Download the ESMO 2012 Abstract Book - Oxford Journals

survival rates of 13-25% (Prieto et al, CCR, 2012). Evaluation of the long-term
survival benefit of IPI for MM pts from 3 additional completed phase II studies
continues and we now report 5-yr survival data.
Methods: Trials included: (1) CA184-008, a single-arm study of IPI at 10 mg/kg in
previously treated pts; (2) CA184-022, a dose-ranging study of IPI at 0.3, 3, or 10
mg/kg in previously treated pts with crossover from lower doses to 10 mg/kg allowed
upon disease progression; and (3) CA184-007, a study of IPI at 10 mg/kg +/prophylactic
budesonide in treatment-naive and previously treated pts. IPI was given
q 3 wks x 4 (induction); eligible pts could receive reinduction or maintenance IPI q
12 wks from Week 24. The analysis reports OS with updated last known alive date or
death based on data collected through March 2012. Safety data for these trials have
been previously published.
Results: Survival rates at 5 yrs ranged from 12-28% in pretreated pts and 38-50% in
treatment-naive pts. The table summarizes median OS, previously reported yearly
survival rates, and current 5-yr survival rates.
Trial N
IPI dose
(mg/kg)
Median
OS,
months
OS rate, %
1-yr 2-yr 3-yr 4-yr 5-yr
008 155* 10 10.2 47.2 32.8 23.3 19.7 18.2
022 72* 10 11.4 48.6 29.8 24.8 21.5 21.5
72* 3 8.7 39.3 24.2 19.7 18.2 16.5
73* 0.3 8.6 39.6 18.4 13.8 13.8 12.3
007 57 (total) 10 + placebo 19.3 62.4 41.8 34.4 32.0 32.0
32 †
30.5 71.4 56.6 42.5 37.7 37.7
25* 14.8 50.8 24.2 24.2 24.2 24.2
58 (total) 10 + 17.7 55.9 41.1 38.7 36.2 36.2
21 budesonide
†
45.0 65.9 57.7 57.7 49.5 49.5
37* 8.5 49.9 31.6 28.4 28.4 28.4
*Previously treated; †Treatment naive.
Conclusions: Across studies, 5 yr OS rates appear similar to those at 4 years,
suggesting that IPI monotherapy may result in prolonged survival in some pts with
MM. Further research and analyses are needed to identify the pt population with
MM most likely to achieve long term survival benefit from IPI therapy.
Disclosure: C. Lebbe: Participation to BMS Advisory Boards. J.S. Weber: For BMS,
only honoraria less than 10K dollars and advisory role in Ad Boards. M. Maio: Paid
Advisor in Boards from BMS and Roche. B. Neyns: No conflict of interest to report
with regards to this abstract. K. Harmankaya: With Regards to conflict of interest in
this abstract, unpaid investigator on this study. K. Chin: Employed by BMS; Stock
ownership. D. Opatt McDowell: Employed by BMS; Stock ownership. L. Cykowski:
Employed by BMS; Stock ownership. M.B. McHenry: Employed by BMS, Stock
OwnershipJ.D. Wolchok: I am a consultant to Bristol-Myers Squibb, Merck and
Glaxo SmithKline. I receive research funding from Bristol-Myers Squibb.
1117PD LONG TERM SURVIVAL AND IMMUNOLOGICAL
CORRELATES IN METASTATIC MELANOMA TREATED
WITH IPILIMUMAB AT 10 MGS WITHIN AN EXPANDED
ACCESS PROGRAM
A.M. Di Giacomo 1 , L. Calabrò 1 , R. Danielli 1 , I. Pesce 1 , E. Fonsatti 1 , E. Bertocci 1 ,
D. Giannarelli 2 , M. Biagioli 3 , M. Altomonte 1 , M. Maio 1
1 Department of Oncology, Medical Oncology and Immunotherapy,University
Hospital of Siena, Siena, ITALY, 2 Medical Oncology, Regina Elena National
Cancer Institute, Roma, ITALY, 3 Department of Medicine, Dermatology, University
Hospital of Siena, Siena, ITALY
Background: Ipilimumab (IPI) has shown long lasting responses in metastatic
melanoma (MM) patients (pts) in phase II/III trials (Hodi et al., NEJM 2010; Robert
et al. NEJM 2011; Prieto et al., CCR 2012). We have previously reported a significant
clinical efficacy of IPI in 27 heavily pre-treated MM pts enrolled within the 10 mgs
expanded access program (EAP) available at the University Hospital of Siena; one
and 2-years survival were 34.8% and 23.5%, respectively (Di Giacomo, et al. CII,
2010). In spite of its clinical efficacy, no definitive predictive markers of response
have been identified yet. We report here the survival follow-up of this cohort of MM
pts and preliminary data on immunological correlates.
Methods: Twenty-seven pts with stage III (2) or IV (25) MM, progressing to a
median of 3 (1-5) systemic therapies, were treated according to the EAP with IPI at
10 mg/Kg i.v. given at weeks (wks) 1, 4, 7, 10 during the induction phase, and every
12 wks from W24 in the maintenance phase. Peripheral blood mononuclear cells
were collected from 17 pts at baseline, W7, W12 and W24 and analyzed by flow
cytometry for ICOS expression on both CD4 and CD8+ T cells. The ratio between
absolute blood neutrophils (N) and lymphocytes (L) count was determined at
baseline, W4, W7 and W10 for 27, 27, 24 and 23 pts, respectively.
