the provisional X-RTequivalent iso-effective doses (given in 2 Gy fractions) for late effects in normal brainwould be 45.8, 49.8 and 61.7Gy, respectively. This assumes a CBE of 1.3 and a beamRBE 1.5. When the dose is given as 2 fractions, each of 3.5 hr, at a dose-rate of 2.5Gy hr-1, the X-RT equivalent iso-effective dose would be 58.2 Gy, based on an assumptionthat repair of sub-lethal DNA damage is mono-exponential (T1/2 of 3 hours).There is considerable scope for radiobiological modelling in BNCT in order to optimisetherapy, to compare BNCT alone with XRT, and in the assessment of BNCT used asa boost with XRT.9:50 AM -SIMS Ion Microscopy Imaging of Boronophenylalanineand 13C15N-Labeled Phenylalanine in Human GlioblastomaCells: A Comparison of Subcellular Distributionand Cellular Uptake at 2 and 6 Hours. S. Chandra*, D.R.Lorey II, and D.R. SmithCornell SIMS Ion Microscopy Laboratory, Department of Earth and Atmospheric Sciences,Snee Hall, Cornell University, Ithaca, NY 14853The boron-10 analogue of the amino acid L-phenylalanine (p-boronophenylalanine,BPA) is a clinically approved Boron Neutron Capture Therapy (BNCT) agent for braintumors and malignant melanomas. Recent Swedish clinical trials on glioblastoma multiformeused a 6 hr BPA infusion protocol rather than 2 hr. This protocol was supportedin part by subcellular secondary ion mass spectrometry (SIMS) imaging studies onanimal and cell culture models of human glioblastoma multiforme (1, 2). These studiesindicated that exposure of T98G human glioblastoma cells and rat models to BPA for 6hr resulted in higher uptake of 10B in cells and tumor tissues, including the infiltratingtumor cells in the normal brain, as compared to 2 hr. The present study was designedto understand the reasoning behind this increased 10B uptake in T98G cells. Cellswere exposed to 110 ppm boron equivalent of BPA or the identical concentration of

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - AMPre-Clinical & Translational13C15N isotopically labeled L-phenylalanine for 2 and 6 hrs. After cryogenic samplepreparation, the cells were analyzed with the SIMS ion microscope for 10B from BPAand the 28(13C15N) label from 13C15N-labeled phenylalanine. Images of 39K, 23Na,and 40Ca were also recorded to verify that SIMS measurements were made in healthy,well-preserved cells. T98G cells reveal well-characterized subcellular compartments:a nucleus or multiple nuclei, a mitochondria-rich perinuclear region, and the remainingcytoplasm (3). Overall cellular signals of 10B increased approximately 1.6-foldbetween the 2 and 6 hr BPA exposures. Similarly, the 28CN images of individual cellsindicated a comparable increase in the net entry of 13C15N-labeled phenylalanine at6 hr in comparison to 2 hr. However, the subcellular distribution of 10B from BPA wasdistinctly different from the 28CN distribution from labeled phenylalanine in the mitochondria-richperinuclear cytoplasmic region. The 10B from BPA was depleted in thisregion, but 28CN was distributed throughout the cell without any noticeable gradientin the mitochondria-rich perinuclear region. These observations indicate that (i) thehigher uptake of BPA and 13C15N-labeled phenylalanine at 6 hr vs. 2 hr plausibly representsa basic similarity between BPA and phenylalanine in a time-dependant entrymechanism through the plasma membrane in response to cellular requirements for theamino acid and (ii) intracellular processes recognize BPA as a different molecule thanphenylalanine and the metabolism of BPA is distinctly different from that of phenylalanine.(Supported by the U.S. Department of Energy).References:1. Smith, D.R., Chandra, S., Barth, R.F., Yang, W., Joel, D.D., and Coderre, J.A. Cancer Res. 61:8179-8187 (2001)2. Chandra, S., Lorey II, D.R., and Smith, D.R. Radiation Res. 157: 700-710 (2002)3. Chandra, S., Kabalka, G.W., Lorey II, D.R., Smith, D.R., and Coderre, J.A. Clin. Cancer Res. 8:2675-2683 (2002).10:30 AM -User Center for Neutron Capture Therapy Research byOtto Harling. O.K. Harling a, *, K.J. Riley b , P.J. Binns b , H. Patel c , J. A.Coderre aa Department of Nuclear Engineering, Massachusetts Institute of Technology, Cambridge,MA 02139, USA, b Nuclear Reactor Laboratory, Massachusetts Institute ofTechnology, Cambridge, MA 02139, USA, c Beth Israel Deaconess Medical Center,330 Brookline Avenue, Boston, MA 02215, USAThe Massachusetts Institute of Technology (MIT) with funding from the US Departmentof Energy has established a user center for neutron capture therapy research.The user center comprises a variety of irradiation facilities, laboratories and analyticaltools in addition to the scientific support and institutional oversight that is provided forpre-clinical and clinical NCT research. The irradiation facilities for experimental andclinical use are housed in the multipurpose MITR-II Research Reactor which schedulescontinuous operation at 5 MW for approximately 300 days per year. High intensity,high purity thermal and epithermal neutron beams are equipped with automated controlsystems to precisely deliver the desired dose and variable collimators that can bechanged as needed for experiments. Gross boron concentration is assayed by eitherprompt gamma neutron activation analysis (PGNAA) or inductively coupled plasmaatomic emission spectroscopy (ICP-AES). Dedicated resources and procedures arebeing developed to offer high-resolution quantitative auto-radiography (HRQAR),which is useful for determining the cellular 10B micro-distribution within samples oftumor or normal tissue. General purpose and cell culture laboratories are availablein adjacent buildings as well as accredited animal care and surgical facilities that arecentrally located on the MIT campus. These facilities are complemented by in-housescientific support that is available for planning, execution and data analysis of theexperiments or clinical program. Expertise within the responsible MIT review boardsand the MIT NCT research staff can help to facilitate the institutional approvals that areoften needed by both centers for conducting research. Within the currently availableresources the facilities can be made available to qualified researchers without incurringcharges that would normally be applied. The level of support in the form of MITstaff and services is limited, however MIT will make every effort to meet the needs ofall interested researchers.10:50 AM -Results of the European research project: therapeuticstrategies for boron neutron capture therapy: boron imaging.A. Wittig a , J. Michel b , K. Zierold c , H. Arlinghaus d , R. Moss e , K. Appelmannf , M. Malago g , S. Grabbe h , P. Therasse i , W. Sauerwein aa Department of Radiation Oncology, Universitätsklinikum Essen, Essen, Germany,b INSERM ERM 0203 Université de Reims Champagne-Ardenne, Reims, France,c Max Planck Institute for Molecular Physiology, Dortmund, Germany, d PhysikalischesInstitut, Universität Münster, Münster, Germany, e Joint Research Centre, Institutefor Energy, Petten, Netherlands, f Nuclear Research and consultancy Group (NRG),Petten, Netherlands, g Dept. of Surgery, University Hospital Essen, Essen, Germany,h Dept. of Dermatology University Hospital Essen, Essen, Germany, i EORTC DataCenter, Bruxelles, BelgiumIntroduction: Improvement of anti-cancer therapies, therapeutic and diagnostic strategiesrely on the ability to selectively deliver compounds to target cells. Imaging of suchtargeting compounds and knowledge of their distribution as well as that of intrinsic elementsand molecules in cells is of great interest. The aim of this project is to developtools for imaging and quantifying of atoms and molecules within tissues and singlecells.Materials and Methods: Electron energy loss spectroscopy (EELS), time of flight secondaryion mass spectrometry (TOF-SIMS) and laser secondary neutral mass spectrometry(Laser-SNMS) - each capable of simultaneously detecting all masses withvery high sensitivity and sub-cellular resolution - were developed to be used for imagingand quantifying atomic species in biological samples. Prompt gamma ray spectroscopywas used to assess the boron concentration in macroscopical volumes. Alltissues were characterised histopathologically.Results: For analysis of cell cultures and tissues dedicated cryo-fixation methods aremandatory. With TOF-SIMS and Laser-SNMS high-resolution elemental images andmass spectra were obtained in cells and tissues. The measurement of the Na/K ratio,which had to be similar to the one of living cells was used to select cells for detailed investigation.Incubation of melanoma cells with the boronated compound BSH resultedin a boron signal inside the cells, which was lower than the one outside cells in theculture medium. The results with EELS give evidence of an inhomogeneous spot-likedistribution of boron inside single cells.Discussion: Successful sample preparation preserving the chemical integrity of cellshas been demonstrated. Imaging of boron inside single cells was possible with the 3dedicated methods used, thus demonstrating their suitability for imaging elements inbiological material. Collection and storage of samples from patients in clinical trials arespecial aspects of translational research. The application of a boronated drug prior tosurgery has no individual benefit for the patient. Successful handling of the relatedethical, legal and regulatory aspects need the support of a dedicated organisationsuch as the EORTC. The realisation of such research requires an especially closeinterdisciplinary collaboration between clinicians and basic scientists.11:10 AM -Neutron Capture Therapy for Liver Cancer Metastases.Tazio PinelliI.N.F.N. and University of Pavia, ItalyAn original research on the therapy of the human organs affected by incurable diffusedmetastases is resumed. The project is based on the idea to treat the removed organ ina thermal neutron field in such a way to realize the integral irradiation of the metastaticsystem including the undetected nodules of very small dimension. Of course this procedurerequires a necessary and strong support for auto-transplant surgery .In order to carry out a feasibility study on the project, the case of the colon cancerspread to the liver only was first taken into consideration.This particular radiotherapy meets the crucial problem of preserving the normal hepaticparenchyma while adequate doses are given to the tumor tissues. The solutionwas reached by setting up an original application of the BNCT (boron neutron capturetherapy).In the following the method to assess the procedure is detailed: experimental study ofboron uptake, realization of a neutron field suitable for the organ irradiation, evaluationof doses to give the organ, methods and instruments to control any parameter of thetreatment.Results and conclusions, achieved during 15 years of research activity, prove the possibilityto set up a treatment fit to face the above dramatic pathology.In conclusion the first human procedures and the starting clinical trials organized togetherwith Italian Ministry of health, will be discussed.11:30 AM -BNCT dose distribution in liver with epithermal D–D andD–T fusion-based neutron beams. H. Koivunoro a, *, D.L. Bleuel a ,U. Nastasi b , T.P. Lou a , J. Reijonen a , K-N. Leung aa Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Mail Stop 5R0121,Berkeley, CA 94720, USA, b Ospedale San Giovanni A.S., Via Cavour 31, 10100Torino, ItalyRecently, a new application of Boron Neutron Capture Therapy (BNCT) treatment hasbeen introduced. Results have indicated that liver tumors can be treated by BNCTafter rejection of the liver from the body. Because of the complexity of the surgicalremoval of the liver and its high risks of complications, in vivo liver BNCT, withoutremoving the liver from body, may be preferable, if a low dose to healthy tissue and52

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - AMPre-Clinical & Translationalsufficient tumor dose all over the liver could be assured. In conventional radiotherapy,the tolerance dose (TD(5/5)) for liver has found to be 30Gy in whole liver irradiations.BNCT could be considered a favorable treatment modality, if a tumor dose >30Gy(W)can be assured in the liver tumor.At Lawrence Berkeley National Laboratory, compact neutron generators based on2H(d,n)3He (D-D) or 3H(t,n)4He (D-T) fusion reactions are being developed. Preliminarysimulations of the applicability of 2.45MeV D-D fusion and 14.1MeV D-T fusionneutrons for in vivo liver tumor BNCT, without removing the liver from the body, havebeen carried out.MCNP simulations were performed in order to find a moderator configuration for creatinga neutron beam of optimal neutron energy and to create source model for dose calculationswith the SERA (Simulation Environment for Radiotherapy Applications) treatmentplanning program. SERA dose calculations were performed in a patient modelbased on CT scans of the body. The BNCT dose distribution in liver and surroundinghealthy organs was calculated with rectangular beam aperture of sizes of 20cm*20cmand 25cm*25cm. Collimator thicknesses of 10 and 15 cm were used. Dose calculationswere performed using single beams and combination of three beams of samesize. The maximum healthy liver dose was limited to 12.5Gy(W). The dose for neutronsand photons was plotted in the beam centerline and in addition, the isodosecontours were printed over the CT images and dose volumes in liver were calculatedfor tumor dose and healthy tissue dose.When the collimator length was increased from 10cm to 15cm, the treatment time wasincreased by 70-110%, but the maximum tumor dose was increased only by 10%-15%with the D-D source and 1-3 % with the D-T source. With the three beam combinations,tumor dose greater than 30Gy(W) was delivered only to maximum of 58% ofliver volume with either the D-D or the D-T beams when healthy liver dose was limitedto 12.5 Gy (W).Overall, treatment times with D-D neutron source of neutron yield 1012 n/s were notsensible (all > 58 hours). D-D neutron yield should be increased ~30-60 times to obtainrealistic treatment time of 1-2 hours. With the used D-T neutron source yield, treatmenttimes were acceptable (28-72 minutes with single beam and 63-128 minutes with 3beams of different direction). However, tumor dose of 30 Gy(W) can be achieved inthe whole liver area with these beams only, when the maximum healhty tissue dose (to~12% of total liver volume) is increases from 12.5 Gy(W) to 36 Gy(W). In this case theaverage tissue dose to liver would be 21Gy(W).53

54Session Chairs: Martti Kulvik, Andrea WittigEleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 5 - B i o l o g y & C h e m i s t r y2:00 PM -Selective irradiation of the blood vessels by using BNCR® - development of the method and its effects on tumorand normal tissues. Koji Ono a , Shin-ichiro Masunaga a , Yuko Kinashia , Minoru Suzuki a , Kennji Nagata a , Yoshinori Sakurai a , Akira Maruhashia , Satoshi Kasaoka b , Kazuo Maruyama ba Kyoto Univ. Research Reactor Institute, Osaka 590-0494, Japan, b Teikyo Univ.Faculty of Pharmacy, Kanagawa 199-0195, JapanIntroduction: Tumor blood vessels are immature, and then fragile to anti-tumor treatment.If any B-10 compound stays in the blood vessel and receives thermal neutrons,the emitted 4He and 7Li nucleuses destroy this fragile structure from the inside. Wehave examined the effects of BSH-containing large size liposome (=300nm) PEGliposome(BSH).This type liposome escapes from phagocytosis by macrophage.Materials & Methods: C3H/He mice and SCCVII tumors were used for the study. Theeffects on tumors and normal tissues were examined by in vivo ®¢ in vitro colonyformation assay, tumor growth delay assay, survival assay of mice. All experimentswere performed according to the rule decided by Kyoto University Experimental AnimalCommittee.Results: The B-10 concentration ratio between blood and tumor 30 minutes afterPEG®¢liposome(BSH) administration was 35-40. The surviving fraction of tumor cellsin the PEG-liposome(BSH) neutron group was slightly further suppressed in comparisonwith that of neutron irradiation alone, however, the tumor growth was remarkablysuppressed. No reactions of the skin covering tumors were observed. The mousegroup of 28-36 Gy to the systemic endothelial cells of blood vessel did exhibit no death,and in the groups of 73-91and 87-109 Gy, all mice died. No diarrhea and bloody analdischarge did appeared at all. Cause of the death seemed bone marrow damage.Discussion: BNCT intends to kill tumor cells selectively. Three conditions are necessaryfor this treatment to be successful. The first is selective accumulation of boroncompound in tumor cells. The second is amount of B-10 per gram tissue of the tumors.The third is uniform accumulation of B-10 compound through all tumor cells. Thesethree conditions generally conflict with each other. BPA accumulates in tumor cells athigh selectivity but the extent of accumulation in quiescent tumor cells is very low. BSHis not taken in tumor cells. Based upon the above consideration, we have devised anidea of tumor blood vessel targeted therapy by B-10 neutron capture reaction. Thetumor growth was significantly suppressed without any radiation damage on tumorcoveringskin. The effects on whole body are also mild. In BNCT, neutron is deliveredto limited volume of the body involving tumors. Accordingly, significant therapeutic gainseems to be expected by the combination of tumor blood vessel targeted BNCT andordinary BNCT. We are now planning to examine the effects of combined BNCT onexperimental tumors and normal tissues.Conclusions: By combined use of neutron irradiation and large size PEGliposome(BSH),the blood vessels can be irradiated selectively. Selective irradiation oftumor vessels suppressed tumor growth effectively, but it does not give acute effectson normal tissue if neutron field is limited.2:20 PM -Targeting liposomes to tumor endothelial cells for neutroncapture therapy. Gerben A. Koning a,b, *, Marjan M. Fretz b ,Urszula Woroniecka a , Gert Storm b , Gerard C. Krijger aa Department of Radiochemistry, Interfaculty Reactor Institute, Delft University ofTechnology, Mekelweg 15, NL-2629JB Delft, The Netherlands, b Department of Pharmaceutics,Utrecht University, PO Box 80082, NL-3508 TB Utrecht, The NetherlandsThe growth of solid tumors strongly depends on the in growth of new blood vesselsin the process of angiogenesis. Inhibition of angiogenesis has proven a powerful approachfor the inhibition of tumor growth. Recent studies point towards an importantrole of endothelial cell apoptosis on the response of the total tumor on radiotherapy,indicating that tumor endothelial cells are sensitive to radiation induced damage.The aim of our work is to target 10B to the tumor vasculature for neutron capturetherapy (NCT) causing inhibition of the growth of angiogenic endothelial cells, damageto tumor blood vessels and ultimately tumor regression. For this purpose, we are usingliposomal targeting devices to improve tumor specificity and to deposit high quantitiesof 10B into endothelial cells of tumor vasculature.RGD-peptides, specific for alpha (v) integrins expressed on angiogenic endothelialcells, were coupled to liposomes encapsulating dodecahydrododecaborate. RGD-liposomesstrongly associated with endothelial cells (HUVEC) and were internalized.Proliferating HUVEC proved sensitive to treatment with gamma-irradiation resulting indecreased cell viability pronounced inhibition of DNA-synthesis with increasing dose.Irradiation of 10B-RGD-liposome incubated HUVEC with neutrons strongly inhibitedendothelial cell viability. Recent in vivo studies proved that these liposomes are alsoable to efficiently bind to tumor blood vessels. These results suggest that efficient NCTcan be achieved by targeting 10B-liposomes to angiogenic endothelium in tumors.Future studies will address the in vivo efficacy of RGD-liposomes for neutron capturetherapy. Besides RGD-liposomes, targeting the alpha (v) integrins on endothelial cells,other receptors that are specifically expressed on angiogenic endothelium, such asVEGF receptors and endoglin as well as novel molecular markers identified by phagedisplay technology will be targeted. In