Adding bemarituzumab to chemotherapy as first-line treatment for advanced gastric and gastroesophageal junction (GEJ) cancers improved survival over chemotherapy alone in the phase 2 FIGHT trial.
Among 155 patients followed for a median of 10.9 months, the combination of bemarituzumab and modified FOLFOX chemotherapy (leucovorin, fluorouracil, and oxaliplatin) resulted in a median progression-free survival (PFS) of 9.5 months, compared with 7.4 months with chemotherapy alone.
The median overall survival (OS) was not reached in the combination arm but was 12.9 months for patients treated with modified FOLFOX plus placebo.
Wainberg explained that bemarituzumab is an IgG1 antibody targeted specifically to the FGFR2b receptor to block growth factor signaling. The findings from FIGHT suggest FGFR2b is a new biomarker and therapeutic target for advanced gastric and GEJ cancers.
“The FIGHT trial results support a prospective, randomized, phase 3 study in gastric and gastroesophageal adenocarcinoma and the evaluation of bemarituzumab in other FGFR2b tumor types,” Wainberg said.
FGFR2b is a splice isoform of FGFR2 that is thought to be overexpressed in anywhere from 3% to 61% of gastric cancers, depending on the tumor stage and assay used, he explained at the meeting sponsored by the American Gastroenterological Association, the American Society for Clinical Oncology, the American Society for Radiation Oncology, and the Society of Surgical Oncology.
Table of Contents
Study Downgrading and Design
The FIGHT trial was originally designed as a phase 3, randomized trial with the primary endpoint of OS, but it was converted to a phase 2 trial with a PFS primary endpoint and OS secondary endpoint.
In the question-and-answer phase following his presentation, Wainberg explained that the trial was downgraded because “we wanted to get a little more comfortable with the biomarker,” and because the prevalence of FGFR2b, which was unknown during the planning phase, played a large role in the statistical assumptions.
Patients were eligible for FIGHT if they had received no prior therapy for unresectable, locally advanced or metastatic gastric or GEJ cancers. They also had to have FGFR2b overexpression on immunohistochemistry and/or FGFR2 gene amplification by circulating tumor DNA. They had to have good Eastern Cooperative Oncology Group performance status (0 or 1) and could not have tumors positive for HER2.
Patients were allowed to have one cycle of FOLFOX while their FGFR2b status was being determined, prior to randomization.
After stratification for geographic region, FOLFOX use during screenings, and prior adjuvant or neoadjuvant chemotherapy, the patients were randomized to receive FOLFOX with either placebo or 15 mg/kg of bemarituzumab every 2 weeks, plus an additional 7.5-mg/kg dose on day 8 of cycle 1.
A total of 77 patients were randomized to the bemarituzumab arm, and 78 were randomized to the placebo arm.
Patient Status at Cutoff
As of the September 2020 cutoff, 14 patients continued on bemarituzumab, and 63 were off the assigned drug: 26 because of radiographic progression, 3 for consent withdrawal, 5 because of death, 20 because of adverse events, and 9 for other unspecified causes.
There were 42 patients in the bemarituzumab arm who were still being followed at the cutoff, and 35 were off study. Of this latter group, 28 died, 6 withdrew consent, and 1 was off study for an unspecified cause.
In the placebo arm, 8 patients were still receiving the assigned therapy, and 70 were off that treatment. Of the patients who discontinued study treatment, 4 died, 39 had radiographic progression, 7 withdrew consent, 3 had adverse events, and 17 discontinued for other unspecified reasons.
A total of 27 patients assigned to placebo were still on follow-up at the data cutoff. Of the remaining 51 patients, 40 died, 10 withdrew consent, and 1 had other, unspecified causes for going off study.
Primary Endpoint Met
The PFS rate at 9 months was 52.5% in the experimental arm and 33.8% in the control arm. The median PFS was 9.5 months with bemarituzumab and 7.4 months with placebo, which translated to a hazard ratio of 0.68 (P = .0727). This met the prespecified level of statistical significance for the phase 2 version of the trial, which required a two-sided alpha of 0.2.
OS at 12 months was also superior in the experimental arm, at 65.3% with bemarituzumab and 56.9% with placebo (HR, 0.58; P = .0268).
The benefit of bemarituzumab for both PFS and OS was greatest among patients whose tumors had the highest levels of FGFR2b expression.
Overall response rates were 47% in the experimental arm and 33% in the placebo arm. Among patients with measurable disease at baseline, the respective response rates were 53% and 40%. The median duration of response was 12.2 months in the bemarituzumab arm and 7.1 months in the placebo arm.
Corneal Toxicities, Stomatitis
The incidence of grade 3 or greater adverse events was 82.9% in the experimental arm and 74% in the placebo arm. Five patients in the bemarituzumab arm and four in the placebo arm died from treatment-related causes.
Grade 3 or greater stomatitis occurred in seven patients in the experimental arm, compared with just one patient in the placebo arm. Grade 3 or greater dry eye occurred in two patients in the experimental arm but none in the placebo arm.
Corneal adverse events occurred in 67.1% of patients treated with bemarituzumab and 10.4% of patients on placebo. Respective rates of grade 3 or greater corneal events were 23.7% and 0%.
Twenty patients assigned to bemarituzumab had to discontinue the drug because of corneal problems, compared with no patients on placebo.
At last follow-up, corneal adverse events had resolved in 60% of patients, while 40% had continuing problems. The median time to resolution of corneal adverse events was 27 weeks.
“FIGHT study is the first study to demonstrate that biomarker selection exists beyond HER2,” commented invited discussant Rutika Mehta, MD, of the Moffitt Cancer Center and University of South Florida, both in Tampa.
“These are promising results, with toxicities that need to be watched out for. It calls for a larger phase 3 study with special attention to the toxicity profile for bema[rituzumab],” she added.
The trial was supported by Five Prime Therapeutics. Wainberg disclosed relationships with Five Prime and other companies. Mehta disclosed relationships with several companies, not including Five Prime.
SOURCE: Wainberg ZA et al. GI Cancer Symposium 2021. Abstract LBA160.