This article is more than 5 years old. Information may no longer be current.
Rilotumumab fails to improve outcomes in MET-positive gastric cancer
Click Here to Manage Email Alerts
CHICAGO — The addition of rilotumumab to epirubicin, cisplatin and capecitabine chemotherapy yielded significantly shorter OS and worse toxicity in patients with advanced MET-positive gastric or gastroesophageal junction adenocarcinoma, according to the results of the phase 3 RILOMET-1 study presented at the ASCO Annual Meeting.
Rilotumumab (AMG102, Amgen) is a fully human monoclonal antibody to hepatocyte growth factor. A previous phase 2 study (Iveson T, et al. Lancet Oncol. 2014;doi:10.1016/S1470-2045(14)70023-3.) demonstrated the addition of rilotumumab to epirubicin, cisplatin and capecitabine (ECX) chemotherapy extended OS and PFS in this setting, thus prompting a phase 3 trial, according to study background.
“This study was designed to hopefully confirm the results from the phase 2 trial,” David Cunningham, MD, FRCP, head of the gastrointestinal unit at The Royal Marsden Hospital in Surrey, U.K., said during a presentation. “Our hypothesis was that rilotumumab, in combination with ECX, would significantly improve OS in the first-line treatment of MET-positive gastroesophageal junction adenocarcinoma.”
Cunningham and colleagues evaluated data from 609 patients with previously untreated advanced gastric or gastroesophageal junction adenocarcinoma in the double-blind, placebo-controlled RILOMET-1 trial. Eligibility criteria included ECOG performance score of 0 to 1, MET-positive tumor status and HER-2–negative tumor status.
Researchers randomly assigned patients 1:1 to a treatment regimen of ECX (50 mg/m2 epirubicin on day 1, 60 mg/m2 cisplatin on day 1 and 625 mg/m2 oral capecitabine on days 1-21) plus 15 mg/kg rilotumumab or placebo every 3 weeks.
Researchers stratified patients based on disease extent (locally advanced vs. metastatic) and ECOG score (0 vs. 1).
OS served as the primary outcome. PFS, 12-month survival rate, objective response rate, pharmacokinetics and safety served as secondary endpoints.
Researchers observed 128 deaths in the rilotumumab arm and 107 deaths in the placebo arm at data cutoff. Patients in the rilotumumab arm achieved a median OS of 9.6 months, whereas patients in the placebo arm achieved an 11.5-month median OS (HR = 1.37; 95% CI, 1.06-1.78).
Significantly more patients in the placebo arm also achieved 12-month OS (49.7% vs. 28.4%; P = .016) and demonstrated an objective response (39.2% vs. 30%; P = .027).
No subgroups benefitted from treatment with rilotumumab, including those with greater MET expression, Cunningham said.
A higher incidence of fatal adverse events occurred in the rilotumumab arm, including peripheral edema, hypoalbuminemia, deep vein thrombosis and hypocalcemia. Researchers attributed the higher proportion of fatal adverse events to increased disease progression in the rilotumumab arm.
The researchers closed the trial early due to the imbalance of deaths between the study arms, with MET biomarker and pharmacokinetics analyses left pending.
“The primary reason for deaths in the rilotumumab arm was disease progression,” Cunningham said. “Further analysis is clearly warranted to understand the results from this trial from a biological and clinical perspective.” – by Cameron Kelsall
Reference:
Cunningham D, et al. Abstract 4000. Presented at: ASCO Annual Meeting; May 29-June 2, 2015; Chicago.
Disclosure: The study was funded by Amgen. Cunningham reports receiving research funding from Amgen, AstraZeneca, Celgene, MedImmune, Merck Serono, Merrimack and Sanofi. Please see the abstract for a list of all other researchers’ relevant financial disclosures.
Perspective
Back to Top
Davendra Sohal, MD, MPH
The RILOMET-1 study final results were actually quite surprising in that they were completely negative. Patients who were actually selected for high MET expression based on the phase 1 and phase 2 findings — which were actually quite encouraging — found that in this phase 3 study, rilotumumab was actually inferior to placebo. Not only did it not result in benefit, it likely led to harm in patients. The results are surprising, but sobering — it is educational in that it shows how much we do need randomized phase 3 trials to move the field forward. It shows that we cannot simply rely on early phase trial results.
Perspective
Back to TopJoseph Chao, MD
The RILOMET-1 study crossed a boundary in terms of inferiority with the rilotumumab arm. I think it really speaks to the fact that — at least in targeting MET in gastric cancers — there still needs to be a lot of work done in finding the right way to measure MET as a biomarker. Cunningham and colleagues embarked on the phase 3 study based on at least some promising initial phase 2 data, in terms of looking at MET protein overexpression by immunohistochemistry. What is of interest is that they did decrease the cutoff for MET overexpression that may have diluted the proportion of tumors that are truly MET-driven. Unfortunately, there are more emerging data, at least pre-clinically, that looking at gene amplification by fluorescence in situ hybridization may be superior in selecting patients who may benefit from MET inhibitors. The researchers at least did go back and conduct an exploratory analysis to determine any benefit of adding rilotumumab to MET-amplified patients, but they still failed to see a benefit even then.
Click Here to Manage Email Alerts