Joseph Chao, MD

The randomized phase 2 FIGHT trial investigated the addition of the anti-FGFR2b agent bemarituzumab to first-line chemotherapy and represents an important incremental step forward for stage IV gastroesophageal adenocarcinoma.

Novel precision medicine strategies are still limited, as evidenced by the few regulatory approvals for targeted agents, despite genomic profiling efforts such as The Cancer Genome Atlas (TCGA) denoting potential molecular targets. High-level amplification of the wild-type FGFR2 gene has been described in the gastric literature and historically has been associated with poor prognosis in retrospective analyses. Previous efforts with tyrosine kinase inhibitors against this target have been unsatisfactory and, as such, bemarituzumab demonstrating clinical benefit in this proof-of-concept randomized study represents a meaningful path forward in improving outcomes for patients harboring this biomarker.

With 30% of patients testing positive for FGFR2b overexpression by a central laboratory assay, this represents a novel assessment methodology to characterize this target as an oncogenic driver. This contrasts with HER2, where immunohistochemistry positivity typically associates with HER2 gene amplification and where trastuzumab [Herceptin, Genentech] benefit predominantly is observed.

FGFR2 gene amplification, as assessed by circulating tumor DNA in the FIGHT trial, also was reported, but it represented only a minority (17%) of the trial population and appears in line with previous biomarker analyses when one factors in the proportion for all patients screened for the trial.

Oncogenic copy number alterations such as HER2 and FGFR2 have been well-characterized in the chromosomal instability (or CIN) molecular subtype of gastroesophageal adenocarcinoma from genomic sequencing efforts such as the TCGA. As such, FGFR2b protein overexpression without corresponding gene amplification raises one new potential avenue of research into post-transcriptional or post-translational mechanisms driving gastroesophageal adenocarcinoma oncogenesis. Further biomarker research efforts will also need to explore intratumoral genomic heterogeneity of this target, as heterogeneity has been well-recognized as a barrier to precision medicine efforts in this disease.

With investigator-assessed PFS as a primary endpoint and OS as a secondary endpoint, a follow-up phase 3 study will likely be necessary to validate this approach as a change in standard of care. However, the standard of care in first-line therapy is shifting with recent reporting of the phase 3 CheckMate 649 trial, where OS improved with the addition of nivolumab [Opdivo, Bristol Myers Squibb] to chemotherapy in PD-L1-overexpressing gastroesophageal adenocarcinoma tumors with a combined positive score cutoff of 5 or higher.

Data are still limited on whether overlap exists between high PD-L1 and FGFR2 overexpression, presenting questions on choice of control arm in a follow-up study if a patient’s tumor harbors both biomarkers. Combining both bemarituzumab and immune checkpoint inhibitors with first-line chemotherapy may be a strategy to overcome this conundrum and may augment antitumor immune response, even in low PD-L1 but FGFR2b-overexpressing tumors. Proof of principle with this approach has been supported by phase 2 trials exhibiting encouraging clinical activity of anti-HER2 and anti-PD-1 combinatorial approaches. However, safety will still need to be carefully examined, as in the phase 2 FIGHT trial, higher toxicities — particularly stomatitis and corneal adverse events — were noted with the addition of bemarituzumab to chemotherapy.

Follow-up phase 3 studies with larger patient numbers will hopefully provide reassurance that corneal toxicities are predominantly reversible with discontinuation of bemarituzumab, as reported in the FIGHT study.

Reference:

Moehler M, et al. Abstract LBA6_PR. Presented at European Society of Medical Oncology Virtual Congress; Sept. 19-21, 2020.