Activity of cabozantinib (CABO) plus durvalumab (DURVA) in patients (pts) with advanced urothelial carcinoma (UC) or non-UC variant histologies (VH) after platinum chemotherapy: Interim results from the phase 2 ARCADIA trial.

Background: Combining multitargeted receptor tyrosine kinase inhibitor (TKI) with checkpoint inhibitors has shown synergistic effect in pts with UC due to the immunomodulatory propriety of VEGFR inhibitors. We investigated if the combination of CABO and DURVA in pts with advanced UC and non-UC histology (VHs) in a phase II study (NCT03824691). Herein the preliminary results of the interim analysis. Methods: Patients affected by UC and VHs recurred or progressed after failure of at least one line of platinum-based chemotherapy for metastatic disease have been treated with CABO 40 mg daily, orally, and are administered DURVA 1500 mg IV, q28 days, until disease progression (PD, by RECIST 1.1) or onset of unacceptable toxicity. Key inclusion criteria were ECOG-PS 0-1 and adequate organ function. Response was evaluated by RECIST criteria v.1.1 every 8 wks. The primary endpoint of the study was OS. Secondary endpoints included safety (CTCAE v.4.03), objective response-rate (ORR), duration of response (DoR), progression-free survival (PFS). The assumption was to detect an improvement in the median OS from ≤6 months (H0) to ≥9 months (H1). Results: Between September 2019 and February 10, 2023, the ARCADIA study enrolled 63 pts: this interim analysis was performed after obtaining at least one post-baseline tumor assessment data from 58 pts. The median follow-up was 23.5 mos (interquartile [IQ] range: 7.7-27.7mos): 27% were female, the median age was 64 yrs (range: 42-81), 20 pts (34%) had a pure/predominant non-UC VH: 9 (45%) a squamous differentiation/sarcomatoid tumor, 5 (25%) an adenocarcinoma, 4 (20%) a small-cell neuroendocrine tumor, 1 (5%) a clear-cell tumor, and 1 a nested VH (5%). Eleven pts (19%) had received ≥2 prior systemic anticancer therapies. In 58 response-evaluable pts, 12 (20%) complete responses (CR) and 11 partial responses (PR) were obtained, the ORR being 39.7% (95%CI: 27.1-53.4) and disease control rate was 69% (95%CI: 55.5-80.5). In the cohort of pts with VH, the ORR was 45% (95%CI: 23.1-68.5). Median PFS was 7.6 mos (95%CI: 4.6-13-6 mos) and median OS was 11.6 (95% CI 6.8-20.3 months). Median exposure of treatment was 4.8 mos; 35 pts (55.5%) of 63 pts had all-grade treatment-related adverse events (AE) and 4 pts (6.3%) reported grade 3 (treatment related) AE with no grade 3-5 events. Dose-reductions of CABO were needed in 25 pts (39%). No treatment-related deaths were reported. Conclusions: CABO in combination with DURVA showed promising preliminary activity with a manageable safety profile in pts with advanced UC and in non-UC VH after previous chemotherapy exposure, deserving further evaluation of the combination that is ongoing with ARCADIA and other clinical trials. More mature results with longer follow-up will be presented. Clinical trial information: NCT03824691.