Real-world outcomes and prognostic factors for immune checkpoint inhibitors in dMMR/MSI-H metastatic colorectal cancer (mCRC).

101 Background: Mismatch repair deficiency (dMMR) or microsatellite instability-high (MSI-H) status predicts response to immune checkpoint inhibitors (ICIs) in mCRC. In the landmark KEYNOTE177 study, a mPFS of 16.5 months was observed with pembrolizumab, with a 29% progressive disease (PD) rate and 14% treatment discontinuation rate. In this study, we evaluated real-world efficacy and safety of single-agent pembrolizumab in dMMR/MSI-H mCRC and explored patient and tumor factors associated with outcomes. Methods: We conducted a retrospective study of patients (pts) with dMMR/MSI-H mCRC treated with pembrolizumab between 02/2022-11/2024. BC Cancer is the sole funder of oncology drugs across the province, treated pts were identified through the pharmacy database. Demographics, tumor characteristics, and treatment patterns were collected. Descriptive statistics were used for baseline characteristics. Progression-free (PFS) and overall survival (OS) were estimated using Kaplan–Meier methods and compared by log-rank tests. Cox proportional hazards models were used to identify prognostic features. Results: Among the 129 pts included, median age at diagnosis was 71 years (24-88) and 57% were female. BRAF and RAS mutations were identified in 53% and 29% of patients, respectively. The primary tumor was more frequently right-sided (53%). Common metastatic sites included non-regional lymph nodes (37%), liver (33%), and peritoneum (30%). Median PFS was 19.3 months (95% CI, 11.9–NE). Estimated PFS at 3 and 24 months was 80% (95% CI, 73–87) and 49% (95% CI, 40–60), respectively. 14% of patients experienced progression at first restaging. Median OS was not reached (95% CI, 22.4 months–NE), with estimated OS at 3 and 24 months of 87% (95% CI, 82–93) and 57% (95% CI, 47–68). Treatment-related toxicity occurred in 65% of pts; 16% required hospitalization and 16% discontinued treatment. The most frequent toxicities were dermatitis (13%), fatigue (12%), colitis (10%), and hypothyroidism (5%). On multivariable analysis, older age (HR 1.05, 95%CI: 1.02-1.08, p<0.001, per year) and the presence of liver metastasis (HR 2.03, 95%CI: 1.02 – 4.03, p = 0.043) were associated with inferior PFS, whereas the occurrence of any grade of treatment-related toxicity was associated with improved outcomes (HR 0.19, 95%CI – 0.09 – 0.40), p<0.001). Conclusions: In this population-based cohort, pembrolizumab demonstrated durable efficacy and manageable safety, comparable or better than that seen in KEYNOTE 177 with a mPFS of 19.3 mo and 16% discontinuation rate. A primary PD rate of 14% was lower than expected. Age, ECOG performance status, and liver metastases were significant prognostic factors, while development of treatment-related toxicity was associated with improved survival. Now with the availability of dual IO, these factors may help inform IO treatment strategy in the 1L for dMMR/MSI-H mCRC.