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

240s Gastrointestinal (Noncolorectal) Cancer
4004 Oral Abstract Session, Sat, 3:00 PM-6:00 PM
Optimum time to assess complete clinical response (CR) following chemoradiation
(CRT) using mitomycin (MMC) or cisplatin (CisP), with or without
maintenance CisP/5FU in squamous cell carcinoma of the anus: Results of
ACT II. Presenting Author: Robert Glynne-Jones, Mount Vernon Centre for
Cancer Treatment, Middlesex, United Kingdom
Background: In the ACT I trial, relapse-free survival was improved with CRT
compared to RT alone, but the CR rate at 6 weeks was similar (39% CRT,
30% RT alone; p�0.08). Most studies assessed CR between 6 and 12
weeks following completion of CRT. We investigated the association
between observation of CR at 3 different time-points and progression-free
and overall survival (PFS, OS), to determine the optimal time to assess this
early endpoint. Methods: ACT II recruited 940 pts between 2001 and
2008. It compared, in a factorial design, CisP versus MMC when combined
with 5-FU CRT, and two cycles of maintenance chemotherapy versus no
maintenance. CR (complete absence of tumour and node negative) was
assessed by rectal exam or imaging at 11, 18 and mandated CT at 26 weeks
after the start of CRT. Median follow-up was 5 years. Pts were excluded
from analysis if assessment was not done/ missing at one or more time
points (n�245). Data from 695 pts were analysed. Results: Pt characteristics
(all 940) median age 58 years; tumour site – canal (84%), margin
(14%); stage T1-T2 (52%), T3-T4 (46%); N� (32%), N0 (62%). CisP did
not improve OS (HR: 0.96, p�0.89) or PFS (HR: 0.85, p�0.43) compared
to MMC. This is consistent with the lack of an absolute difference at 18 and
26 weeks, but not at 11 weeks (where a difference was found). CR patients
(compared to not-CR) had a significantly lower risk of progression/death.
The association between these outcomes and response was strongest at 26
weeks. 202/695 (29%) pts not in CR at 11 weeks were CR at 26 weeks.
Conclusions: 29% of pts not in CR at 11 weeks achieved CR at 26 weeks.
Early surgical salvage would not have been appropriate for these pts. We
recommend assessment at 26 weeks in future trials.
Absolute risk
HR (95% CI)
difference (95% CI)
(CR vs not-CR)
Pts with CR CR rate % MMC CisP PFS OS
Week 11 429 65.6 57.9 7.7% (0.52, 14.9) 0.74 (0.56, 0.97) 0.68 (0.48, 0.97)
p�0.04
p�0.03
p�0.03
Week 18 527 75.4 76.2 0.8% (-7.2, �5.6) 0.54 (0.40, 0.73) 0.44 (0.31, 0.64)
p�0.81
p�0.001
p�0.001
Week 26 582 83.5 84.0 0.5% (-5.9, �5.1) 0.26 (0.19, 0.35) 0.23 (0.16, 0.33)
p�0.88
p�0.001
p�0.001
4006 Oral Abstract Session, Sat, 3:00 PM-6:00 PM
Tivantinib (ARQ 197) versus placebo in patients (Pts) with hepatocellular
carcinoma (HCC) who failed one systemic therapy: Results of a randomized
controlled phase II trial (RCT). Presenting Author: Lorenza Rimassa,
Department of Oncology, Humanitas Cancer Center, Rozzano, Italy
Background: Tivantinib (T) is a selective, oral inhibitor of c-Met, the tyrosine
kinase receptor for hepatocyte growth factor involved in tumor cell
migration, invasion, proliferation and angiogenesis. T has shown promising
results in HCC in phase 1 studies as monotherapy and in combination with
sorafenib. Methods: This multi-center RCT, enrolled pts with unresectable
HCC, 1 failed systemic therapy, ECOG PS �2. Child-Pugh B-C were
excluded. Pts were randomized 2:1 to oral T {360 mg bid (A), 240 mg bid
