Targeting FGFR Alterations in Gastroesophageal Cancer

PROJECT SUMMARY Gastroesophageal adenocarcinoma (GEA) is the 4th leading cause of cancer deaths worldwide with a median survival of 6.2 months for patients who present with metastatic disease. Up to 10% of GEA patients show activation of fibroblast growth factor 2 (FGFR2) signaling due to FGFR2 gene amplification and/or fusion. FGFR2

overexpression is associated with lower response to chemotherapy and a poor prognosis. Clinical trials targeting FGFR2 in GEA include the use of both antibodies and small molecules. A Phase 2 clinical trial combining standard-of-care chemotherapy with an anti-FGFR2 monoclonal antibody (bemarituzumab) showed meaningful

but not yet statistically significant improvements in progression-free survival. Recent clinical data demonstrate significant improvements in progression-free survival in GEA patients treated with an FGFR2 selective tyrosine kinase inhibitor (TKI), RLY4008. However, resistance to TKIs emerge rapidly and we have identified on-target

resistance mutations in the FGFR2 kinase domain in GEA patients who progress on RLY4008. One approach to overcome TKI resistance mutations is through combination therapy with antibody drug conjugates (ADC). Standard antibodies, including bemarituzumab, do not significantly attenuate FGFR2 signaling likely due to the

bivalent nature of the antibodies that maintain receptor dimers in the absence of ligand. Biparatopic antibodies (BipAbs) however, recognize two distinct epitopes on the same protein leading to enhanced pharmacologic activity. Thus, we created 15 BipAbs from combinations of 6 FGFR2 antibodies targeting the extracellular domain

of FGFR2 which is retained in both FGFR2 amplified and FGFR2 fusion-positive GEA. Two of our FGFR2 BipAbs show markedly increased avidity, receptor degradation mediated by lysosomal pathways, and in vitro and in vivo activity against FGFR2-fusions. Furthermore, initial experiments show these two bipAbs are active in GEA cell

lines as ADCs. The significant clinical activity of the ADC trastuzumab-deruxtecan in HER2+ GEA suggests that this linker-payload combination targeting FGFR2 should be tested in FGFR2+ GEA. In this proposal, we will 1) utilize novel immune competent mouse models of FGFR2 amplified GEA or FGFR2 fusion-positive GEA to:

develop and test highly efficacious FGFR2 bipAbs and antibody-drug conjugates alone or in combination with immune checkpoint inhibitors, and 2) develop combination strategies with FGFR2 bipAbs and TKIs by understanding the mechanisms of resistance to FGFR2 selective TKI. We hypothesize that both approaches will

lead to enhanced responses to FGFR2 TKIs and delay the emergence of TKI resistance.