Safety population indicates all randomized patients who received at least 1 dose of study treatment; intent-to-treat (ITT) population, all randomized patients. mFOLFOX6 indicates fluorouracil, leucovorin, and oxaliplatin.
Median overall survival was 11.3 months for placebo plus fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) vs 11.0 months for onartuzumab plus mFOLFOX6, and median progression-free survival was 6.8 vs 6.7 months, respectively. Dashed lines indicate median survival, and HR, hazard ratio.
Size of the data marker corresponds to the number of patients in each subgroup. ECOG indicates Eastern Cooperative Oncology Group; IHC, immunohistochemistry; MET, mesenchymal-epithelial transition; NE, not estimable.
eTable 1. Baseline Characteristics of the Intent-to-Treat Population
eTable 2. MET IHC Scoring Qualification of Central Laboratory Pathologists
eTable 3. All Grade and Grade ≥3 AEs
eFigure 1. Study design
eFigure 2. Met IHC scoring criteria and representative staining
eFigure 3. OS (A) and PFS (B) in the MET 2+/3+ subpopulation
eFigure 4. Overall Survival by MET status, region, and prior gastrectomy
Customize your JAMA Network experience by selecting one or more topics from the list below.
Shah MA, Bang Y, Lordick F, et al. Effect of Fluorouracil, Leucovorin, and Oxaliplatin With or Without Onartuzumab in HER2-Negative, MET-Positive Gastroesophageal AdenocarcinomaThe METGastric Randomized Clinical Trial. JAMA Oncol. 2017;3(5):620–627. doi:10.1001/jamaoncol.2016.5580
Does onartuzumab, a mesenchymal-epithelial transition (MET) antibody, improve efficacy outcomes in human epidermal growth factor receptor 2–negative, MET-positive gastroesophageal adenocarcinoma?
This randomized clinical trial included 562 patients (intent-to-treat population), of whom 214 had MET-positive tumors. The addition of onartuzumab to first-line fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) did not significantly improve overall survival, progression-free survival, or objective response rate of mFOLFOX6 plus placebo in the intent-to-treat or MET-positive populations.
Further research is required to better understand which patient populations may potentially derive clinical benefits from MET-targeting agents.
Dysregulation of the mesenchymal-epithelial transition (MET) signaling pathway is associated with poor prognosis in gastroesophageal adenocarcinoma (GEC). We report results of METGastric, a phase 3 trial of the MET inhibitor onartuzumab plus standard first-line chemotherapy for human epidermal growth factor receptor 2 (HER2)-negative, MET-positive, advanced GEC.
To determine whether the addition of onartuzumab to first-line fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) improves efficacy compared with mFOLFOX6 plus placebo in HER2-negative, MET-positive GEC.
Design, Setting, and Participants
Randomized, double-blind, multicenter trial conducted from November 2012 to March 2014. Patients were 18 years or older with an adenocarcinoma of the stomach or gastroesophageal junction with metastatic disease not amenable for curative therapy. Tumor samples were centrally tested for MET expression using Ventana anti-Total c-MET (SP44) rabbit monoclonal antibody, HER2 status, and Lauren histologic subtype. MET-positive tumors were defined as at least 50% of tumor cells showing weak, moderate, and/or strong staining intensity (MET 1+/2+/3+, respectively) by immunohistochemistry.
Patients with HER2-negative, MET-positive GEC were enrolled and randomized 1:1 to receive mFOLFOX6 with or without onartuzumab (10 mg/kg).
Main Outcomes and Measures
Co–primary end points: overall survival in the intent-to-treat (ITT) population and in patients with MET 2+/3+ GEC. Secondary end points: progression-free survival (PFS), overall response rate (ORR), and safety.