(B)} or placebo (P), stratifying by PS and vascular invasion (VI). Treatment
continued until disease progression (PD) or unacceptable toxicity. RECIST
1.1 response was evaluated by CT / MRI every 6 weeks. Crossover to
open-label T was allowed after PD. Primary endpoint was time to tumor
progression (TTP) in the intent-to-treat (ITT) population by central radiology
review. Other endpoints included disease control rate (DCR), PFS, OS,
efficacy in Met� (Met � 2� in �50% of tumor at immunohistochemistry)
pts, safety. Results: Characteristics of the 107 enrolled HCC pts (A� 38,
B� 33, P� 36) in T/P: 58/28 male; median age 70/68; PS 0 41/21; VI
22/13; Met� 22/15; Met- 27/13. Dose A was reduced to B in all pts due to
G�3 neutropenia (NEUT) rate. Major TTP, DCR and PFS benefits were
obtained in Met� pts, with preliminary OS trend favoring T (HR�0.47) and
no detrimental effect in Met- pts. DCR (95% CI) in T/P was 44 (31-56) /
31(16-48)% for ITT and 50 (28-72) / 20 (4-48)% in Met� pts. Most
common AEs in T were asthenia (26.8%), NEUT (25.4%), low appetite
(25.4%); most common drug-related AEs were NEUT (25.4%), anemia
(15.5%). Most frequent drug-related serious AE was neutropenic sepsis
(4.2%). Efficacy was similar inA/Bwith less frequent NEUT in B (21.1% /
6.1%). Final OS, PK, biomarker data will be presented. Conclusions:
Compared to P, T significantly benefited second-line HCC pts, especially if
Met�, with manageable safety profile at 240 mg BID.
Median:
TTP (mos)
T arm P arm HR CI* Log-rank p value
ITT 1.6 1.4 0.64 0.43-0.94 0.04
Met �
PFS (mos)
2.9 1.5 0.43 0.19-0.97 0.03
ITT 1.7 1.5 0.67 0.44-1.04 0.06
Met � 2.4 1.5 0.45 0.21-0.95 0.02
*90%CI for primary TTP. 95%CI for other endpoints.
4005 Oral Abstract Session, Sat, 3:00 PM-6:00 PM
Evaluation of MET pathway biomarkers in a phase II study of rilotumumab
(R, AMG 102) or placebo (P) in combination with epirubicin, cisplatin, and
capecitabine (ECX) in patients (pts) with locally advanced or metastatic
gastric (G) or esophagogastric junction (EGJ) cancer. Presenting Author:
Kelly S. Oliner, Amgen Inc., Thousand Oaks, CA
Background: R is an investigational, fully human, monoclonal antibody to
HGF/SF, the ligand of the MET receptor. A double-blind, P-controlled,
phase 2 study randomized 121 G/EGJ cancer pts 1:1:1 to 15 mg/kg R �
ECX (Arm A, n � 40), 7.5 mg/kg R � ECX (Arm B, n � 42), or P � ECX (Arm
C, n � 39). OS and PFS improved with the addition of R to ECX (Iveson et
al, European Multidisciplinary Cancer Congress 2011; abstr 6.504). High
tumor MET levels have been associated with poor prognosis in G cancer
(Nakajima et al, Cancer 1999;85:1894-1902). We explored MET pathway
biomarkers to identify pts who may benefit from R. Methods: MET protein
levels and gene copy numbers were measured in archival tumor samples by
immunohistochemistry (IHC) and fluorescence in situ hybridization, respectively.
High and low MET subgroups were defined by several predefined
strategies. Total HGF and soluble MET (sMET) protein in plasma were
measured by ELISA and MSD assays, respectively. Treatment and biomarker
effects on OS and PFS were analyzed by Cox proportional hazard
models and Kaplan-Meier estimates. Results: Tumor samples were evaluable
for MET protein from 62 pts in Arms A � B and 28 pts in Arm C. Pts
with METHigh tumors (� 50% tumor cells positive) in Arms A � B had
improved median OS than those in Arm C (11.1 mo [80% CI: 9.2-13.3] vs
5.7 mo [80% CI: 4.5-10.4]; HR � 0.29, 95% CI: 0.11-0.76, p � 0.012).