these approaches the presented 10B-containingliposomes will serve as a platform for the attachment of the various targeting ligands.2:40 PM -Boronated epidermal growth factor as a delivery agentfor neutron capture therapy of EGF receptor positivegliomas. Weilian Yang a , Rolf F. Barth a, *, Gong Wu a , Achintya K.Bandyopadhyaya b , B.T.S. Thirumamagal b , Werner Tjarks b , Peter J. Binns c ,Kent Riley c , Hemant Patel e , Jeffrey A. Coderre d , Michael J. Ciesielski f , RobertA. Fenstermaker fa Department of Pathology, The Ohio State University, 165 Hamilton Hall, 1645 Neil Avenue,Columbus, OH 43210, USA, b College of Pharmacy, The Ohio State University,Columbus, OH 43210, USA, c Nuclear Reactor Laboratory, Massachusetts Instituteof Technology, Cambridge, MA 02139, USA, d Department of Nuclear Engineering,Massachusetts Institute of Technology, Cambridge, MA 02139, USA, e Department ofRadiology, Beth Israel-Deaconess Medical Center, Boston, MA 02215, USA, f Departmentof Neurosurgery, Roswell Park Cancer Institute, Buffalo, NY 14263, USAWe have been interested in targeting either the epidermal growth factor receptor(EGFR) or its mutant isoform EGFRvIII by means of either boronated EGF or monoclonalantibodies (MoAbs) for boron neutron capture therapy of brain tumors. Both ofthese receptors frequently are overexpressed in human glioblastomas, which makesthem attractive targets for the treatment of brain tumors. Cetuximab (Erbitux), previouslyknown as IMC-C225, is a chimeric MoAb directed against both wildtype EGFRand EGFRvIII. Cetuximab is particularly attractive as a boron delivery agent for neutroncapture therapy (NCT) of gliomas since it also can interfere with critical signaltransduction events by blocking the binding of EGF and TGF-a to EGFR.The purpose of the present study was to evaluate the boronated monoclonal antibody(MoAb), cetuximab, as a boron delivery agent for NCT of brain tumors. Twenty-fourhours following intratumoral (i.t.) administration of boronated cetuximab (C225-G5-B1100), the mean boron concentration in rats bearing either F98EGFR or F98WTgliomas were 92.3±23.3 µg/g and 36.5±18.8 µg, respectively. In contrast, the uptakeof boronated dendrimer (G5-B1000) was 6.7±3.6 µg/g. Based on its favorable in vivouptake, C225-G5-B1100 was evaluated as a delivery agent for BNCT in F98EGFRglioma bearing rats. The mean survival time (MST) of rats that received C225-G5-B1100, administered by convection enhanced delivery (CED), was 45 ± 3 d comparedto 25 ± 3 d for untreated control animals. A further enhancement in MST to >59 d wasobtained by administering C225-G5-B1100 in combination with i.v. boronophenylalanine(BPA). One possible explanation for this enhanced survival is that the F98EGFRglioma contained a heterogenous population of tumor cells, including EGFR positiveand negative cells, and that the BPA targeted the latter. Alternatively, the enhancedsurvival may have been due to the additional radiation dose contributed by 10B deliveredto both receptor (+) and (-) cells within the tumor.The combination of high molecular weight EGFR targeting agents, such as boronatedcetuximab or EGF, administered i.c. by CED, together with systemic administrationof BPA, could provide a way to target the heterogenous population of cells that arefound within gliomas. Studies to quantify tumor and normal brain 10B concentrationsfollowing CED of C225-G5-B1100 will be carried out in the near future. These datashould allow us to estimate the physical dose delivered to the tumor. However, sinceit is impossible with currently available technology to quantify in real time the in vivocellular uptake and chemical forms of the 10B, precise dosimetric calculations areimpossible. Further studies are planned to optimize treatment of EGFR(+) gliomas byusing combinations of agents to target various subpopulations of tumor cells. (Supportedby National Institutes of Health grant 1R01 CA098945-01 (R.F.B.) and UnitedStates Department of Energy grants DE-FG02 98ERG2595 (R.F.B.) and DE-FG-07-02ID14420 (to Dr. Otto Harling).3:00 PM -Boron biodistribution in Beagles after intravenous infusionof 4-dihydroxyborylphenylalanine–fructose complex.M.E. Kulvik a, *, J.K. Vähätalo b , J. Benczik c , M. Snellman c , J. Laakso d ,R. Hermans e , E. Järviluoma f , M. Rasilainen f , M. Färkkilä a , M.E. Kallio aa Department of Neurology, Helsinki University Central Hospital, PL 340, FIN-00029HUS, Finland, b Laboratory of Radiochemistry, University of Helsinki, PL 55, FIN-00014, Helsinki, Finland, c Faculty of Veterinary Medicine, University of Helsinki, PL

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 5 - B i o l o g y & C h e m i s t r y57, FIN-00014 Helsinki, Finland, d Institute of Biomedicine, Pharmacology, Universityof Helsinki, PL 63, FIN-00014 Helsinki, Finland, e ETLA, The Research Institute of theFinnish Economy, Lonnrotinkatu 4B, FIN-00120 Helsinki, Finland, f HUCH Pharmacy,Helsinki University Central Hospital, PL 200, FIN-00029 HYKS, FinlandAssessment of tumour boron levels is required for dosimetric modelling in BNCT.Whole blood concentrations are used as a surrogate for determining the in vivo tissueboron content. It is assumed that each of the various regions of interest has aneven average boron concentration. The irradiation time for BNCT is adjusted on thebasis of the pre- and per-irradiation whole blood boron concentrations, assuming amean boron concentration ratio of 1:1 for blood to healthy tissue and 1:3.5 for bloodto tumour tissue. This data is mainly derived from preclinical studies in tumour animalmodels. In order to verify the whole blood to normal tissue boron concentration ratioswe performed a boron biodistribution study on beagles.Boron biodistribution after intravenous infusion of 4-dihydroxyúborylphenylúalanineBPA–F complex was investigated in six dogs. Blood samples were evaluated duringand following doses of BPA of 205 and 250 mg/kgbw as a thirty minute infusion, and500 mg/kgbw as a one hour infusion. Samples from whole blood, urine, brain tissuesand other organs were analysed for their boron content after varying times followingthe end of the infusion.Whole blood boron concentrations ranged from 8.4 ppm to 27 ppm 9 min to 130 minafter the end of infusion ; the values showed a descending pattern with time. The boronconcentrations in different organs varied greatly, with the exception of brain and kidneytissue. These concentrations seemed to relate to the corresponding blood boronvalues. Kidneys showed high boron values, 2.5 to 5 times that of corresponding bloodvalues. However, boron concentrations in brain tissue as a function of time seemed tofollow a opposite path as compared to blood boron levels (Table 1).In order to elaborate on this finding, a Pearson correlation calculation was performedusing all tissues as variables from which we had full series of samples. Whole bloodboron concentrations showed a significant positive correlation with the boron concentrationsin lung, liver, and kidney.Blood boron concentrations and time after onset of infusion was negatively correlated(r=-0.86, p=0.03), but brain boron concentrations correlated positively with time afterthe onset of infusion (r=0.81, p=0.05). These trends can be clearly visualized in apresented blood boron plot. In order to try to identify possible distorting dependencies,we performed a principal component (PC) analysis including all collected variablesyielding same results as the correlation analysis. However, the confidence interval forour data can be relatively large as the number of dogs and samples is limited.The concentration in brain is of particular interest for brain tumor BNCT. The findings inthis study suggest that whole blood boron concentrations might not reflect accuratelythe boron concentration in dog brain tissue at respective time points; and the ratio betweenwhole blood and brain grey matter boron concentrations change with time. Thepossible implications of this finding on radiobiological studies using BPA and neutronirradiation go, however, beyond the scope of this paper.3:20 PM -T cell uptake for the use of boron neutron capture as animmunologic research tool. E. Binello a, *, R.N. Mitchell b , O.K.Harling ca Harvard-MIT Division of Health Sciences & Technology, Cambridge/Boston, MA,USA, b Department of Pathology, Brigham & Women’s Hospital, Boston, MA, USA,c Department of Nuclear Engineering, MIT, Cambridge, MA, USAIntroduction: Boron uptake by specific immune cells is critical to the development of thetechnique using the boron neutron capture reaction as an immunologic research tool.It is proposed in the context of heart transplantation research, in order to investigatethe temporal relationship between parenchymal rejection (representing immune cellinfiltration) and transplantation-associated arteriosclerosis (characterized by progressivevascular occlusion). The aim of this work was to evaluate the potential of selectedboronated nucleosides for this novel boron neutron capture application. Advantages ofboronated nucleosides are the reduced risk of leakage and effective dose delivery dueto their incorporation into the nuclear material of a cell.Materials and Methods: T cells were isolated from murine splenocytes and stimulatedwith an antibody to CD3 (anti-CD3). The boron compounds used were hydrophilicallyenhanced carboranyl thymidine analogs denoted as N4-2OH and N5-2OH. Uptake bycells was measured after a 12-hour incubation ending on days 2 and 3 post stimulation.Boron concentration of the supernatants was also measured. Finally, the effectof uptake on production of interferon-gamma (IFNg) was determined using enzymelinkedimmunosorbent assays.Results: Experiments indicate high uptake in vitro with comparatively minimal boronconcentrations in the supernatants. Specifically, the anti-CD3-stimulated T cell uptakeof N4-2OH ranged from approximately 75 ppm to 290 ppm over the course of the secondand third day post-stimulation while the uptake of N5-2OH ranged from approximately140 ppm to 250 ppm. The overall average concentration was approximately200 ppm for both compounds