Background: Existing literature has reported differences in clinical outcomes by ethnicity in patients receiving immune checkpoint inhibitors (Olsen et al Front Oncol 2021). We investigated real-world outcomes between Latinx and non-Latinx mRCC patients treated with first-line nivo/ipi within a safety-net healthcare system and at a tertiary care center in Southern California. Methods: We performed a retrospective analysis of mRCC patients who received nivo/ipi within the Los Angeles County Department of Health Services (DHS), a safety-net healthcare system, and the City of Hope Comprehensive Cancer Center (COH), a tertiary oncology center, between Jan. 1, 2015 and Dec. 31, 2021. Patients were identified using institutional databases and clinical data were compiled from electronic health records. Patients with pathologic diagnosis of mRCC, age > 18 years and receipt of nivo/ipi as first-line therapy were included. Progression-free survival (PFS) and overall survival (OS) were analyzed using the Kaplan-Meier method, log-rank test, and Cox proportional hazards model with adjustments for other covariates. Results: Of 94 patients, 66 (70%) were male, 90 (94%) had clear-cell histology, and 87 (93%) had IMDC intermediate/poor risk disease. Forty patients (43%) were Latinx. Fifty (53%) and 44 (47%) patients received their care at a tertiary care center and within a safety-net healthcare system, respectively. Most Latinx patients (95%) were treated at DHS, and most non-Latinx patients (89%) were treated at COH. Latinx patients were significantly older than non-Latinx patients (59.5 vs 55 years, p=0.008). IMDC risk classification, body mass index, history of nephrectomy, and number of comorbidities were similar between both groups. Pooled analysis by ethnicity demonstrated significantly shorter PFS in Latinx versus non-Latinx patients (10.1 vs 25.2 months, HR 3.61, 95% CI 1.96-6.66, p= <0.01). Adjusting for age, gender, IMDC risk classification, history of nephrectomy, and number of co-morbidities, multivariate analysis revealed a HR of 3.41 (95% CI 1.31-8.84; p=0.01). At a median follow up of 11.0 months, the median OS was not met in either arm at the time of data cutoff (NR vs. NR, HR 1.34, 95% CI 0.44-4.11). Conclusions: Compared to non-Latinx patients,Latinx patients demonstrated shorter PFS; no difference was observed in OS although these data were immature. As the majority of Latinx patients received their care at DHS, our data suggest that disparities in access to care may significantly contribute to differences in clinical outcomes of mRCC patients receiving nivo/ipi.