Conversely, pts with METLow tumors (� 50% positive) in Arms A � B had a
trend toward unfavorable OS compared with those in Arm C (HR � 1.84,
95% CI: 0.78-4.34). In the chemotherapy only arm (Arm C), pts with
METHigh tumors had poorer OS (HR � 3.22, 95% CI: 1.08-9.63) than pts
with METLow tumors. Similar trends were seen with PFS. Predefined
dichotomization schemes for tumor MET gene copy number and baseline
plasma levels of total HGF or sMET did not correlate with OS or PFS.
Conclusions: High MET expression by IHC may predict clinical benefit to R
� ECX in G/EGJ cancer pts. High MET expression may also be associated
with poor prognosis for ECX-treated pts. Further investigation is needed to
confirm these findings.
4007 Oral Abstract Session, Sat, 3:00 PM-6:00 PM
Activity of cabozantinib (XL184) in hepatocellular carcinoma: Results from
a phase II randomized discontinuation trial (RDT). Presenting Author: Chris
Verslype, Hepatology, University Hospitals Gasthuisberg, Leuven, Belgium
Background: Cabozantinib (cabo) is an oral, potent inhibitor of MET and
VEGFR2. MET over-expression has been observed in advanced hepatocellular
carcinoma (HCC). Anti-VEGF pathway agents have shown clinical
benefit in pts w/ HCC. Simultaneous targeting of the MET and VEGF
signaling pathways with cabo may therefore be a promising treatment
strategy. Methods: Eligible HCC patients (pts) were required to have
measurable disease per RECIST, � 1 prior systemic regimen and Child-
Pugh score of A. Pts received cabo at 100 mg qd over a 12 wk Lead-in
stage. Tumor response (mRECIST) was assessed q6 wks. Treatment � wk
12 was based on response: pts with PR continued open-label cabo, pts with
SD were randomized to cabo vs placebo, and pts with PD discontinued.
Primary endpoint in the randomized phase was progression free survival
(PFS). The primary endpoint was overall response rate (RR) per mRECIST
in the Lead-in stage. Results: Enrollment has been completed (n � 41); all
pts are unblinded. Median age: 61 years (33 to 83). Males: 76%; Asian:
37%. HCC etiology: Hep B 24%; Hep C 22%; alcohol abuse 20%; other
38%. Extra-hepatic spread observed in 70%. Median number of prior
systemic treatments was 1; prior sorafenib was 51%. Median baseline AFP
was 368 ng/mL (3 – 259,298); 86% had elevated AFP at baseline. Median
follow-up was 5.5 mos (0.8 -18.5). 29 pts (71%) completed the Lead-in
stage. Median PFS from Study Day 1 was 4.2 mos. 2/36 pts evaluable for
tumor assessment at 12 weeks achieved a confirmed PR (cPR) by original
RECIST (RR 5%). One more pt randomized at Week 12 achieved a cPR at
18 weeks. 28/36 pts (78%) with �1 post-baseline scan had tumor
regression (with no apparent relationship to prior sorafenib therapy). The
overall disease control rate (DCR � PR�SD) at Week 12: was 68% (Asian
subgroup: 73%). AFP responses (defined as reduction from baseline of
�50% in pts with elevated AFP at baseline) in 26 pts with �1 postbaseline
result: 10/26 (38%). Most common Gr 3/4 AEs: diarrhea (17%),
palmar-plantar erythrodyesthesia (15%), and thrombocytopenia (10%).
Conclusions: Cabo treatment exhibits activity in HCC pts regardless of prior
sorafenib treatment. The safety profile was comparable to that of other
VEGFR TKIs.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.