Enrollment was stopped early due to sponsor decision, which was agreed with an independent data monitoring committee. At the data cutoff (April 25, 2014) there were 562 patients in the ITT population (n = 283 placebo plus mFOLFOX6 [median age, 58 y; 65% male]; n = 279 onartuzumab plus mFOLFOX6 [median age, 60 y; 67% male]); 109 (38.5%) and 105 (37.6%) of the ITT population were MET 2+/3+, respectively. Addition of onartuzumab to mFOLFOX6 did not significantly improve OS, PFS, or ORR vs placebo plus mFOLFOX6 in the ITT (OS hazard ratio [HR], 0.82; 95% CI, 0.59-1.15; P = .24; PFS HR, 0.90; 95% CI, 0.71-1.16; P = .43; ORR, 46.1% vs 40.6%) or MET 2+/3+ populations (OS HR, 0.64; 95% CI, 0.40-1.03; P = .06; PFS HR, 0.79; 95% CI, 0.54-1.15; P = .22; ORR, 53.8% vs 44.6%). Safety was as expected for onartuzumab.
Conclusions and Relevance
Addition of onartuzumab to first-line mFOLFOX6 did not significantly improve clinical benefits in the ITT or MET 2+/3+ populations.
clinicaltrials.gov Identifier: NCT01662869
Gastroesophageal adenocarcinoma (GEC), which comprises tumors of the gastroesophageal junction and the stomach, is the fifth most frequently diagnosed cancer worldwide and is the third-highest cause of cancer mortality.1 The median overall survival (OS) for patients with human epidermal growth factor receptor 2 (HER2)-negative metastatic GEC treated with chemotherapy is approximately 8 to 11 months in Europe and 10 to 16 months in Asia.2 For patients with unresectable, metastatic GEC, the main therapeutic option is doublet or triplet combination chemotherapy containing a platinum-fluoropyrimidine combination, such as the mFOLFOX6 regimen (fluorouracil, leucovorin calcium, and oxaliplatin).3 Chemotherapy plus trastuzumab is standard of care for patients with HER2-positive metastatic GEC.4 Other effective therapeutics for the treatment of gastric cancer in the refractory setting include chemotherapy such as taxanes and irinotecan hydrochloride,5-8 and the vascular endothelial growth factor receptor 2 inhibitor ramucirumab (in the second-line setting).9,10 Despite all efforts, prognosis for patients with HER2-negative GEC is poor, with a 5-year OS rate of less than 10%.11
The hepatocyte growth factor (HGF) receptor that activates key oncogenic pathways through RAS, PI3K, and STAT3 plays an important role in tumorigenesis and is encoded by the mesenchymal-epithelial transition (MET) oncogene.12 Signaling through the HGF/MET pathway stimulates tissue repair and regeneration in normal tissue but can promote proliferation, survival, and metastasis in tumor cells.13,14 In GEC, dysregulation of the MET/HGF pathway is associated with poor prognosis and more aggressive disease, with MET activation stimulating tumor invasiveness.15-17 In a randomized phase 2 study in patients with advanced gastric or gastroesophageal junction cancer, nonsignificantly improved survival was observed with the anti-HGF monoclonal antibody, rilotumumab, plus epirubicin hydrochloride, cisplatin, and capecitabine vs placebo plus epirubicin hydrochloride, cisplatin, and capecitabine in patients with MET-positive tumors.18
Onartuzumab is a recombinant, fully humanized, monovalent monoclonal antibody that binds with the extracellular domain of MET. By binding with MET, onartuzumab prevents MET from binding with HGF and restricts cellular signaling via the MET pathway.19 Results of a phase 2 study demonstrated that second-/third-line onartuzumab in combination with the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib hydrochloride improved progression-free survival (PFS) and OS vs placebo plus erlotinib in patients with MET-positive (50% of tumor cells staining with an immunohistochemistry [IHC] intensity of 2+/3+) non–small-cell lung cancer (NSCLC).20
To investigate whether onartuzumab has clinical efficacy in patients with HER2-negative, MET-positive GEC, METGastric was designed to determine whether the addition of onartuzumab to first-line mFOLFOX6 improves efficacy outcomes when compared with mFOLFOX6 plus placebo.
