Annual Meeting Proceedings Part 1 - American Society of Clinical ...

Developmental Therapeutics—Clinical Pharmacology and Immunotherapy
2593 General Poster Session (Board #8B), Mon, 8:00 AM-12:00 PM
Phase I and pharmacokinetic (PK) study of pazopanib (P) in combination
with two schedules of ifosfamide (I) in patients (pts) with advanced solid
tumors (STs). Presenting Author: Paul Hamberg, Erasmus University
Medical Center, Daniel den Hoed Cancer Center, Rotterdam, Netherlands
Background: P is a tyrosine kinase inhibitor of PDGFR, VEGFR and KIT. This
study aimed to determine the recommended phase II dose (RP2D) of P
combined with I. To assess the impact of scheduling on tolerability, two I
schedules were explored. Methods: Pts with progressive ST, no standard
therapy available, PS 0-1 and good organ function were eligible. Pts
received daily P with 3-weekly I, 9 g/m2 /cycle, either continuously (Iciv) or
as 3 hrs boluses (Ibolus) for 3 days. P was escalated in serial cohorts of 3-9
pts per I-schedule. If in 3-6 pts dose-limiting toxicity (DLT) of 33.3%
within 1st cycle was seen 3 more pts at that dose level (DL) were enrolled.
RP2D was exceeded if in � 1/3 pts, � 2/6 pts or � 3/9 pts DLT occurred. P
was maximized at 1000 mg/day. At RP2D, an expansion cohort of 6-9 pts
was studied to confirm safety. PK samples of P and I were collected in all
pts. Sampling allowed intra-individual comparison of I-PK with or without P
and vice versa. Results: 47 pts were studied. For Iciv (25 pts), RP2D was
the maximum dose of P (1000 mg), higher than the single agent RP2D. At
RP2D 2 DLTs in 15 pts occurred: 1 febrile neutropenia (FN); 1 encephalopathy.
With Ibolus (22 pts), P 400 mg with or without GCSF was not tolerated
(5 DLTs in 10 pts: 3*FN, 1 encephalopathy; 1 proteinuria). At RP2D (P
200 mg � Ibolus � GCSF) 3 DLTs in 12 pts (FN; encephalopathy, renal
insufficiency) were observed. Toxicity occurring in �30% of pts in Iciv (all
grades/grade 3-4,%) were neutropenia (92/92) anemia (92/8); thrombocytopenia
(35/12), elevated ASAT, ALAT, alkaline phosphatase (73/4; 73/0;
62/2), nausea (58/0), vomiting (73/0), diarrhea (54/0) and hypertension
(31/12). 4/25 pts in Iciv evaluable for response had PR (2 sarcoma, 2
ovary) and 10 prolonged (�3 mths) SD. In Ibolus, 6/20 pts had PR (2
urothelial, 1 sarcoma, 1 ovary, 1 CUP, 1 mesothelioma) and 3/20
prolonged SD. Preliminary PK analysis showed no effect of P on I-exposure,
neither of Ibolus or Iciv on P. Conclusions: P is tolerated at a higher dose
combined with Iciv compared to Ibolus (1000 vs 200 mg). PK cannot
explain this difference. This study underlines the impact of scheduling on
the tolerability of VEGFR-TKI and cytotoxic drug combinations.
2595 General Poster Session (Board #8D), Mon, 8:00 AM-12:00 PM
Pharmacokinetic modeling and simulation (PK M&S) supported dose
escalation of PF-03446962, a monoclonal antibody (mAb) against activin
receptor like kinase 1, in patients with solid tumors. Presenting Author:
Erjian Wang, Pfizer Inc., San Diego, CA
Background: PF-03446962 is a first in class mAb. The PK of mAbs is
complex, as their clearance (CL) can be linear or non-linear, besides
extremely slow. Caution must be exercised when designing mAbs’ dose
escalation schemes. Along with safety and activity evaluation, we used PK
M&S in real time to ensure that subsequent recommended doses did not
expose too many patients to subtherapeutic levels while preserving safety.
Methods: The model incorporated both Michaelis-Menten approximation
and linear components. A prespecified target exposure was defined by
preclinically projected therapeutic and toxic concentrations. The PK M&S
results were used to make dose escalation decisions. Results: Eight dose
levels (0.5, 1, 2, 3, 4.5, 6.75, 10, and 15 mg/kg) were studied in 44
patients. CL of PF-03446962 exhibited nonlinear characteristics at low
doses and linear at high doses. CL was saturated at dose levels of 3 mg/kg
and above. Mean half-life was 2.6 days for the first dose level and 12-15
days after CL saturation. The dose ratio of 15 to 0.5 mg/kg was 30, while its
corresponding Cmax and AUC ratios were 46 and 97, respectively.
However, the PK M&S accurately predicted exposure following each dose.
95% of the observed drug concentrations across all dose levels were within
the predicted values. Concentrations at the 4.5 mg/kg dose exceeded the
predicted efficacious level over the dosing period. None of 44 patients in
the study were exposed to the toxic level derived from a study in monkeys.
