Background: Dostarlimab is a programmed death receptor-1 (PD-1) blocking antibody that is approved in the US as a monotherapy in adult patients (pts) with mismatch repair deficient (dMMR) recurrent or advanced endometrial cancer that has progressed on or following prior treatment with a platinum-containing regimen or dMMR recurrent or advanced solid tumors that have progressed on or following prior treatment and who have no satisfactory alternative treatment options. Here, we report on the antitumor activity and safety of dostarlimab monotherapy in pts with dMMR colorectal cancer (CRC), a post-hoc subgroup analysis of cohort F of the GARNET trial. Methods: GARNET is a phase 1, multicenter, open-label, single-arm study of dostarlimab monotherapy in pts with advanced or recurrent solid tumors. Cohort F of the GARNET expansion cohorts enrolled pts with dMMR/MSI-H or POLε mutated non-endometrial solid tumors, including pts with CRC. Pts must have progressed per blinded independent central review (BICR) following prior systemic therapy for advanced disease. Pts with CRC were required to have progressive disease after or been intolerant to fluoropyrimidine, oxaliplatin, and irinotecan. Pts received 500 mg intravenous dostarlimab every 3 weeks for 4 cycles, followed by 1000 mg every 6 weeks until discontinuation. Primary endpoints were objective response rate (ORR) and duration of response by BICR per RECIST v1.1. Pts were included in the efficacy analysis if they received ≥1 dose of dostarlimab, had measurable disease at baseline, and had ≥24 weeks of follow-up. All pts who received ≥1 dose of dostarlimab were included in the safety analysis. Results: As of the March 01, 2020 interim analysis data cut, 141 pts with dMMR non-endometrial solid tumors were included in the safety analysis, with 106 in the efficacy analysis. Of the pts in the efficacy analysis, 69 (65.1%) had CRC. Confirmed ORR by BICR per RECIST v1.1 in pts with dMMR CRC was 36.2% (95% CI, 25.0%–48.7%). There were 3 complete responses and 22 partial responses. At the data cut, 23 pts (92%) were still on treatment. Median duration of response had not been reached for pts with CRC. In the overall dMMR non-endometrial solid tumor population, treatment-related adverse events (TRAEs) were reported in 68.1% of pts, with 8.5% of pts experiencing at least 1 grade ≥3 TRAE. The most common grade ≥3 TRAE was lipase increased in 2 (1.4%) of pts. Only 5 pts (3.5%) discontinued dostarlimab due to at TRAE. Treatment-related serious AEs were reported in 9 (6.4%) pts. Conclusions: Dostarlimab demonstrated durable clinically meaningful antitumor activity in pts with dMMR CRC, which was consistent with that seen in patients with dMMR non-CRC solid tumors. No new safety signals were detected in patients with dMMR non-endometrial solid tumors. Clinical trial information: NCT02715284.