over both days. The concentration in the supernatantswas within 5% to 15% of the concentration measured in cells for both compounds overboth days. The concentration of IFNg present in the supernatants following addition ofmedium alone did not significantly change between days (p> 0.5) and was an average12,700 ± 1,700 pg/ml. Addition of either boronated nucleoside does not significantlyaffect IFNg production on any day (p> 0.5).Discussion: The particularly high uptake observed in the murine T cells may be attributableto the dramatic increase in proliferation following anti-CD3 stimulation. Thetendency for the uptake to be higher on day 3 compared to day 2 may be due to thetiming of the incubation, in that the day 3 pulse have included a fuller period of activeproliferation compared to day 2. The comparatively minimal supernatant boron concentrationsindicate retention against a gradient and are consistent with the degree ofapoptosis that normally occurs after stimulation. The lack of an effect on IFNg productionis particularly significant in this application where the goal is to have fully functionalT cells present for determined times in the graft prior to inactivation using the boronneutron capture reaction.Conclusions: High uptake may be achieved using carboranyl thymidine analogs designedto incorporate into cellular nuclear material. This, combined with the lack ofsignificant effects on T cell function, indicate that they are attractive compounds forthe use of the boron neutron capture as an immunologic research tool. Further developmentof this technique, which could have wide application in immunology, is warranted.3:55 PM -Evaluation of poly(ADP-ribose) polymerase-1 inhibitionon the induction of chromosomal aberrations and micronucleiby the boron neutron capture reaction. NunoG. Oliveira a , Matilde Castro a , António S. Rodrigues b , Isabel C. Gonçalves c ,José M. Toscano Rico d , José Rueff ea Faculty of Pharmacy, University of Lisbon, Lisbon, Portugal/Department of Genetics,Faculty of Medical Sciences, New University of Lisbon, b Department of Genetics,Faculty of Medical Sciences, New University of Lisbon/ University Lusófona, Lisbon,Portugal, c Nuclear and Technological Institute, Portuguese Research Reactor,Sacavém, Portugal., d Faculty of Medicine, University of Lisbon, Lisbon, Portugal,e Department of Genetics, Faculty of Medical Sciences, New University of LisbonThe boron neutron capture (BNC) reaction results from the interaction of boron-10 withlow energy thermal neutrons and gives rise to lithium-7 and alpha-particles. Theseparticles have a high linear energy transfer (LET) and a short range being highly damagingto cells. Accordingly, the BNC reaction has been used in clinical trials - boronneutron capture therapy (BNCT) - to treat some aggressive types of cancer, namelymalignant melanoma and high grade glioma. Mechanistic knowledge on DNA and celldamage induced by alpha-particles remains limited. It is well known that high-LET radiationinduces both DNA single and double strand breaks, being the latter frequentlyassociated with cell death. The repair of these DNA lesions and especially doublestrand breaks are thus fundamental for the understanding of high-LET radiation effects.PARP-1 is considered to be a constitutive factor of the DNA damage surveillancenetwork present in eukaryotic cells acting through a DNA break sensor function.The aim of this work was the evaluation of PARP-1 inhibition on the genotoxic effectsof the boron neutron capture reaction in V79 Chinese hamster cells using chromosomalaberrations and micronuclei as the end- points. The BNC reaction was carriedout at the vertical access of the thermal column of the Portuguese Research Reactor(RPI) using different concentrations of p-borono-L-phenylalanine (BPA) and differentperiods of irradiation with low energy thermal neutrons. A clear dose response in thefrequencies of aberrant cells excluding gaps (%ACEG) and chromosomal aberrationsexcluding gaps per cell (CAEG/cell) induced by the BNC reaction was observed fordifferent BPA concentrations and thermal neutrons fluences. The pattern of chromosomalaberrations induced by the BNC reaction included not only breaks and dicentricchromosomes and rings but also an important number of chromosomal rearrangements(e.g. triradial, quadriradial, complex rearrangements) which appear to be typicalfeatures of high LET radiation and formed probably due to the erroneous rejoiningof double strand breaks. However, the presence of different concentrations of theclassical PARP inhibitor 3-aminobenzamide (3-AB) did not lead to an increase in the%ACEG or CAEG/cell, neither did it changed the pattern of the induced chromosomalaberrations. In addition, two novel and very potent PARP-1 inhibitors were studied, 5-aminoisoquinolinone (5-AIQ) and PJ-34. Results with these compounds agreed withthe results obtained with 3-AB.55

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 5 - B i o l o g y & C h e m i s t r yUsing the cytokinesis-blocked micronucleus assay, the combined effect of both PARP-1 and DNA-dependent protein kinase (DNA-PK) inhibitors (3-AB and wortmannin, respectively)on the genotoxicity of the BNC reaction was further studied. DNA-PK is alsoactivated by DNA breaks and binds DNA ends, playing a role of utmost importance inthe repair of double strand breaks. The results obtained with micronuclei corroboratethat PARP-1 inhibition does not enhance the genotoxicity of the BNC reaction, and thatPARP-1 inhibition together with a concomitant inhibition of DNA-PK revealed the samesensitizing effect as DNA-PK inhibition per se.4:15 PM -Metabolomics and Proteomics for BNCT Studies. P. L.Mauri a , F. Basilico a , A. Zonta b , C. Ferrari b , Anna Maria Clerici b , A. Wittig cand W. Sauerwein ca ITB-CNR, Via Fratelli Cervi 93, 20090 Segrate Milano, Italy, b Università di Pavia,Dipartimento di Chirurgia, sezione di Chirurgia epato-pancreatica, Pavia, Italy,c Strahlenklinik, Universitätsklinikum Essen, GermanyIn the development of synthesis and delivery drug strategy, it is of primary importanceto determine the absorption levels and localization of an administered drug in biologicalfluids and tissues, in order to be able to characterize its potential utility, check itsconcentration in the tumor and optimize its administration. For this reason, efforts haveto be made to measure pharmaco-kinetics and to be able to quantify drugs in bloodsamples and in tissue extracts.In addition, to understand the biochemical and physiological differences between tumorcells and their normal counterparts and to be able to use these differences incompound design, synthesis and targeting, it is very important to investigate proteinprofiles related to different physiological states (normal and tumoral) as well as tostudy protein-protein interactions and protein modifications. For these reasons, wehave developed an integrated approach for characterization and quantification of boroncompounds in biological samples and the related protein profiles. Although several10B containing compounds belonging to different chemical families have been synthesized,there are still only two products currently used in BNCT trials for the treatmentof glioblastoma multiforme and melanoma: BSH (closo-undecahydro-1-mercaptododecaborate)disodium salt and BPA (boro-phenylalanine).In particular, BSH and BPA derivatives were analyzed by means of ion trap massspectrometry, using direct injection of samples. This approach allows their completecharacterization in a short time (about 2 min for analysis) and high sensitivity (limit ofdetection: 50 fmol). Moreover, for the same samples, protein analysis is performedusing the LC/LC-MS/MS (also called MudPIT, Multidimensional Protein IdentificationTechnology) approach. It consists of one single direct analysis of peptides obtainedfrom digestion of a protein extract. Peptides are separated first by ion exchange andthen by reversed-phase chromatography. The eluted peaks from 2DC (LC/LC) areanalyzed directly by mass spectrometry to obtain molecular weight and fragmentationof each peptide.This proteomic approach allows simultaneously the separationof digested peptides, their sequencing and then the identification of related proteinspresent in the samples (cell lines and urine). Using this methodology it is also possibleto detect protein modifications due to administration of boron compounds.The main objectives of a “metabolomics and proteomics” integrated approach, is todevelop and apply innovative methodologies for increasing sensitivity, accuracy andresolution to accelerate the development of tumor targeting drugs and drug deliverysystems suitable for Boron Neutron Capture Therapy.4:35 PM -Recommended Infusion Protocol for Neutron CaptureTherapy. L.F.Miller a , A.Rahim a , M.K.Khan a , T.L.Nichols b , G.W.Kabalka ca Nuclear Engineering Department, University of Tennessee, Knoxville. USA.,b Radiology & Internal Medicine, University of Tennessee Knoxville Medical Center.USA., c Chemistry Department, University of Tennessee, Knoxville. USADynamic PET scans of several GBM patients demonstrate that the biokinetics of tumorand healthy tissue are significantly different. It is noted that rates of uptake in tumorregions are approximately twice as fast as in normal tissue and that two rates of uptakeare clearly identified in each tissue region and in blood. Eigenvalues associated withexponential fits to time-dependent PET data are used to facilitate the identification ofrate constants. These rate constants are used in conjunction with second and fourthorder models that are used to simulate cellular-level distributions of pharmaceuticalsfor investigation of BNCT treatment protocols. Results from evaluations of multipleand single infusion protocols indicate that concentrations of pharmaceuticals in tumor,inter-cellular space, normal tissue, and blood depend significantly on rate constantsand that very long infusion times, of about 16 hours, would significantly improve theefficacy of BNCT.4:55 PM -Current clinical results of the Tsukuba BNCT trial.T. Yamamoto a , A. Matsumura a, *, K. Nakai a , Y. Shibata a , K. Endo a , F. Sakuraib , T. Kishi b , H. Kumada b , K. Yamamoto b , Y. Torii ba Department of Neurosurgery, Institute of Clinical Medicine, University of Tsukuba,Tenno-dai 1-1-1, Tsukuba City, Ibaraki 305-8575, Japan, b Department of ResearchReactor, Tokai Research Establishment, Japan Atomic Energy Research Institute, Tokai-mura,Naka-gun, Ibaraki 319-1195, JapanThe treatment protocol had been designed to determine the therapeutic dose of intaoperativeboron neutron capture therapy (IOBNCT) using a mixed thermal-epithermalbeam at JRR-4 in the Japan Atomic Energy Research Institute (JAERI) as well as tooptimize the JAERI Computational Dose Planning System (JCDS). The procedure ofIOBNCT and a current results of the clinical trial will be also described. The treatmentprotocol had been designed to determine the therapeutic dose of intaoperative boronneutron capture therapy (IOBNCT) using a mixed thermal-epithermal beam at JRR-4 in the Japan Atomic Energy Research Institute (JAERI) as well as to optimize theJAERI Computational Dose Planning System (JCDS). The procedure of IOBNCT anda current results of the clinical trial will be also described.Nine high grade gliomas (5 glioblastomas and 4 anaplastic astrocytomas) were treatedwith BSH-based IOBNCT. BSH (100 mg /kg body weight) was intravenously injected,followed by single fraction irradiation using the mixed thermal/epithermal beam. Theblood boron level at the time of irradiation averaged 29.9 (18.8-39.5). The peak thermalneutron flux by the post-irradiation measurements varied from 1.99 to 2.77 x109n cm-2sec-1. No serious BSH related toxicity was observed in this series. The interimsurvival data in this study showed median survival times of 23.2 months for glioblastomaand 25.9 months for anaplastic astrocytoma, results which are consistent withthe current conventional radiotherapy with/without boost radiation. Of 4 residual tumors,2 showed complete response (CR) and 2 showed partial response (PR) within6 months following BNCT. No linear correlation was proved between the dose and theoccurrence of early neurological events. The maximum boron dose of 11.7 Gy to 12.2Gy in the brain related to the occurrence of radiation necrosis.The clinical application of mixed thermal / epithermal beam and JRR-4 facilities inBSH-based IOBNCT proved to be safe and effective. After optimization of JCDS andcareful evaluation of the clinical data, a new protocol can be planned to introduceBSH-based IOBNCT using an epithermal neutron beam in which JCDS-based doseplanning and calculation are utilized.5:15 PM -Clinical Trial of BNCT for the Head and Neck Tumor. J.Hiratsuka a , T. Aihara b , N. Morita a , M. Uno b , T.Harada b , Y. Imajo a , Y. Imahori c ,T. Asano d , Y. Sakurai e , A. Maruhashi e , K. Ono ea Department of Radiation and Oncology, b Department of Otorhinolaryngology, KawasakiMedical School, c Department of Neurosurgery, Kyoto Prefectual Universityof Medicine, d Division of Applied Biological Chemistry, Osaka Prefectual University,e Research Reactor Institute, Kyoto UniversityBoron neutron capture therapy (BNCT) has been used clinically to treat patients withprimary brain tumors and a much smaller number of patients with cutaneous melanomaswho have not been candidates for conventional therapy. In order to rapidly adoptBNCT into our program of therapy, we decided to apply it to a head and neck tumorbecause our experimental data from hamsters suggested high boron accumulationinto neck tumors with the use of BPA. The number of patients with head and necktumors is much larger than that of those with malignant melanoma in Japan, and headand neck tumors existing superficially are most suitable for BNCT. In Japan, Kato et al.began using BNCT with both BSH and BPA for recurrent parotid gland carcinoma forthe first time and reported excellent preliminary results. On the basis of the encouragingresults of their pioneering clinical trials and the trend toward emphasizing quality oflife after treatment, we also began treating our patients with BNCT using BPA alone.We report the clinical results of our first case with a recurrent head and neck cancerrelapsed after operation, and note several problems that need to be solved in the nearfuture.The present case was 49-year old woman with recurrent submandibular gland carcinoma.She had presented in March 2003 with swelling of the right submandibular glandwith a suspicion of malignancy. The lesion was found to be a submandibular glandmucoepidermoid (high grade) carcinoma (T2N2bM0, Stage ?A). Tumor resection andright radical neck lymphnode dissection were carried out in April 2003. In September2003, a subcutaneous tumor was noted on the right neck. Thereafter it showed rapidgrowth. The tumor was inoperable because it had invaded the carotid and plexusbrachialis. The patient also strongly rejected re-operation. Radiological examinationrevealed no distant metastases. With the approval of the medical ethics committee,56

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 5 - B i o l o g y & C h e m i s t r ywe confirmed the BPA accumulating capacity of the tumor by 18F-BPA-PET(Tumor/Normal tissue ratio:2.9). She received 250mg/Kg of BPA. Thereafter, the tumor wasirradiated with epithermal neutron 5MW for 90 minutes. BNCT was repeated again usingepithermal neutron after one month later because the first BNCT was incompletedue to the patient’s movement during the irradiation. After the second BNCT, the tumorgradually decreased in size and disappeared. There have been no complications orside effects. Despite the high possibility of radio-resistance, BNCT was effective forthe treatment of this submandibular gland mucoepidermoid (high grade) carcinoma.Additional long-term follow-up is required to assess this treatment.57

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 6 - N u c l ea r E n g i n e e r i n g & P hy s i c sSession Chairs: George Borisov, Mark Rivard2:00 PM -Neutron fibres and possible applications to NCT. RamonF. Alvarez-Estrada a, *, Maria L. Calvo ba Departamento de Fisica Teorica I, Facultad de Ciencias Fisicas, Universidad Complutense,28040 Madrid, Spain, b Departamento de Optica, Facultad de CienciasFisicas, Universidad Complutense, 28040 Madrid, SpainWe summarize various previous researches regarding neutron guides of small transversecross section (neutron fibres), smaller than those of the standard hollow guidesand collimators employed currently. Specifically: 1) The confined propagation of thermalneutrons in suitable neutron fibres and a simple conjecture regarding possibleapplications to BNCT were formulated by 1984-86. 2) By 1992, certain experiments(employing bundles of polycapillary glass fibres) allowed to focalize a flux of confinedneutrons onto a region of size about 1mm, at a distance about 1cm from the exit end ofthe device. Those studies may not be widely known in the NCT community, but theymay be interesting for it. In what follows, to fix the ideas, we take the view that onepolicapillary glass fibre constitutes a physical implementation of a neutron fibre. Wepresent new estimates of neutron fluxes and of the total number of neutrons transmittedduring times T ( of the order of typical NCT durations), propagating confined alongneutron fibres. If the neutron flux generated by the source is not small, the total numberof those neutrons propagating confined in about an hour (for a bundle containing asuitable number of polycapillary glass fibres ) may have an order of magnitude aboutthat required for a standard BNCT treatment of a small tumour having size about 1mm. We point out and discuss some new possible specific applications of those neutronfibres to NCT at present (in typical NCT durations). They could allow to deliver andconcentrate neutron beams selectively in suitable regions of tissue of size smaller than1 mm, thereby reducing the undesirable delivery of radiation to healthy cells aroundregions with malignant tissue. Specifically, it is not claimed that the use of those neutronfibres could replace the current NCT treatments of the bulk of tumours, havingtypical sizes about some cm. The latter are adequately treated by means of standardtherapies employing the standard collimators (with transverse dimension from severalcm down to some mm). Rather, we entertain the possibility that neutron fibres could beuseful for: i) treating small tumours (of size smaller than 1 mm), ii) complementary therapiesof rather thin borders of tumours (not fully treated by the beam dealing with thebulk). A quick overview of spatial sizes in NCT, and eventual improvements in spatialresolutions in some aspects ot it, may suggest not to disregard the potential interest(for NCT, at least) of parallel research about improved focalizations and concentrationsof neutron beams on small spots (having size, at least, below the 1 mm scale).2:20 PM -Assessing Estimate Possibilities of Implementing InvasionNeutron Capture Therapy (INCT) Using CapillaryNeutron Optical Systems. G.I. Borisov, M.A. Kumakhov, R.I. Kondratenko,R.A. SpryshkovaInstitute for Roentgen Optics, RRC “Kurchatov Institute”, Moscow, Russian Federation/ Unisantis SA, Geneva, SwitzerlandDuring INCT, the exposed object represents an almost ideal neutrons trap, where thenumber of neutrons that entered the object through the inlet hole is equal to the numberof nuclear reactions, in which they are absorbed.With the help of this model and nuclear data on all nuclides under consideration fornormal biological tissue and in case of administration of various concentrations ofdosage-forming preparations on the basis of nuclides B(10), Gd(157) and U(235),atheoretic calculation was done in respect of the following INCT characteristics:- values of the effective masses irradiated by thermal neutrons;- energies absorbed in the object when hit by one thermal neutron;- partial compositions of main dosage-producing reactions;- partial compositions of absorbed dosage of thermal neutrons generated by maindosage-forming reactions;- quantities of thermal neutrons required to create a full absorbed