Results: With a median follow-up of 9.6 months the median OS was 9.6 months
(95% CI: 4.2-16.1; range: 3.1-51.2; three and 4-years survival rates were 20.9%, with 5
long-term survivors (>4 yrs). A significant (p < 0.05) increase in the percentage and
absolute number of CD4 + ICOS+ and CD8 + ICOS+ circulating T cells was observed
Annals of Oncology
from W7 on. Compared to baseline, pts with a fold increase in CD4 + ICOS+ and
CD8 + ICOS+ T cells higher than 4 at W7 and W12 experienced a clinical benefit
(SD, PR and CR). Pts with a N/L ratio lower than median at W7 and W10 had a
significantly better survival.
Conclusions: ipi can induce long-term survivals in heavily pre-treated MM pts.
Circulating ICOS+ T cells and N/L ratio in the course of treatment with ipi may
represent predictive markers of clinical effectiveness in the daily practice.
Disclosure: M. Maio: Advisory Board and Honoraria from Bristol-Myers Squibb and
Roche. All other authors have declared no conflicts of interest.
1118PD MULTIPLE PRIMARY MELANOMAS FROM SAME PATIENTS
PRESENT DISCREPANT SOMATIC ALTERATIONS IN MAIN
CANDIDATE GENES
M. Colombino 1 , M. Sini 1 , V. De Giorgi 2 , A. Lissia 3 , D. Massi 4 , C. Rubino 5 ,
A. Cossu 3 , F. Ayala 6 , P.A. Ascierto 6 , G. Palmieri 1
1 Unit of Cancer Genetics, Institute of Biomolecular Chemistry, CNR, Sassari,
ITALY, 2 Department of Dermatology, University of Florence, Florence, ITALY,
3 Institute of Pathology, University of Sassari, Sassari, ITALY, 4 Institute of
Pathology, University of Florence, Florence, ITALY, 5 Plastic Surgery, University of
Sassari, Sassari, ITALY, 6 Melanoma, National Tumor Institute Fondazione
Pascale, Naples, ITALY
Background: A series of patients with multiple primary melanoma (MPM) were
screened for the involvement of the key-regulator genes in susceptibility (CDKN2A)
and pathogenesis (BRAF, cKIT, CyclinD1) of such a disease.
Methods: Genomic DNA from peripheral blood of 63 MPM patients (54 cases with
two primary melanomas, 8 with three, and 1 with four) were screened for germline
mutations in p16 CDKN2A and p14 CDKN2A genes by automated DNA sequencing.
Melanoma families were identified according to standardized criteria: 9 (14%)
patients were classified as familial cases. Paired synchronous and/or asynchronous
MPM tissues (N = 100) from same patients (N = 46) were analyzed for somatic
mutations in BRAF gene and FISH-based amplifications in cKIT and CyclynD1
genes.
Results: Overall, 6 (10%) different CDKN2A germline mutations were identified: 5
inp16 CDKN2A and 1 inp14 CDKN2A . The age of onset was significantly lower and the
number of primary melanomas higher in patients with mutations. CDKN2A
mutations were significantly more frequent in patients with familial history of
melanoma (5/9; 56%) compared with patients without (1/54; 2%) (P < 0.001), and in
patients with more than two melanomas (3/9; 33%) compared with patients with
only two melanomas (3/54; 6%) (P = 0.012). The debated A148T polymorphism was
found at low level (2/54; 4%) in our series. Regarding genetic alterations at somatic
level, BRAF mutations were identified in 36/100 (36%) primary melanoma tissues,
whereas amplification of cKIT and CyclinD1 genes was observed in 2/88 (2%) and
10/88 (11%) analyzed tissue samples, respectively. Considering all types of genetic
events, paired samples presented a poorly consistent distribution of somatic
alterations in same patients (52% consistency).
Conclusions: Coexistence of MPM and familial recurrence of melanoma as well as
the presence of more than two melanomas seem to be strong indications to address
patients to CDKN2A mutational screening. The low consistency in genetic patterns
of primary tumors from the same patients provide additional evidence that
pathogenetic mechanisms of melanomagenesis are heterogeneous and molecularly
different cell types may be generated in multiple primary melanoma.
Disclosure: All authors have declared no conflicts of interest.
1119P ASSOCIATION OF GENETIC POLYMORPHISMS IN TUMOR
SUPPRESSORS, ERP29 AND PTCH1, AND DNA
TRANSCRIPTION FACTORS, IKBKAP AND ZNF415, WITH
CUTANEOUS MELANOMA RISK
E.F.D. Costa, G.J. Lourenço, C. Oliveira, J.A. Rinck-Junior, A.M. Moraes,
G.A.S. Nogueira, C.S.P. Lima
Clinical Medical, State University of Campinas, Campinas, BRAZIL
Introduction: Recently, we found that genetic single nucleotide polymorphisms
(SNPs) in tumor suppressors ERP29 c.293A > G, PTCH1 g.79755C > T and g.79456C
> T, and in genes of DNA transcription IKBKAP p.Cys1072Ser and ZNF415 c.443A
> G, alter the oropharynx cancer risk. The study was conducted using high resolution
DNA microarrays genotyping (SNP 5.0 array, Affymetrix®) and the quantities and
functions of the proteins encoded by distinct alleles are being examined by our
research group. Objective: To investigate the influence of the referred SNPs in the
risk of cutaneous melanoma (CM).
Materials and methods: Genomic DNA from 153 CM patients and 153 controls
were analyzed by TaqMan® genotyping assays. The differences between groups were
analyzed by the logistic regression model. Power analysis (PA) was used to verify the
effect of sample size on the results obtained in the study.
ix364 | Abstracts Volume 23 | Supplement 9 | September 2012