Limited prospective data about the activity of immune checkpoint inhibitors (ICIs) are available for elderly patients. The aim of our analysis was to determine the relative efficacy of ICIs versus available standard therapies [standard of care (SOC)] in subgroups defined by patients' age. Searching the MEDLINE/PubMed, Cochrane Library, and American Society of Clinical Oncology (ASCO) Meeting abstracts randomized clinical trials were identified. Data extraction was conduced according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement. The measured outcome was overall survival (OS). Twenty-nine randomized clinical trials (18,839 patients) were selected. As for the distribution of patients by age, all but 3 of the selected studies considered young the patients younger than 65 years (n=10,832) and elderly those with 65 years and older (n=7723); 7 studies identified a third subgroup of very elderly patients aged 75 years and above (n=421). In elderly and very elderly patients ICIs significantly reduced the risk of death by 23% compared with SOC [hazard ratio (HR), 0.77; P<0.00001)]. On the contrary, a lack of a survival advantage of immunotherapy was observed in the subgroup of very elderly patients (HR, 0.85; P=0.39). When comparing the efficacy of ICIs between the 2 subpopulations (elderly vs. young), no significant difference in OS was observed (HR, 0.76; P=0.66). ICIs prolonged OS compared with SOC in both elderly and young patients affected by lung cancer, melanoma, and renal carcinoma, regardless of the age. In conclusion, ICIs (as monotherapy or combinations) significantly improved OS compared with SOC in both young and elderly patients with advanced cancers, regardless of the tumor type. The magnitude of this benefit is debated in patients aged 75 years and above.