Background: Pembrolizumab (pem) was initially approved in the first line setting for cisplatin-ineligible metastatic urothelial carcinoma (mUC) patients (pts) with high PD-L1 expression and platinum ineligible pts regardless of PD-L1 status; this later changed to only platinum ineligible pts. There is a high disease progression (PD) rate with pem monotherapy. Cabozantinib (cabo) is a multiple receptor tyrosine kinase inhibitor targeting MET and VEGFR2 approved in multiple other malignancies as monotherapy or in combination with PD-1 inhibition. We hypothesized that combination of pem and cabo therapy will be safe and efficacious in patients with treatment naïve mUC. Methods: This is a phase II, open label, multi-center, single arm trial evaluating the tolerability and activity of pem 200 mg every 3 weeks and cabo 40 mg daily as first-line treatment for patients with mUC. Key inclusion criteria were histologically proven locally advanced or mUC, ECOG-PS 0-2, cisplatin-ineligible (including patient refusal of cisplatin), treatment naïve, and no prior PD-1/L1 therapy. The primary endpoint was objective response rate (ORR). Key secondary endpoints included progression-free survival (PFS) and overall survival (OS). There was a lead-in safety cohort of 6 pts. We hypothesized that combination therapy would improve ORR to > 32%. With 35 evaluable subjects, the lower bound of the 95% confidence interval would extend no more than 26% from the observed proportion. With ≥17 objective responses, the confidence interval excludes 32%. Results: Between December 26, 2018, and April 19, 2023, 36 pts were enrolled of which 35 were evaluable for response with the median follow up of 14.4 months (95% CI 12.2 – 16.2). The median age was 72.5y [range, 47-82]; ECOG PS 0-1 in 89% pts. Responses were observed in 15 pts (ORR= 42.8%) including complete responses in 5 pts (14.3%). Stable disease was seen in 10 pts (28.6%). The disease control rate was 68.5% (25/35). Median PFS was 7.7 months (95% CI 4.2 – 11.2). Fifty-two percent of pts developed any grade treatment-related adverse events (TRAE) attributable to either cabozantinib or pembrolizumab. The most common grade 1/2 TRAE were diarrhea, anorexia, dysgeusia, weight loss, and nausea. Forty-four percent pts developed grade 3/4 TRAEs most common being hypertension, hypophosphatemia, alanine transaminase elevation, diarrhea, and fatigue. Conclusions: This novel phase II trial of pem + cabo demonstrated a manageable toxicity profile and promising efficacy as first-line therapy in mUC including patient’s ineligible for cisplatin. Further investigation with a focus on predictive biomarkers is ongoing. Clinical trial information: NCT03534804.