OBJECTIVES: Response to immune checkpoint inhibitor (ICI) remains limited to a subset of patients and predictive biomarkers of response remains an unmet need, limiting our ability to provide precision medicine. Using real-world data, we aimed to identify potential clinical prognosticators of ICI response in solid tumor patients. METHODS: We conducted a retrospective analysis of all solid tumor patients treated with ICIs at the Mount Sinai Hospital between January 2011 and April 2017. Predictors assessed included demographics, performance status, co-morbidities, family history of cancer, smoking status, cancer type, metastatic pattern, and type of ICI. Outcomes evaluated include progression free survival (PFS), overall survival (OS), overall response rate (ORR) and disease control rate (DCR). Univariable and multivariable Cox proportional hazard models were constructed to test the association of predictors with outcomes. RESULTS: We identified 297 ICI-treated patients with diagnosis of non-small cell lung cancer (N = 81, 27.3%), melanoma (N = 73, 24.6%), hepatocellular carcinoma (N = 51, 17.2%), urothelial carcinoma (N = 51, 17.2%), head and neck squamous cell carcinoma (N = 23, 7.7%), and renal cell carcinoma (N = 18, 6.1%). In multivariable analysis, good performance status of ECOG ≤ 2 (PFS, ORR, DCR and OS) and family history of cancer (ORR and DCR) associated with improved ICI response. Bone metastasis was associated with worse outcomes (PFS, ORR, and DCR). CONCLUSIONS: Mechanisms underlying the clinical predictors of response observed in this real-world analysis, such as genetic variants and bone metastasis-tumor microenvironment, warrant further exploration in larger studies incorporating translational endpoints. Consistently positive clinical correlates may help inform patient stratification when considering ICI therapy.