METGastric (NCT01662869) (study protocol in Supplement 1) was a randomized, double-blind, multicenter, placebo-controlled, phase 3 study designed to evaluate the efficacy and safety of onartuzumab plus mFOLFOX6 vs placebo plus mFOLFOX6 in patients with metastatic HER2-negative and MET-positive GEC. Patients were enrolled between November 19, 2012, and March 7, 2014. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The study was approved by the local institutional review board/ethics committee at each participating site. All patients provided written informed consent.
Eligible patients were 18 years or older with an adenocarcinoma of the stomach or gastroesophageal junction with metastatic disease not amenable for curative therapy. Other eligibility criteria included Eastern Cooperative Oncology Group performance status of 0 or 1, measurable disease or nonmeasurable but evaluable disease, adequate organ function, and no previous treatment for metastatic disease. Only patients with HER2-negative, MET-positive (as assessed by IHC; score of 1+, 2+, or 3+) GEC were enrolled. Key exclusion criteria included a history of HER2-positive tumors, previous chemotherapy for locally advanced or metastatic gastric carcinoma, a significant history of cardiac disease, peripheral neuropathy, and/or active (significant or uncontrolled) gastrointestinal bleeding.
Patients were randomized using a permutated block method in a 1:1 ratio to receive placebo plus mFOLFOX6 (oxaliplatin 85 mg/m2 intravenous [IV], day 1, fluorouracil 400 mg/m2 IV bolus followed by 2400 mg/m2 over 48 hours starting on day 1, leucovorin 400 mg/m2 IV, day 1) or onartuzumab (10 mg/kg IV on day 1, every 14 days) plus mFOLFOX6 for a maximum of 12 cycles (eFigure 1 in Supplement 2). Patients whose disease had not progressed after 12 cycles could continue treatment with either onartuzumab or placebo until disease progression, unacceptable toxic effects, or death. Stratification was performed by MET IHC expression status score (1+ vs 2+/3+), region (Asia-Pacific vs other), and prior gastrectomy (yes vs no). The first dose of study drug had to be administered within 3 days of randomization. No modifications of the onartuzumab/placebo dose were allowed. If onartuzumab/placebo was discontinued due to tolerability, patients could continue mFOLFOX6 treatment (if onartuzumab/placebo was discontinued during the first 12 cycles of study treatment) or fluorouracil/folinic acid if agreed on by the investigator and patient.
The provision of tumor samples (archival tissue block or 15 serial cut unstained slides) was mandatory. Tumor samples were tested centrally (Targos, Germany) to determine MET expression status (using the validated Ventana anti-Total c-MET [SP44] rabbit monoclonal antibody IHC assay21), HER2 status (using the Ventana anti-HER2 [4B5] rabbit monoclonal antibody IHC assay), and the Lauren histologic subtype status. To be eligible for the study, patients were considered positive if their tumor samples demonstrated at least 50% of cells staining positive with an intensity of 1+ or above using the Ventana IUO assay system guidelines. A MET IHC-positive subgroup, for the purpose of testing the MET hypothesis, was defined as those patients with at least 50% of tumor cells showing moderate and/or strong staining intensity (MET 2+/3+).
The co-primary end points were OS in the intent-to-treat (ITT) population and in the subgroup of patients with MET 2+/3+ GEC. Secondary end points included PFS, overall response rate (ORR), disease control rate, duration of response, patient-reported outcomes, and safety.
Tumor response and progression were assessed according to the Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1. Evaluation of tumor response was documented every 6 weeks for the first 12 months. For patients who continued to receive study treatment and were followed up for more than 12 months, chest, abdomen, and pelvis computed tomography or magnetic resonance imaging scans were taken every 12 weeks. Overall survival was defined as the time from randomization to death from any cause. Progression-free survival was defined as the time from randomization to the first occurrence of disease progression, as determined by investigator-assessed RECIST v1.1 or death from any cause, whichever occurred first. Duration of response was defined as the time from first occurrence of a documented objective response to disease progression as determined by investigator-assessed RECIST v1.1, or death from any cause during the study. Disease control rate comprised complete response, partial response, and stable disease according to RECIST. Adverse events (AEs) were graded according to the National Cancer Institute Common Terminology Criteria for AEs, version 4.0. An independent data monitoring committee (IDMC) monitored patient safety data approximately every 6 months during the course of the study.