dosage of 5 Gr inthe focal spot of 0,28 cm2 in area;- number of nuclear reactions of neutrons with dosage-forming nuclides of preparationsper one cell based on cell volume estimate of 7x7x7 m.If follows from the data obtained that the dosage-forming nuclide B(10) is not only theoptimal but also the sole one suitable for INCT.Using the results obtained, one can assess the parameters of capillary neutron opticalsystems (CNOS) required to implement INCT at specific experimental channels (EC)of research nuclear reactors.2:40 PM -Moderated 252 Cf neutron energy spectra in brain tissueand calculated boron neutron capture dose. Mark J.Rivard*, Robert G. ZamenhofDepartment of Radiation Oncology, Tufts-New England Medical Center #246, TuftsUniversity, 750 Washington Street, Boston, MA 02111, USACf-252 is a fast neutron emitter of relatively low average energy, and there is potentialto augment Cf-252 brachytherapy with boron neutron capture reaction (BNCR) doseenhancement for treatment of malignant disease. Cf-252 neutrons are readily moderatedby tissue, and may then be captured through the B-10(n,alpha)Li-7 nuclearreaction, Q = 2.79 MeV. Following this nuclear capture, a 477.6 keV photon is emitted93.7% of the time by relaxation of the excited 7Li nucleus. However, the large majorityof dose is locally absorbed high-linear energy transfer (LET) radiation, that is depositedby the alpha particle and recoiling lithium ion. Consequently, by incorporating B-10 loaded drugs designed to have affinity for malignant tumor cells, it is possible thatCf-252 brachytherapy may benefit from BNC dose enhancement. Calculations of thispotential dose enhancement were made for a variety of B-10 loadings, phantom diameters,and tumor diameters ranging from 2 to 6 cm. Phantom diameters ranged from10 to 30 cm, and the unencapsulated Cf-252 neutron energy spectrum was modeledwith an isotropic Maxwellian distribution. Neutron transport was calculated with ICRU44 brain material having a mass density of 1.04 g/cm^3 with uniform B-10 concentrationranging from 1 to 500 microgram B-10 per gram of brain tissue. An additionalstudy was performed to examine BNCR dose as a function of radius within a 15-cmdiameter head phantom, in which 0.010 mg B-10 per gram of brain tissue was uniformlydistributed. A centrally positioned tumor was modeled with a B-10 loading of 30ppm with tumor diameter ranging from 2 to 6 cm. These loadings were used to modela clinical environment.3:00 PM -Neutron Activation of Patients following BNCT at theHFR Petten. R.L.Moss a, *, F.Stecher-Rasmussen b,c , K.Appelman c , A.Roca a,d ,W.Sauerwein e , A.Wittig ea Joint Research Centre, Institute for Energy, Postbus 2, 1755ZG, Petten, Netherlands,b NCT Physics, Alkmaar, Netherlands, c Nuclear Research And Consultancy Group(NRG), Postbus 1, 1755ZG, Petten, Netherlands, d University of Al.I.Cuza,70050 Iasi,Romania, e Universitaetsklinikum, Strahlenklinik, Hufelandstr. 55, 45122 Essen, GermanyBoron Neutron Capture Therapy (BNCT) involves the irradiation of cancer patients bymeans of neutron radiation produced currently by nuclear research reactors. The treatmentis a bi-modal form of therapy, whereby the patient is infused with a boron-loadedcompound prior to irradiation. The beam itself is a mix of neutrons and gammas, butin most cases predominately epithermal neutrons, i.e in the energy range 1eV to 10keV. With a higher concentration of 10B atoms in the tumour, in comparison to healthytissue, neutron capture in the 10B atoms gives an effective tumoricidal dose to thetumour cells, whilst sparing the cells of the healthy tissue. However, due of the naturalpresence of other elements in the tissue, neutron capture also occurs in some of theisotopes of these elements. The patient is therefore, following BNCT, radioactive andconsequently, radiation protection measures need to be taken. At the HFR Petten,procedures are in place to perform appropriate measurements immediately after andsometime thereafter on the patients. Likewise, out of scientific curiosity, 3 patientshave undergone direct measurements following treatment using a portable gammaspectrometry system to identify which elements and confirm which isotopes are activated.The main isotopes were, as expected, identified as 38Cl, 49Ca and 24Na.However, in two patients, the isotopes 198Au and 116mIn, were also present. To someextent, 18O and other minor isotopes contributed to the overall activity, though theircontribution is negligible.As standard procedure, measurements using a hand-held dosimeter were taken on all27 patients treated at Petten. Each patient in the Petten protocol received 4 fractions.As such, over 100 radiation days have been performed. A summary of all these measurementsis presented here, as well as more in-depth information and conclusionsfrom the gamma ray spectroscopy measurements. In summary, peak levels, i.e. atcontact and directly after radiation, are of the order of 40-60 micro Sv/h, falling to lessthan 10 micro Sv/h some 30-50 minutes after treatment. The initial activity is predominatelydue to 49Ca, whilst the remaining activity is predominately due to 24Na.58

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 6 - N u c l ea r E n g i n e e r i n g & P hy s i c s3:20 PM -A benchmark of treatment planning system THORplanwith SERA. Y-W H. Liu, T.Y. LinDepartment of Engineering and System Science, National Tsing Hua University, Hsinchu,Taiwan, ROCA treatment planning system for boron neutron capture therapy, THORplan, is developedat Tsing Hua University at Taiwan. It uses MCNP4C for the flux/dose calculationwith pre-processor and post-processor developed for handling the patient image andexhibiting the dose distribution.The source spectrum at the exit of the new THOR epithermal neutron beam is generatedfor use in the treatment planning system. The performance of the new beam forBNCT is evaluated by THORplan and the existing treatment planning system SERA.The epithermal neutron flux near the surface calculated by THORplan is ~8% higherthan the SERA result, due to the fluctuated fast neutron flux calculated by SERA. Thethermal neutron flux at the peak (2.25cm) on the central axis calculated by THORplanis 5% lower than the SERA result, which is believed to be due to the difference of materialcomposition mock up of patient head by the two codes.The dose in THORplan is directly calculated by using the KERMA for individual isotopeprovided in the MCNP/ENDF60 library. The neutron dose of SERA near the surface islower than THORplan because the KERMA for hydrogen in SERA is lower than that inENDF60 at energies below 10eV, which causes the hydrogen dose of epithermal neutronis somewhat underestimated in the SERA calculation. The gamma ray dose andthe boron dose around the peak calculated by THORplan are 5~6% lower than SERAdue to the 5% difference in thermal neutron flux between the two codes.For patient irradiated at 10 cm away from the beam exit, THORplan calculation showsthat the advantage depth is 8.9cm, and advantage ratio is 5.8 if boron concentrationin tumor and normal tissue are assumed to be 65 ppm and 18 ppm. The maximumdose rate for normal tissue is ~50 cGy/min. The maximum therapeutic ratio is 6. Theseparameters are comparable with those of the current BNCT facilities at Finland andSweden.3:55 PM -Neutron radiography of human liver metastases afterBPA infusion. S. Altieri a,b, *, A.Braghieri b , S. Bortolussi a , P. Bruschi a ,F.Fossati a , P. Pedroni b , T.Pinelli a,b , A. Zonta c , C.Ferrari c , U.Prati d , L.Roveda e ,S.Barni f , P.Chiari f , R.Nano fa Dipartimento di Fisica Nucleare e Teorica dell’Università, via Bassi, 6 27100 Pavia,Italy, b Istituto Nazionale di Fisica Nucleare-Sezione di Pavia, via Bassi, 6 27100Pavia, Italy, c Dipartimento di Chirurgia dell’Università, via Aselli, 45 Pavia, Italy,d Chirurgia Sperimentale e Tecnologie Chirurgiche Innovative IRCCS Policlinico S.Matteo, Pavia, Italy, e Chirurgia Epato-Pancreatica IRCCS Policlinico S. Matteo,Pavia, Italy, f Dipartimento di Biologia Animale e Centro di Studi per l’IstochimicaIGM-CNR, Pavia, ItalyA precise knowledge of the Boron concentration, both in tumor and healthy tissues,is an essential requirement for the BNCT treatment of cancer. Many methods can beused to detect the Boron signal coming from biological samples, but it is very importantto know what type of tissues are present in the sample. In the case of liver metastases,for example, tumor cells are disseminated through the healthy tissues. Tumor cells,normal hepatocytes, blood cells and necrotic material can be simultaneously presentinside a metastatic nodule having a diameter of a few millimeters.In this paper we show the results of the analysis of tumor samples from human liverinfused by BPA that was performed using neutron radiography. We have found thatthe boron concentration is a very different in the different biological materials. Thisdemands to take into account the morphology of the sample for the correct evaluationof the boron concentration.Keywords: BNCT; liver metastases; imaging; neutron radiography4:15 PM -MicroPET-based pharmacokinetic analysis of the radiolabeledboron compound [ 18 F]FBPA-F in rats with F98glioma. J.C. Chen a, *, S.M. Chang a , F.Y. Hsu b , H.E. Wang a , R.S. Liu ca Department of Medical Radiation Technology, Institute of Radiological Sciences,National Yang-Ming University, 155 Li-Nong Street, Sec 2, Taipei, 112, Taiwan, b Departmentof Radiological Technology, Yuanpei University of Science and Technology,Taiwan, c School of Medicine, National Yang-Ming University, Taipei, TaiwanBoron neutron capture therapy (BNCT) uses an epithermal neutron source to bombard10B atoms inside a patient’s brain and this produces short range alpha particles viaa nuclear reaction that can kill tumor cells effectively. BNCT has been proposed forselective destruction of the infiltrating cells in brain tumors since 1936. In Taiwan, theopportunity to start BNCT research occurred in 1998. BNCT in Taiwan is a new technologyand there is a need to design a local accurate treatment planning system. Theaim of this