Background: Immune checkpoint inhibitors (ICIs) are part of standard of care in many advanced solid malignancies. Patients (pts) with poor performance status or organ dysfunction have been usually ineligible in pivotal clinical trials of ICIs. Methods: In this retrospective cohort study, we assessed ICI uptake and effectiveness among 34,131 pts diagnosed with advanced non-targetable non-small cell lung (NSCLC), urothelial cell (UCC), renal cell (RCC), or hepatocellular (HCC) carcinoma between January 1, 2014 and December 31, 2019. Pts were identified from the Flatiron Health electronic health record-derived database, consisting of pts receiving care in 280 mostly community oncology practices in the US. 9,318 (27.3%) pts were trial-ineligible based on ECOG performance status ≥ 2 or liver (total bilirubin ≥ 3 mg/dL) or kidney (eGFR < 30 ml/min/1.73m²) dysfunction. The association between trial-ineligibility and ICI monotherapy uptake was assessed using inverse probability weighted logistic regression. We hypothesized that there would be no differences in overall survival between ICI and non-ICI therapies in the trial-ineligible cohort. To test this hypothesis, we used inverse probability of treatment weighted survival analyses. Because we observed non-proportional hazards, we report 12- and 36-month restricted mean survival times (RMSTs) and time-varying hazard ratios < and ≥ 6 months. Results: Between 2014-2019, the proportion of pts who received ICI monotherapy rose from 0% to 30.2% among trial-ineligible pts and from 0.1% to 19.4% among trial-eligible pts. Trial-ineligible pts received ICI monotherapy more frequently than trial-eligible pts (22.5% vs 12.7%, adjusted odds ratio [aOR] relative to non-ICI therapy 1.8, 95% CI 1.7-1.9). Among trial-ineligible pts, there were no OS differences between ICI monotherapy, ICI combination therapy, and non-ICI therapy at 12 months (RMST 7.8 vs. 7.7 vs. 8.1 mos) or 36 months (15.0 vs. 13.9 vs. 14.4 mos). Relative to non-ICI therapy, ICI monotherapy was associated with early mortality (adjusted hazard ratio [aHR] within 6 months 1.19) but late benefit (aHR after 6 months 0.80) in trial-ineligible pts (Table). Conclusions: Trial-ineligible pts with advanced malignancies preferentially receive first-line ICI, despite the lack of a survival benefit and possible early mortality associated with ICI use. Positive results for ICI in phase III trials may not translate to this vulnerable population.