Background: In cancer treatment (tx), loss of T-cell function is associated with progressive expression of checkpoint (CP) molecules, and resistance or relapse following PD-(L)1 inhibitor therapy is a high unmet need. LAG-3 and TIM-3 are key CP pathways implicated in resistance to PD-1 inhibitors and have shown synergistic effects upon triple inhibition with PD-1 in a murine model. This open-label, nonrandomized phase 1/2 study (NCT04370704) aims to determine optimal doses and preliminary safety and efficacy for the monoclonal antibody combinations of INCAGN02385, INCAGN02390, and retifanlimab. Here we report initial safety and efficacy results. Methods: Study consists of 4 parts in phase 1 and 2 expansion cohorts in phase 2. In phase 1, parts 1 and 2 determined safety of INCAGN02385 350 mg Q2W + INCAGN02390 400 mg Q2W and INCAGN02385 350 mg Q2W + INCAGN02390 400 mg Q2W + retifanlimab 500 mg Q4W, respectively; parts 3 and 4 will test additional dosing schedules. Pts enrolled had previously treated (including PD-[L]1) locally advanced unresectable or metastatic solid tumors, with no prior LAG-3– or TIM-3–directed therapy. Results: As of June 28, 2021, 21 pts were enrolled in parts 1 (n=10) and 2 (n=11); 62% of pts were <65 years and 38% were male. Pts had median 3 (range 1-9) prior lines of therapy, including prior anti–PD-(L)1 (part 1, 8 pts; part 2, 11 pts); pt tumor types are listed. Nineteen pts discontinued tx, 1 due to toxicity. Nine pts in part 1 and 10 pts in part 2 had tx-emergent adverse events (TEAEs), most commonly anemia and diarrhea (each 30%) in part 1 and anemia (36%) in part 2 (all tx-unrelated). Serious tx-related TEAEs occurred in 1 pt in part 1 (grade [G] 3 vasculitis) and 1 pt in part 2 (G3 myocarditis and pericardial effusion). Most common G≥3 TEAE was G3 anemia (part 1, 2 pts; part 2, 3 pts); 1 pt in part 1 had a G5 TEAE of sepsis unrelated to tx, with COVID-19 infection. In part 2, 1 pt had an infusion-related reaction, and 1 pt had immune-related TEAEs of pericardial effusion and myocarditis (both G3 and tx-related), which resolved after steroid therapy and tx discontinuation. Best overall response in part 1 was stable disease, and in part 2 was a confirmed and durable partial response in 1 pt with CP therapy–refractory melanoma. Conclusions: Doublet anti–LAG-3 + anti–TIM-3 and triplet anti–LAG-3 + anti–TIM-3 + retifanlimab regimens in the study were generally well tolerated in heavily pretreated pts, with a safety profile consistent with that of CP monotherapy and no novel toxicities. Optimal doses of both combinations are to be further evaluated in the ongoing study in first-line tx of melanoma. Clinical trial information: NCT04370704.