study is to investigate the use of 4-borono-2-[18F]-fluoro-L-phenylalaninefructose([18F]-FBPA-Fr) in glioma-bearing rats using in vivo small animal positronemission tomography (PET) imaging (microPET).Methods: We used microPET R4 (Concorde, U.S.A.) supported by PET gene probecore to do high-resolution rat imaging. Male Fischer 344 rats with F98 glioma in theleft brain were used for the studies. Dynamic PET images of the F98 glioma-bearingrats were performed on the 13th day after tumor inoculation using the microPET R4system that produced 63 image slices over a 7.89 cm axial field of view (FOV). Timeactivitycurves (TACs) were plotted for both the tumor and the normal tissue locatedin the nontumoral control area. Regions of interest (ROIs) of tumor and normal tissuewere identified from each image plane in which tumor was visible on the final timeframe. The mean of radioactivity concentration of the tumor ROIs and the normal ROIsat different time frames were calculated. Quantified knowledge of the tissue kineticparameters in the regions of the brain is able to offer information such as the metabolicrate or 10B level and is useful in clinical applications. Dynamic microPET imaging withinjection of radioactive tracer can be used for this measurement. We used a modifiedthree-compartment physiological model of [18F]FBPA-F with K1 (ml/g/min), k2 (min-1),k3 (min-1), and k4 (min-1) obtained from arterial blood samplings as input function fortracer kinetic modeling. Results: The accumulation ratios of [18F]FBPA-F for gliomato-normalbrain approached 3. The uptake characteristics of BPA-F and[18F]FBPA-Fwere similar to a previous study. The results indicate that 4 h after BPA-F injectionwould be the optimal time for BNCT. Rate constants [K1 (ml/g/min), k2 (min-1), k3(min-1), k4 (min-1)] were estimated for the tumor and normal tissue.Conclusion: This microPET imaging with [18F]-FBPA-F can be used as a probe for10B-BPA-F in BNCT. This preclinical study provides useful information for the futureclinical application of BNCT in patients with brain tumors.4:35 PM -Development and application of an unconstrained techniquefor patient positioning in fixed radiation beams.W.S. Kiger III a, *, J.R. Albritton b , X.Q. Lu a , M.R. Palmer ca Department of Radiation Oncology, Beth Israel Deaconess Medical Center, HarvardMedical School, 330 Brookline Avenue, Shapiro-505, Boston, MA 02215, USA,b Nuclear Engineering Department, Massachusetts Institute of Technology, 77 MassachusettsAvenue, Cambridge, MA 02139, USA, c Department of Radiology, BethIsrael Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue,Boston, MA 02215, USAA flexible technique for positioning patients in fixed orientation radiation fields such asthose used in neutron capture therapy (NCT) has been developed. The positioningtechnique employs reference points marked on the patient in combination with a 3Ddigitizer to determine the beam entry point and a template fitted to the patient’s head isused to determine the proper beam orientation. A coordinate transformation betweenthe CT image data and reference points on the patient determined by a least squaresalgorithm based on singular value decomposition is used to map the beam entry pointfrom the planning system onto the patient. The technique was validated in a phantomstudy where the mean error in entry point placement was 1.3 mm. Five glioblastomamultiforme patients have been treated with NCT using this positioning technique.Key words: Boron Neutron Capture Therapy (BNCT), patient positioning, coordinatetransformation, fixed radiation beams4:55 PM -MCNP study for epithermal neutron irradiation of anisolated liverat the Finnish BNCT facility. P. Kotiluoto*,I. AuterinenVTT Technical Research Centre of Finland, P.O. Box 1608, FIN-02044 VTT, FinlandIntroduction: Boron neutron capture therapy (BNCT) has been mainly used to treatmalignant brain tumors. Only quite recently, BNCT has been applied to treat multipleliver metastases. This was first done in Italy, where the liver was removed from thepatient, irradiated with thermal neutrons, and then reimplanted.Based on ICRP Publication 89, the reference mass of the liver for adult male is 1800 gand for adult female 1400 g. Thus, liver is rather large organ, and one could expect thatby using epithermal neutrons as primary incident radiation, the neutron radiation wouldpenetrate much deeper to the cancerous tissue. To test this assumption, we have carriedout Monte Carlo simulations with MCNP code, both for the actual epithermal neutronsource of the FiR 1 reactor, and for a pure Maxwellian thermal neutron source.59

Eleventh World Congress on Neutron Capture TherapyThursday, October 14, 2004 - PMParallel Session 6 - N u c l ea r E n g i n e e r i n g & P hy s i c sMaterials and methods: A simplified MCNP model of a human liver was obtained fromHanna Koivunoro, Lawrence Berkeley Laboratory, USA. It simply consists of an octantof an ellipsoid. The maximal dimensions of the liver in this model are 22.5x12x12 cm.The liver mass is about 1800 g, corresponding to the size of the liver of a healthy adultmale.The liver model was placed into an MCNP model of the FiR 1 epithermal beam, insidethe bismuth collimator cone. Some modifications were done for the current beam usedfor patient irradiations, such as replacing part of the aperture with a 10 cm thick cylindricalbismuth frontplate, in order to create a more homogeneous epithermal cavity.The material definition for liver was taken directly from ICRU-46, as well as neutronKERMA factors and energy-mass absorption coefficients for photons.In order to be able to compare the results for the epithermal irradiation situation to thethermal irradiation of the liver, another simulation was performed with the same livermodel now surrounded by a spherical inward directed thermal neutron source.The TPDmax, the dependence of TPD on BDE thickness, and the BDE(TPDmax)were established as appropriate BDE optimization parameters. Based on these criteriaand other practical considerations, the suitable choice as BDE among the candidatematerials considered in this study for treatments involving tumors located at shallowdepths would be (C2H4)n while beryllium metal was judged as more appropriate fortreatment of deep-seated tumor.60Results: The calculated neutron, photon, and boron dose profiles will be presented.The results clearly show, that the epithermal irradiation leads to a build-up distributionof the boron dose, whereas with incident thermal radiation the boron dose maximum isonly at the liver surface, but deeper in the tissue becomes attenuated.Discussion and conclusions: There is some clear advantage for using epithermal neutronsas incident radiation for treating isolated liver. The epithermal field penetratesdeeper into the liver and creates a build-up distribution of the boron dose, which isquite the opposite compared to the incident thermal irradiation. The best result mightbe achieved by mixing epithermal and thermal irradiation in the right proportions.Our results strongly encourage further studying of irradiation arrangement of an isolatedliver with epithermal neutron fields.Aknowledgements: We authors wish to thank Hanna Koivunoro for kindly letting us touse her MCNP liver model in our calculations.5:15 PM -Optimization parameters for BDE in BNCT using nearthreshold 7 Li(p,n) 7 Be direct neutrons. Gerard Bengua a, *,Tooru Kobayashi a , Kenichi Tanaka b , Yoshinobu Nakagawa ca Research Reactor Institute, Kyoto University, Sennan-gun, Kumatori-cho, Oaza,Okubo, 1726-2, Epaule Sakaue Rm 201, Kumatori-cho, Osaka 590-0401, Japan,b Research Institute for Radiation Biology and Medicine, Hiroshima University, Japan,c National Kagawa Children’s Hospital, JapanThe dose contribution of 10B(n,a)7Li reaction in BNCT using near threshold7Li(p,n)7Be direct neutrons can be increased through the use of materials referred toas boron dose enhancers (BDE). In this paper, possible BDE optimization criteria weredetermined from the characteristics of candidate BDE materials namely (C2H4)n,(C2H3F)n, (C2H2F2)n, (C2HF3)n, (C2D4)n, (C2F4)n, beryllium metal, graphite, D2Oand 7LiF. The treatable protocol depth (TPD) was used as the assessment index forevaluating the effect of these materials on the dose distribution in a medium undergoingBNCT using near threshold 7Li(p,n)7Be direct neutrons.The particle fluence in a cylindrical water phantom was generated using MCNP-4Btransport code. The absorbed dose distributions were obtained from the particle fluenceand the respective dose conversion factors for gamma rays and heavy chargedparticles. The TPD for particular BDE thicknesses were computed based on the doseprotocol being used for intra-operative BCNT in Japan where the treatment dose fortumor from heavy charged particles is set at 15Gy and the tolerable dose for normaltissue due to heavy charged particles and gamma rays were separately set at 15Gyand 10Gy respectively. The boron dose concentration in the tumor and normal tissuewere assumed to be 30ppm and 10ppm, respectively.The maximum TPD (TPDmax) did not exhibit an explicit dependence on material typeas evidenced by its small range and arbitrary variations. The TPDmax generated by aBDE material was deemed as an important consideration in choosing the suitable BDEmaterial for treatment since a high TPDmax would be advantageous for treatmentsinvolving deep-seated tumors.The dependence of TPD on BDE thickness was influenced by the BDE material usedas indicated by the sharply peaked TPD versus BDE thickness curves for materialswith hydrogen compared to the broader curves obtained for those without hydrogen.A small dependence of TPD on BDE thickness will ensure that large variations in TPDwill not occur for small deviations in the intended BDE thickness to be used for actualtreatment.A Thinner BDE thickness required to achieve TPDmax (BDE(TPDmax)) is likewise favoredsince they result in less reduction of the dose rate within the irradiated medium.For the candidate BDE materials considered in this study, BDE(TPDmax) was foundto be thinner for materials with hydrogen.

