Derazantinib, an inhibitor of fibroblast growth factor receptors 1-3, synergises with paclitaxel in pre-clinical gastric tumor models

Background: Derazantinib (DZB) is an oral fibroblast growth factor receptor (FGFR) inhibitor with clinical activity in intrahepatic cholangiocarcinoma. Kinase assays indicate activity against other important targets in oncology, including CSF1R and VEGFR2. We have shown that DZB can inhibit phosphorylation of CSF1R upon ligand stimulation in mouse macrophages ex vivo (GI50=100 nM); suggesting tumor-associated macrophages (TAMs) as an important target for DZB. Paclitaxel also reduces M2-TAM function, suggesting a potential synergy when combined with DZB. DZB monotherapy showed strong efficacy in some gastric (GA) PDX tumor models, so we have investigated the combination of DZB and paclitaxel in several GA-models in vivo. Materials and Methods: Four human tumor GA cell lines were studied in vitro: SNU-16, Fu97, AGS and KATOIII. SNU-16 (FGFR2-fusion) was grown s.c. in Balb/c mice as a xenograft (CDX), and 5 different patient-derived xenografts (PDX) with various FGFR-aberrations (fusion, amplification, over-expression or mutation) were grown in the same host. Mice were treated with different doses of DZB (p.o., qd) alone, and/or with paclitaxel (15 mg/kg, i.v., qw) for 3-4 weeks when tumors were 150 mm³. At the endpoint, tumors were ablated and snap-frozen or paraffin-embedded (FFPE) for western-blot/qPCR or immunohistochemistry (IHC), respectively. Efficacy was summarized as the endpoint dT/C, and the interaction assessed formally as synergy/additivity/antagonism by the Clarke-Combination-Index (CCI). The statistical significance of M2-TAMs was assessed using a one-tailed Fisher contingency test. The tumor models were run by CrownBio Inc, which also provided data on the FGFR-aberrations. Results: In vitro, DZB showed synergy with paclitaxel in SNU-16 and Fu97 models at concentrations known to be achievable in mouse plasma. In vivo, three experiments with the SNU-16 model showed reproducible synergy (mean CCI = -0.64) with the combination causing at least stasis and some complete-regressions. A PD-study after 3-days treatment showed a significant decrease in Ki67 and increase in the M1-TAMs in the combination group. Plasma PK showed no indication of a drug-drug interaction between the two compounds. In the 5 PDX-models, the combination showed synergy in three models and additivity in two. IHC analysis of M2-TAM levels in vehicle-treated mice of all 6 models showed that the two additive models had M2-TAM levels ≤0.8%, while synergy was seen in the 4 models with M2-TAMs of 1.1-8.7%. (p=0.03, using a cut-off of 1%). Conclusions: DZB combined with paclitaxel in vivo showed synergy/additivity in GA-tumor models with FGFR aberrations including FGFR2-fusions/amplifications, and FGFR1-3 over-expression or mutations. Additionally, the number of M2-TAMs may also play a role in sensitivity to the combination, which is consistent with the CSF1R kinase being an important target for DZB. Citation Format: Paul M. McSheehy, Mahmoud El Shemerly, Felix Bachmann, Laurenz Kellenberger, Heidi Lane. Derazantinib, an inhibitor of fibroblast growth factor receptors 1-3, synergises with paclitaxel in pre-clinical gastric tumor models [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2021 Oct 7-10. Philadelphia (PA): AACR; Mol Cancer Ther 2021;20(12 Suppl):Abstract nr P238.