Derazantinib (DZB) in combination with atezolizumab (AZB) in patients with solid tumors: Results from the dose-finding phase Ib substudy of FIDES-02.

Abstract

437 Background: DZB is an oral small-molecule FGFR1/2/3 kinase inhibitor, which demonstrated antitumor activity in preclinical and clinical studies (in cholangiocarcinoma patients [pts] with FGFR2-driven tumors and in FGFR1-3-driven urothelial cancer [UC] PDX models). In CSF1-stimulated mouse bone-marrow derived macrophages, DZB reduced CSF1R phosphorylation with a maximal effect similar to the selective CSF1R inhibitor BLZ945, suggesting DZB could have an effect on tumor-associated macrophage regulation. Thus, DZB+AZB is a rationale combination to be investigated in immunogenic and FGFR-driven tumors like UC. Methods: FIDES-02 is a multi-cohort open-label Phase 1b/2 study evaluating the effect of DZB as monotherapy and DZB+AZB in combination. To determine the RP2D of DZB+AZB, a total of 26 pts with UC (N = 4) and other solid tumors (N = 22), of whom 7 pts carried various FGFR genetic aberrations (GA), were enrolled at 2 dose levels (DL) (1200 mg AZB Q3W + 200 mg [DL1] / 300 mg [DL2] DZB QD) and treated until disease progression or unacceptable toxicity. Both DLs were divided into MTD (endpoint: dose-limiting toxicity [DLT] at D21) and expansion cohorts to investigate both acute and delayed adverse events (AEs) per CTCAE v5. Results: In the MTD cohorts of both DL1 (N = 7) and DL2 (N = 6) no DLTs were observed, and the DL1 and DL2 expansion cohorts subsequently enrolled 7 and 6 pts, respectively. The most frequent treatment-emergent AEs across both DLs were fatigue (31%), nausea (27%), diarrhea (23%), the most frequent laboratory abnormalities were increased ALT (58%), and AST (50%). Non-DLT grade (G) ≥ 3 treatment-related AEs (TRAE) across both DLs were G3 diarrhea (4%), G3 nausea (4%), G3 asthenia (4%), oral fungal infection (4%) and one case of immune-related G4 nephritis (4%). Dose interruptions / reductions and discontinuations due to TRAEs occurred in 19% and 8%, respectively. Pharmacokinetic analyses demonstrated that DZB exposure parameters and AZB serum concentrations in pts treated with DZB+AZB were similar to those assessed in pts under DZB / AZB monotherapy. At data cut-off, 2 of 14 (DL1) and 7 of 12 pts (DL2) were still on treatment. Median duration of treatment in DL1 was 9.7 weeks (range, 9-25), and best overall response per investigator assessment was SD in 4 of 10 efficacy-evaluable DL1 pts. DL2 efficacy analysis was uninformative at cutoff. The combination of 1200 mg AZB Q3W with both 200 mg and 300 mg QD DZB was found to be safe and tolerable. Conclusions: The combination of DZB+AZB is safe at both investigated DLs, the RP2D has been determined as 300 mg DZB QD+1200 mg AZB Q3W. DZB can be safely dosed at its monotherapy RP2D in this novel combination with AZB. The anti-tumor efficacy of DZB+AZB is currently being investigated in adequately designed cohorts of FIDES-02 with enrichment for UC pts with FGFR GA. Clinical trial information: NCT04045613.