Session Chairs: Sara Liberman, Akira Matsumura8:30 AM -Preclinical and Translational studies in BNCT (InvitedSpeaker) Koji OnoKyoto Univ. Research Reactor Institute, Osaka 590-0494, JapanEleventh World Congress on Neutron Capture TherapyFriday, October 15, 2004 - AMClinical ApplicationsBuenos Aires, San Martín 5481, 1417 Buenos Aires, Argentina, e Hospital CosmeArgerich, Servicio de Neurocirugía, Almirante Brown 240, 1155 Buenos Aires, Argentina,f Departamento de Radioquímica y Química de las Radiaciones, Centro AtómicoEzeiza, Argentina, g Departamento de Reactores y Centrales Nucleares, CentroAtómico Constituyentes, Argentina9:10 AM -Clinical Results of Modified BNCT for Malignant Gliomausing Two Boron. Shin- Ichi Miyatake a , Yoshinaga Kajimoto a ,Shinji Kawabata a , Kunio Yokoyama a , Toshihiko Kuroiwa a , Shin- IchirouMasunaga b , Yoshinori Sakurai b , Akira Maruhashi b , Yoshio Imahori c , MitsunoriKirihata d , Koji Ono ba Osaka Medical College,Neurosurgery, b Kyoto University research reactorinstitute,Radiation Oncology, c Kyoto Prefectual University of Medicine,Neurosurgery,d Osaka Prefectual university,AgricultureIntroduction: To improve the effectiveness of boron neutron capture therapy for malignantgliomas, we utilized epithermal neutron instead of thermal neutron for deeppenetration and used two different boron compounds, sodium borocaptate (BSH) andboronophenylalanine (BPA) with different accumulation mechanism for compensatingtheir faults each other and achieving higher born levels in tumorsMaterials and Methods: Ten glioblastoma and one anaplastic astrocytoma and oneanaplastic oligoastrocytoma patients were treated with this BNCT, since January 2002to October 2003. Only one glioblastoma patient had no postoperative enhanced lesionon MRI. The patients were received 18F-labeled BPA-positron emission tomography(PET), if available, to assess the accumulation and distribution of BPA before neutronirradiation. The neutron distribution was estimated by dose-planning system SERAbefore irradiation and confirmed by the direct measurement using activation of goldby neutron. The neutron irradiation time was determined not to exceed 13 Gy-Eq (Gyequivalent)to the normal brain and as much as we could to contrast enhanced tumorfor the fresh cases. For the recurrent cases, irradiation time was determined in eachcase according to the previous irradiation dose and field. Five grams of BSH and 250mg /kg of BPA were administrated 12 hours and 1 hour before neutron irradiation,respectively. The effect of BNCT on tumors was assessed by volumetrical methods,on MRI or CT scan. The improvement on images were assessed between 2 to 7 daysafter irradiation as initial effects and also assessed to analyze the maximum effects onserial radiographical images.Results: The lesion/normal brain (L/N) ratio of BPA before BNCT on PET varied from2.65 to 7.8. There was a tendency of low L/N ratio in recurrent cases and high in primarycases. Concentration of BSH and BPA in tumor tissue during neutron irradiation,estimated by repeated venous sampling and L/N ratio on PET, varied from 17 to 31.4ppm, and from 33.6 to 98 ppm, respectively. The neutron irradiation time varied from60 to 120 minutes. The mean of initial tumor volumes prior to BNCT was 45.7 (2.2 to107.6) cm3. Irrespective of initial tumor volume, in every cases who had assessablelesions, the improvements on MRI/CT images were recognized both on initial assessments(17.4 to 71.0 % , mean 51.5 %) and on follow-up assessments (30. 3 to 87.6 %,mean 61.7%). More than 50% of contrast-enhanced lesions disappeared in 7 out of 11patients during the follow-up period.Discussion: Definitive effects on mass reduction and on decrease of edema of surroundingbrain tissue were recognized by this BNCT in every cases. However 4 patientsout of 11 cases had already died by the tumor recurrence or CSF dissemination.We also experienced tumor recurrence on follow-up radiological analysis in another4 patients. The cause of recurrence seemed to be absolute shortage of neutron fluenceespecially in deep part of the tumor and uneven distribution of 10B compoundin tumor tissues.Conclusions: The good improvements of malignant glioma patients on radiographicalimages were obtained by this boron neutron capture therapy.9:35 AM -Biodistribution studies of boronophenylalanine–fructosein melanoma and brain tumor patients in Argentina.S.J. Liberman a,b, *, A. Dagrosa a,c , R.A. Jiménez Rebagliati a,b , M.R. Bonomi d ,B.M. Roth d , L. Turjanski e , S.I. Castiglia a,f , S.J. González a,g , P.R. Menéndez d ,R. Cabrini a,c , M.J. Roberti a,b , D.A. Batistoni a,ba Comisión Nacional de Energía Atómica (CNEA), Avda. del Libertador 8250, 1429Buenos Aires, Argentina, b Departamento de Química, Centro Atómico Constituyentes,Avda, General Paz 1499, 1650 San Martin, Prov. de Buenos Aires, Argentina,c Departamento de Radiobiología, Centro Atómico Constituyentes, Argentina, d Institutode Oncología Angel H. Roffo, Departamento Terapia Radiante, Universidad deThe aim of this study was to implement the 10B-enriched L-p-boronophenylalaninefructosecomplex (10BPA-F) infusion procedure in potential BNCT patients, includingmelanoma of extremities and high-grade gliomas. In both cases the boron concentrationratios tumor/blood (T/B) and also skin/ blood (S/B) in the melanomas weredetermined. Two brain tumors (glioblastoma and ganglioglioma) and four metastaticmelanoma of extremities patients were analyzed.Boron measurements were performed by inductively coupled plasma optical emissionspectroscopy (ICP OES). To predict 10B concentration in blood a pharmacokinetictwo-compartment open model was applied (Kiger et at., J. Neuro-Oncol. 62, 171-186,2003) to fit the experimental biodistribution data.The BPA-F was intravenous infused. Five patients received 100 mg kg-1 over 60 - 90min. One melanoma patient received 300 mg kg-1 in three studies. Blood samples forboron analysis were taken during and for several hours after the end of infusion. Thetissue extraction times by surgery or biopsy ranged between 45 to 400 min after theend of the infusion.The four pharmacokinetics studies with tissue sampling of melanoma showed T/Bratios between 1.4 and 2.6 (2.1 ± 0.4) and S/B ratios from 0.95 to 2.1 (1.5 ± 0.4).The highest T/B values were obtained for a patient who had surgery. The T/B boronuptake in the glioblastoma gave a marked dispersion among the five samples taken(from 1.8 to 3.4) in agreement with that expected for those tumors and their cellularityeffect (Coderre et al., Radiat. Res. 149, 163-170, 1998). The T/B ratios measured forthe ganglioglioma (between 0.7 and 1.2) may indicate that this tumor is not a goodcandidate for BNCT using BPA-F.The maximum boron concentration in blood at the end of BPA infusions of 100 mg kg-1BPA are in the range 5.5 - 9.8 µg g-1, while for 300 mg kg-1 are from 22.1 to 25.3 µgg-1 (for the same patient). Three hours after the end of the infusion concentrations arein the ranges 2.6 - 3.7 and 10.0 - 13.9 µg g-1 respectively.The T/B ratio measured for the nodular metastatic melanoma lies near the lower limit ofthe values reported in the literature. In view of the poorer response to BNCT reportedfor the nodular metastatic melanomas and also for melanomas associated to metastasis,our T/B experimental results may suggest that a tumor to blood ratio slightly lowerthan 3.5 may contribute to other factors in decreasing the tumor response for metastaticnodular type cases. Because the current information on skin melanomas, bothnodular and metastatic is rather limited, we plan to pursue further pharmacokineticevaluations in order to increase the reliability of the estimated T/B and to compare theresults with the clinical outcome.Good fitting for boron distribution data is achieved by applying the open two-compartmentmodel. This procedure gave the best agreement between the proposed and thedelivered skin dose in the retrospective analysis of our first BNCT clinical trial.10:20 AM -The new treatment protocol using epithermal neutronfor intraoperative boron neutron capture therapy at JRRJRR-4. A. Matsumura a, *, T. Yamamoto a , K. Nakai a , H. Kumada a,c , TMizutani b , H Takahashi b , S Kihara b , T. Kageji d , Y. Nakagawa e , K. Yamamotoc , K. Endo a , Y. Shibata a , A. Morita f , T. Todo fa Department of Neurosurgery a and Anesthesiology, b University of Tsukuba, 1-1-1 Tennodai, Tsukuba,Ibaraki 305-8575, Japan Atomic Energy Research Institute,Tokai Establishment, c 2-1 Shirane Tokai, Naka, Ibaraki 319-1195, Department ofNeurosurgery, University of Tokushima, d 3-18-15 Kuramoto, Tokusihima 770-8503,Department of Neurosurgery, Kagawa National Children’s Hospital, e 2603 Zentsuji,Kagawa 765-8501 and Department of Neurosurgery, University of Tokyo, f 7-3-1Hongo, Bunkyo, Tokyo 113-0033, JapanBased on the basic and clinical investigations, the authors proposed a new protocol ofboron neutron capture therapy (BNCT) for malignant gliomas using epithermal neutronbeam at JRR-4. The planned dose in this protocol is based on the data from previousBNCT using thermal and mixed thermal-epithermal neutron protocols.Using the new protocol, the dose distribution in the deeper region is improved and thedistribution to the lateral marginal area is also improved. Moreover, the maximum dosein the target is reduced. The therapeutic procedure has been enabled by applicationof the JCDS treatment planning system in combination with patient setting systemthat was developed by the authors. A multi-institutional phase I/II clinical trial is beingstarted and the preliminary clinical experience will be presented.61

10:45 AM -Clinical results of boron neutron capture therapy usingmixed epithermal- and thermal neutron beams inpatients with malignant glioma. T. Kageji a , S. Nagahiro a , S.Uyama a , Y. Mizobuchi a , H. Toi a , Y. Nakagawa b , H. Kumada ca Department of Neurosurgery, School of Medicine, University of Tokushima, Tokushima,Japan, b Department of Neurosurgery, National Kagawa Children’s Hospital,Kagawa, Japan, c Department of Research Reactor, Tokai Research Establishment,Japan Atomic Energy Research Institute, Tokai, Ibaraki, JapanThe purpose of this study was to clarify the clinical interim results of boron neutroncapture therapy (BNCT) using mixed epithermal- and thermal neutron beams in patientswith malignant glioma. Eighteen patients with malignant glioma (glioblastoman=16, anaplastic ependymoma n=1, PNET n=1) underwent mixed epithermal- andthermal neutron beam and sodium borocaptate between 1998 and 2004. The radiationdose (i.e. physical dose of boron n-alpha reaction) in the protocol used between 1997and 2000 (Protocol A, n=8) prescribed a maximum tumor volume dose of 15 Gy. In2001, a new dose-escalated protocol was introduced (Protocol B, n=4); it prescribes aminimum tumor volume dose of 18 Gy or, alternatively, a minimum target volume doseof 15 Gy. Since 2002, the radiation dose was reduced to 80-90% dose of Protocol Bbecause of acute radiation injury. A new Protocol was applied to 6 glioblastoma patients(Protocol C, n=6).The estimated median survival time after diagnosis and after BNCT in all patients were15.7 and 13.8 months, respectively. In 8 patients of Protocol A, the estimated mediansurvival time after diagnosis was 16.0 months; 1-year and 2-year survival rate were62.5% and 12.5%, respectively. On the other hand, in 10 patients in Protocol B andC, the estimated median survival time after diagnosis was 15.1 months; 1-year and 2-year survival rate were 50.0% and 33.3%, respectively. Our limited clinical evaluationsuggests that BNCT could achieve local control of glioblastoma at the primary site andthat possible dose escalation is limited. While the dose escalation can contribute tothe improvement of survival rate, it results in the radiation injury. We conclude that notonly the radiation dose at the target point, but also the distribution of neutron flux in theradiation field may contribute to the cure of glioblastoma by BNCT.11:10 AM -Trials of the EORTC BNCT Group. W. SauerweinDept. of Radiation Oncology, Universitätsklinikum Essen, Essen, Germany11:35 AM -BNCT trial at MIT. Paul BusseDepartment of Radiation Oncology, Massachusetts General Hospital, Boston, MAEleventh World Congress on Neutron Capture TherapyFriday, October 15, 2004 - AMClinical Applications62