The study was designed to enroll approximately 800 patients and was powered to demonstrate an improvement of median OS from 9.0 to 12.3 months in the ITT population (hazard ratio [HR], 0.73) and 9 to 18 months in the MET 2+/3+ population (HR, 0.49).
The ITT population included all randomized patients, and the safety population included all randomized patients who received at least 1 dose of study treatment. Kaplan-Meier methodology was used to estimate median PFS and median OS for each treatment arm. A stratified Cox regression model was used to estimate HR and 95% confidence intervals of PFS and OS. The 2 treatment comparisons of OS were based on a stratified log-rank test at 1-sided nominal significance level of .00577 for the MET 2+/3+ subgroup and a 1-sided nominal significance level of .02 for the ITT population, using the stratification factors specified.
Enrollment was stopped early due to sponsor decision, which was agreed with the IDMC, due to a lack of efficacy in a phase 2 trial also assessing mFOLFOX6 plus onartuzumab (NCT01590719).22 Patients were given the option of continuing treatment with onartuzumab after enrollment was stopped. The rationale to conduct concurrent phase 2/3 studies was to use data from the phase 2 study to inform the final MET IHC cutoff to be applied to the phase 3 study, and was endorsed by both the US Food and Drug Administration and the European Medicines Agency.
At the data cutoff (April 25, 2014), the ITT population comprised 562 patients, of whom 283 received placebo plus mFOLFOX6 and 279 were allocated to onartuzumab plus mFOLFOX6 (Figure 1). Baseline characteristics between treatment arms were balanced (eTable 1 in Supplement 2). In total, there were 109 (38.5%) and 105 (37.6%) patients with MET 2+/3+ GEC in the placebo plus mFOLFOX6 and onartuzumab plus mFOLFOX6 groups, respectively.
At the data cutoff, in the ITT population, 74 (26.1%) patients in the placebo plus mFOLFOX6 group and 72 (25.8%) patients in the onartuzumab plus mFOLFOX6 group had OS events (defined as death from any cause); 141 (49.8%) and 132 (47.3%) patients had PFS events in the placebo plus mFOLFOX6 and onartuzumab plus mFOLFOX6 groups, respectively. The addition of onartuzumab to mFOLFOX6 did not result in significant differences in OS or PFS compared with placebo plus mFOLFOX6 (Figure 2A and B). Median OS was 11.3 months for placebo plus mFOLFOX6 vs 11.0 months for onartuzumab plus mFOLFOX6 (HR, 0.82; 95% CI, 0.59-1.15; stratified P = .24), and median PFS was 6.8 vs 6.7 months (HR, 0.90; 95% CI, 0.71-1.16; stratified P = .43), respectively.
Subgroup analyses of OS in the ITT population showed no difference between onartuzumab and placebo in any subgroup (Figure 3). There was no difference in ORR between treatment arms in the ITT population for onartuzumab plus mFOLFOX6 compared with placebo plus mFOLFOX6 (100 patients [46.1%] vs 84 patients [40.6%]; P = .25) (Table 1). Four patients in each group achieved a complete response.
MET IHC comprised both staining intensity and the percentage of MET-expressing tumor cells (eFigure 2 in Supplement 2). MET IHC–positive tumors were defined as those with at least 50% of tumor cells showing moderate and/or strong staining intensity (MET 2+/3+). Overall agreement between central laboratory pathologists was more than 95% (eTable 2 in Supplement 2).
