Background: Programmed cell death –1 (PD-1) and its ligand (PD-L1) serve as checkpoints that regulate the interaction between the immune system and tumors. Antibodies blocking PD-1 or PD-L1 enhance antitumor activity in pts with UC. NIVO, a PD-1 blocking antibody, has clinical activity in UC as a single agent. IPI, an antibody blocking the checkpoint molecule CTLA-4, has demonstrated biologic activity in UC. Dual checkpoint blockade with both NIVO and IPI also has clinical activity in UC, but has not been systematically studied in NIVO-refractory UC. We sought to evaluate whether the addition of IPI to NIVO is feasible and clinically beneficial in pts whose UC is refractory to single agent NIVO. Methods: Forty pts with advanced or metastatic UC were treated with NIVO (3 mg/kg every 2 weeks) as part of a single arm, IRB-approved, investigator initiated study (NCT02553642). The study offered the option of treatment with the combination of IPI (1 mg/kg) and NIVO (3 mg/kg) for pts who had confirmed progression after treatment with NIVO monotherapy. The combination was dosed once every 3 weeks for 4 doses and followed by NIVO monotherapy. Pts were monitored for response and toxicity. Results: Ten pts who were refractory to NIVO monotherapy were treated with the combination of IPI and NIVO. There was a modest increase in the frequency of toxicities associated with the combination, compared to NIVO alone. One pt developed grade 3 diarrhea after treatment with the combination. No grade 3/4 toxicities were observed in pts treated with NIVO monotherapy. One pt achieved a confirmed partial response. Three additional pts achieved stable disease after documented progression. Pharmacodynamic changes in peripheral blood immune cells were monitored during treatment with NIVO monotherapy and combination therapy. Conclusions: The combination of IPI and NIVO is well tolerated and may have clinical activity in pts with UC who fail to respond to PD-1 blockade. Further studies to evaluate the magnitude of clinical activity and predictors of response for this combination are needed. Clinical trial information: NCT02553642