Background: Checkpoint inhibitors (CPI) and vascular endothelial growth factor receptor inhibitors (VEGFi) are standard treatments for patients (pts) with mRCC. This phase I/II study evaluated the safety and efficacy of the novel combination of pembrolizumab (pembro) and cabozantinib (cabo). The phase I dose escalation data was presented at ASCO GU 2019. We now report the objective response rate (ORR), progression free survival (PFS), overall survival (OS), and toxicity of patients in the phase II dose expansion. Methods: Eligible pts had metastatic clear cell (ccRCC) or non-clear cell (nccRCC) histology, normal organ function, ECOG 0-1, and no prior exposure to pembro or cabo. Pts could be treatment-naïve or have received prior CPI and/or VEGFi. Pts were dosed at the recommended phase 2 dose of pembro 200 mg IV Q3W in combination with cabo 60 mg PO QD. Scans were obtained every 9 weeks. Treatment beyond progression, in the setting of continued clinical benefit, was allowed. The primary endpoint was ORR. Simon’s two-stage design was implemented to test the null hypothesis that ORR ≤ 0.20 versus the alternative that ORR ≥ 0.50. Results: Forty pts were enrolled, of which 34 pts (85%) had ccRCC and 6 pts (15%) had nccRCC. This was first-line treatment for 15 pts (38%) and second- and subsequent-line therapy for 25 pts (62%). IDMC risk category was favorable in 15%, intermediate in 72.5%, and poor in 12.5% of pts. Prior therapies included VEGFi in 17 pts (43%), CPI in 17 pts (43%), and 9 pts (23%) had both prior VEGFi and CPI in combination or sequentially. At a median follow up of 17.8 months (mo), the ORR was 60% (95% CI 0.458-1.00), clinical benefit rate (CBR) was 92.5% (95% CI 0.817-1.00), median time to response was 4.2 mo; median duration of response was 8.4 mo. Three of six nccRCC pts achieved partial response. Median PFS was 10.4 mo (95% CI 6.3 mo-NR). Median OS was not reached. Twelve patients remain on treatment. The most common grade 1 and 2 (G1/2) treatment-related AEs were diarrhea (53%), fatigue (49%), weight loss (47%), nausea (43%), and dysgeusia (43%). Twenty-five patients (47%) experienced a treatment-related G3 AE and there were no G4 related AEs. Thirteen pts experienced serious adverse events, 8 of which were related to treatment: G3 transaminitis and hypoglycemia were attributed to the combination; G3 pancreatitis, nephritis, and pneumonitis attributed to pembro; G3 pulmonary embolus, confusion due to reversible posterior leukoencephalopathy (RPLS), and stroke attributed to cabo. There was one treatment-related death in the pt with RPLS, possibly related to cabo. Conclusions: This study of the combination of pembrolizumab 200mg and cabozantinib 60mg met the primary endpoint of ORR. Benefit was seen in first- and subsequent-line therapy. The safety profile was manageable. This combination warrants further confirmation in a randomized controlled trial. Clinical trial information: NCT03149822