In the MET 2+/3+ subgroup, 41 (37.6%) patients in the placebo plus mFOLFOX6 group and 35 (33.3%) of patients in the onartuzumab plus mFOLFOX6 group had OS events at the data cutoff. The addition of onartuzumab to mFOLFOX6 did not significantly improve efficacy outcomes compared with placebo in this subgroup (eFigure 3 in Supplement 2). Median OS was 9.7 months for placebo plus mFOLFOX6 vs 11.0 months for onartuzumab plus mFOLFOX6 (HR, 0.64; 95% CI, 0.40-1.03; stratified P = .06); median PFS was 5.7 vs 6.9 months (HR, 0.79; 95% CI, 0.54-1.15; stratified P = .22), respectively.
Post hoc exploratory subgroup analyses of OS by region and prior gastrectomy are shown in eFigure 4 in Supplement 2. In the MET 2+/3+ subgroup, non-Asian patients without prior gastrectomy (n = 125) seemed to derive a clinical benefit from onartuzumab plus mFOLFOX6 (HR, 0.51; 95% CI, 0.29-0.89). There was no difference in ORR between the treatment arms in the MET 2+/3+ subgroup for onartuzumab plus mFOLFOX6 compared with placebo plus mFOLFOX6 (42 patients [53.8%] vs 41 patients [44.6%]; P = .23) (Table 1). Three patients in the MET 2+/3+ subgroup achieved a complete response in the onartuzumab plus mFOLFOX6 group compared with none in the placebo plus mFOLFOX6 group.
Overall, 273 (97.5%) patients in the placebo plus mFOLFOX6 group and 274 (98.2%) patients in the onartuzumab plus mFOLFOX6 group experienced at least 1 AE. Serious AEs were more frequent with onartuzumab than with placebo (100 [35.8%] vs 91 [32.5%], respectively), but the difference between the treatment arms was not statistically significant (Table 2). All-grade AEs and grade 3 and above AEs are summarized in eTable 3 in Supplement 2. Grade 3 and above AEs more commonly observed with onartuzumab included neutropenia (98 [35.1%] vs 82 [29.3%]), hypoalbuminemia (16 [5.7%] vs 1 [0.4%]), peripheral edema (13 [4.7%] vs 1 [0.4%]), thrombocytopenia (12 [4.3%] vs 3 [1.1%]), pulmonary embolism (17 [6.1%] vs 10 [3.6%]), and gastric perforation (2 [0.7%] vs 0).
Further to discussion with the IDMC, enrollment into this trial was terminated prematurely. The addition of onartuzumab to first-line mFOLFOX6 did not improve OS, PFS, or ORR compared with mFOLFOX6 alone in patients with HER2-negative GEC, irrespective of MET expression status. These negative results are in line with other phase 2/3 trials that have reported similarly disappointing results with onartuzumab, including in triple-negative breast cancer,23 recurrent glioblastoma,24 and NSCLC25 after initial phase 2 data in second-line/third-line NSCLC were promising.20
Several hypotheses have been advanced to explain the failure of onartuzumab, as well as other MET-directed drugs, in GEC and other solid tumors.26,27 One explanation is that MET overexpression might not be an appropriate target for onartuzumab in the majority of patients with GEC. Dysregulation of the MET pathway in GEC can arise from MET amplification, or rare MET mutations affecting the extracellular domain (involving the HGF binding domain), the tyrosine kinase domain (leading to a constitutive MET activation), or the juxtamembrane domain (causing a disruption in negative regulation of MET signaling).26,27 Targeting one of these alternative MET pathways may lead to improved clinical outcomes for patients with GEC. For example, because MET-amplified tumors are likely to signal primarily via ligand-independent mechanisms, a small molecule targeting the kinase domain could potentially prove beneficial in these cases.
An alternative hypothesis for the lack of efficacy with onartuzumab in GEC is that overexpression of HGF in the tumor cells or stroma may lead to autocrine or paracrine loop formation.27 In contrast to the situation of oncogene addiction, this inappropriate MET signaling is a consequence, rather than the cause, of the cell transformation, and targeting MET signaling in these tumors will not fundamentally affect tumor behavior or cancer outcomes.26 However, this does leave open the possibility that some tumors may still be dependent on sustained MET signaling for their growth and survival and may therefore be sensitive to MET blockade. Identification of these tumors would require an alternate biomarker to identify tumors that are driven by MET signaling. Other possible explanations for why MET inhibition alone may not yield an improvement in survival in GEC are the potential redundancy of signaling via the MET pathway,28,29 or because the MET IHC assay may not select appropriately for MET-driven tumors. Predictive biomarkers are sought that can more accurately identify patients who are most likely to gain clinical benefit from onartuzumab treatment.
