New Target to Attack in Stomach Cancer

Patients with untreated advanced gastric or gastroesophageal junction (GEJ) cancer lived longer without disease progression with the addition of a fibroblast growth factor receptor (FGFR) inhibitor to chemotherapy, a randomized trial showed.

Median progression-free survival (PFS) increased from 7.4 months with chemotherapy alone to 9.5 months with bemarituzumab. The size of the PFS benefit increased with levels of FGFR2b expression. Median overall survival (OS) was 12.9 months with chemotherapy and not yet reached in patients who also received the FGFR2b inhibitor.

Adverse events (AEs) that occurred more often with bemarituzumab included stomatitis and ocular effects, particularly corneal AEs, reported Zev Wainberg, MD, of the University of California Los Angeles, at the Gastrointestinal Cancer Symposium virtual meeting.

Wainberg noted that "[this] is the first study to evaluate targeting the overexpression of FGFR2b in any cancer, and the first randomized dataset of any fibroblast growth factor receptor inhibitor in any malignancy. In our study, approximately 30% of patients had a first-line non-HER2-positive gastric or gastroesophageal junction adenocarcinoma overexpressing FGFR by a centrally performed IHC (immunohistochemistry) test."

"Bemarituzumab, added to modified FOLFOX6 chemotherapy, led to clinically meaningful and statistically significant improvements in PFS, OS, and overall response rate [ORR]. The results support a prospective randomized phase III study in gastric/gastroesophageal adenocarcinoma and the evaluation of bemarituzumab to treat other FGFR2b-positive tumor types," he stated.

FGFR2b overexpression varies from 3%-60% of gastric cancers, depending on tumor stage and assay. FGFR tyrosine kinase inhibitors have proven active in cancers associated with FGFR mutations, fusions, or translocations.

Bemarituzumab selectively and specifically targets FGFR2b. In a preliminary clinical evaluation, the monoclonal antibody demonstrated single-agent activity in heavily treated FGFR2b-positive gastric cancer, including an objective response rate of 18%. The results supported continued evaluation in the prospective, randomized phase II FIGHT trial.

Investigators in the multicenter trial enrolled patients with untreated unresectable locally advanced or metastatic HER2-negative gastric/GEJ adenocarcinoma and FGFR2b overexpression or gene amplification. Patients were randomized to modified FOLFOX6 paired with bemarituzumab or placebo. The primary endpoint was investigator-assessed PFS. Statistical significance, defined by two-sided alpha of 0.2, was tested sequentially for PFS, OS, and ORR.

Data analysis included 155 randomized patients. In 83.2% of patients, FGFR2b status was confirmed by IHC, 12.9% by IHC and circulating tumor (ct) DNA, and 3.9% by ctDNA.

Analysis of the primary endpoint yielded a 32% reduction in the hazard ratio for disease progression or death among patients allocated to bemarituzumab (95% CI 0.44-1.04, P=0.0727). The proportion of patients who were progression free at 9 months was 52.5% with bemarituzumab and 33.8% with placebo.

Subgroup analysis by FGFR2b overexpression showed larger PFS benefits for bemarituzumab: 10.2 versus 7.3 months for IHC 2+/3+ ≥5% (n=148, HR 0.54, 95% CI 0.33-0.87) and 14.1 versus 7.3 months for IHC 2+/3+ ≥10% (n=96, 95% CI 0.44, 95% CI 0.25-0.77).

Median OS was 12.9 months with placebo but not yet reached in the bemarituzumab arm, representing a 42% reduction in the HR (95% 0.35-0.95, P=0.0268). ORR was 47% with bemarituzumab and 33% with placebo. Median duration of response was 12.2 versus 7.1 months with the anti-FGFR2b antibody and placebo, respectively.

Grade ≥3 treatment-emergent AEs (TEAEs) occurred slightly more often with bemarituzumab (82.9% vs 74.0%), but serious AEs affected similar proportions of patients (31.6% vs 36.4%). More patients in the bemarituzumab arm discontinued any component of chemotherapy because of AEs (46.1% vs 36.4%), and discontinuation of randomized therapy occurred significantly more often with bemarituzumab (34.2% vs 5.2%). Duration of exposure to randomized therapy was similar (24 vs 26 weeks).

Among grade ≥3 TEAEs, stomatitis occurred more often in the bemarituzumab arm (9.2% vs 1.3%). Corneal evaluation was required at baseline and every 8 weeks thereafter until the end of treatment. Overall, corneal events occurred in 67.1% and 10.4% of the bemarituzumab and placebo arms, including grade 3 in 23.7% and 0%.

Wainberg reported that 26.3% of patients in the bemarituzumab arm discontinued treatment because of corneal events, while none in the placebo arm did. The condition resolved in 12 of the 20 affected patients.

The results are consistent with those of other trials that have evaluated chemotherapy-antibody combinations for advanced and metastatic gastric/GEJ cancer, said invited discussant Rutika Mehta, MD, of Moffitt Cancer Center in Tampa, Florida.

"These results mean there is additional biomarker selection beyond HER2," she said. "About 30% of patients can be FGFR2b positive; higher expressers get more benefit. However, there were more adverse events with bemarituzumab. About a fourth of patients had corneal adverse events, and 40% were unresolved."

"A phase III study is warranted to confirm the findings in a larger population with special attention being paid to minimizing adverse events," Mehta concluded.

Comment