PK data also implied a biologically active dose at 2 mg/kg dose and above,
confirmed by observing target toxicities (ie, telangiectasia) and clinical
tumor responses. One patient with PR was at the 2 mg/kg dose and 2
patients with PR were at the 10 mg/kg. The PK M&S well predicted the drug
exposure following multiple doses in patients, indicating that patients did
not develop antidrug immunogenicity. Conclusions: The PK M&S based
approach ensured that the majority of patients were treated at or near the
therapeutic dose. It can be readily implemented in the conduct of dose
escalation trials of antitumor mAbs, while minimizing risk to patients and
maximizing efficiency.
165s
2594 General Poster Session (Board #8C), Mon, 8:00 AM-12:00 PM
Modeling the combined efficacy of rilotumumab (R; AMG 102) plus
epirubicin/cisplatin/capecitabine (ECX) for the treatment of locally advanced
or metastatic gastric or esophagogastric junction (G/EGJ) cancer.
Presenting Author: Sameer Doshi, Amgen Inc., Thousand Oaks, CA
Background: R is an investigational, fully human monoclonal antibody to
hepatocyte growth factor/scatter factor (HGF/SF) that inhibits signaling
though the MET receptor. In a double-blind, placebo-controlled, phase 2
trial of 121 patients with G/EGJ cancer, R (7.5 or 15 mg/kg) � ECX showed
trends for improved progression-free survival (PFS) and overall survival
compared to ECX alone. The combined effects of R and ECX on tumor
reduction and PFS were modeled using the phase 2 data. Methods: A
population pharmacokinetic (PK) model was used to estimate R PK
parameters and individual R concentrations (C) over time. A tumor dynamic
model was developed to characterize the combined effects of R with ECX on
tumor growth and cell death and to evaluate the impact of baseline patient
characteristics and lab values on model parameters. A discrete timesurvival
model was developed with PFS and C data to evaluate the effect of
R, ECX, and the interaction between R and ECX on PFS within a given time
interval. Results: R exhibited linear PKs with an estimated mean clearance
of 0.216 L/d/70 kg. The tumor dynamic model suggested that R and ECX
worked jointly to reduce tumor size from baseline via inhibition of the tumor
growth rate constant (Kgrowth) and stimulation of the death rate of tumor
cells (Kdeath). Assuming the maximal inhibition of Kgrowth is 100%, the
estimated mean C that provided 50% maximal Kgrowth inhibition (EC50) was
6.71 �g/mL. The mean (SD) estimated effect of 7.5 and 15 mg/kg R on
K was 90.1% (3.8%) and 95.5% (1.9%) inhibition, respectively. None
growth
of the tested baseline factors appeared to significantly affect tumor
reduction by R � ECX. In the discrete time-survival model, the R and ECX
combination significantly improved PFS, and the model suggests that the
magnitude of the effect of one treatment was dependent on the other.
Conclusions: R and ECX appeared to work in combination to decrease tumor
size and to improve PFS.
2596 General Poster Session (Board #8E), Mon, 8:00 AM-12:00 PM
Individual PK-guided sunitinib dosing: A feasibility study in patients with
advanced solid tumors. Presenting Author: Nienke A.G. Lankheet, Slotervaart
Hospital, Amsterdam, Netherlands
Background: Sunitinib treatment shows a clear dose-efficacy relationship.
However, due to large inter-individual variations in plasma exposure, target
total trough levels (� 50 ng/mL) are frequently not reached with the current
dosing schedule. Therefore, a prospective Phase I study was performed to
determine the safety and feasibility of PK-guided dosing of sunitinib.
Methods: Thirty patients with solid tumors with a potential benefit from
sunitinib treatment were included. Patients were treated continuously with
sunitinib 37.5 mg once daily. At day 15 and 29 of sunitinib treatment,
plasma trough levels of sunitinib and its active metabolite (N-desethyl
sunitinib) were measured. If the total trough level (TTL) was � 50 ng/mL
and the patient did not show any � grade 3 toxicity (CTCAE 4.02), the daily
sunitinib dose was increased by 12.5 mg. If the patient suffered from
� grade 3 toxicity, the sunitinib dose was lowered by 12.5 mg. After 8
weeks a final TTL evaluation was performed. Results: All 30 patients started
treatment and in 18 patients all 3 TTLs were measured at the time of this
analysis. Given TTL � 50 ng/mL and � grade 3 toxicity, a first sunitinib
dose increase could be applied in 10 patients and a second dose increase
in 2 patients at day 15 and 29, respectively. At day 15, day 29 and 8 weeks
of treatment, mean TTLs were 49.7 ng/mL (coefficient of variation (CV)
27.7%), 62.4 ng/mL (CV 35.5%) and 56.7 ng/mL (CV 47.8%), respectively.
At day 15, using the sunitinib standard dose of 37.5 mg, target TTLs
were reached in 8 patients (44%). At 8 weeks, the mean sunitinib daily
dose intensity was increased to 40.3 mg (standard deviation (SD) � 11.0).
At the end of the 8 week study period, 5 out of 18 patients (28%) were still
at an increased sunitinib dose level without additional side effects. In these
patients the mean daily sunitinib dose was 55.0 mg (SD � 6.8) and their
final TTLs were all � 50 ng/mL (mean: 70.8 ng/mL (CV 29.8%)). Using
PK-guided sunitinib dosing, target TTLs were reached in 10 out of 18
patients (56%) at the final TTL evaluation, compared to 44% at day 15
before any dose adjustment. Conclusions: Individual PK guided dosing is
feasible and enables a significant increase in sunitinib dose intensity
without causing additional toxicity.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.