Although METGastric failed to meet its co–primary end points, further research is required to better elucidate which patient populations may potentially derive clinical benefits from onartuzumab.
Accepted for Publication: October 5, 2016.
Corresponding Author: David Cunningham, MD, Royal Marsden Hospital, Fulham Rd, London SW3 6JJ, England (firstname.lastname@example.org).
Published Online: December 1, 2016. doi:10.1001/jamaoncol.2016.5580
Author Contributions: Drs Shah and Cunningham had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Bang, Lordick, Hack, Bruey, Smith, McCaffery, Phan, Cunningham.
Acquisition, analysis, or interpretation of data: Shah, Bang, Lordick, Alsina, Chen, Hack, McCaffery, Shames, Phan.
Drafting of the manuscript: Shah, Bang, Smith, McCaffery, Shames, Phan, Cunningham.
Critical revision of the manuscript for important intellectual content: Shah, Bang, Lordick, Alsina, Chen, Hack, Bruey,
McCaffery, Shames, Phan, Cunningham.
Statistical analysis: Chen, Hack, Phan.
Obtained funding: Bruey, Phan.
Administrative, technical, or material support: Shah, Hack, Shames, Phan.
Supervision: Lordick, Cunningham.
Conflict of Interest Disclosures: Dr Shah has acted in a consulting or advisory role to Lilly Inc, and received research funding from Berg Pharma, Merck, Roche, and Sanofi. Dr Bang has received honoraria from, acted in a consulting or advisory role to, and received research funding from Roche/Genentech. Dr Lordick has received honoraria from Amgen, Bristol-Myers Squibb, Lilly, Merck Sharp & Dohme, Nordic Group, and Roche Pharma AG, acted in a consulting or advisory role to BioNTech AG, Ganymed Pharmaceuticals, and Roche, received research funding from Boehringer Ingelheim, Fresenius Biotech, GlaxoSmithKline, and Merck Serono, and travel, accommodation, or other expenses from Bayer, Roche, and Taiho Pharmaceutical. Dr Chen is an employee of Roche/Genentech. Dr Hack is an employee of, and holds stock in, Genentech. Dr Bruey is an employee of Genentech and holds stock in Roche Holding. Dr Smith is an employee of, and holds stock in, Genentech, and has received travel, accommodation, or other expenses from Genentech. Dr McCaffery was an employee of Genentech at the time this work was conducted, he holds stock in Roche, has patent or intellectual property interest in Amgen, Celera Genomics, and Genentech, and has received travel, accommodation, or other expenses from Corvus Pharma and Genentech. Dr Shames is an employee of Genentech, holds stock in Roche Holding, and has patent or intellectual property interest in UT Southwestern. Dr Phan is an employee of Genentech. Prof Cunningham has received research funding from Amgen, AstraZeneca, Bayer, Celgene, Medimmune, Merck Serono, Merrimack, and Sanofi. No other disclosures are reported.
Funding/Support: This study was sponsored by F. Hoffmann-La Roche Ltd. David Cunningham is supported by the National Institute for Health Research Biomedical Research Centre based at the Royal Marsden Hospital, London, England, and the Institute of Cancer Research, London, England.
Role of Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Previous Presentation: Preliminary results of this study were presented at the American Society of Clinical Oncology annual meeting; May 30, 2015; Chicago, Illinois.
Additional Contributions: Third-party medical writing assistance, under the direction of the authors, was provided by Fiona Fernando, PhD, of Gardiner-Caldwell Communications, and was funded by F. Hoffmann-La Roche Ltd.