Background. PD-1/PD-L1 inhibitors have made unprecedented progress in the treatment of cancer. Methods. A systemic search was conducted for randomized controlled trials that compared PD-1/PD-L1 inhibitor monotherapy or combination therapy with nonimmunotherapy. Hazard ratios (HRs) of overall survival (OS) according to the sex, age, ECOG PS, smoking status, liver metastasis, PD-L1 expression, EGFR, and KRAS status of patients were analyzed. Results. Totally, 13 studies with monotherapy and 5 with combination regimens were included, and the pooled HRs of OS were 0.74 (P<0.001) and 0.64 (P<0.001), respectively. EGFR wild-type patients could benefit from immunotherapy monotherapy (HR, 0.77; P<0.001) while those of the mutant type had no survival benefit (HR, 1.11; P=0.54), and the difference was statistically significant (interaction, P=0.005). KRAS wild-type patients had no survival benefit from monotherapy (HR, 0.89; P=0.49). For combination therapy, both male and female derived benefits but female had a significantly reduced risk of death (HR, 0.45; P<0.001) compared with male (HR, 0.73; P<0.001; interaction, P=0.004). Nonsmokers derived more survival benefits from combination therapy (HR, 0.29; P<0.001) than smokers (HR, 0.63; P=0.001; interaction, P=0.02). No significant difference was found between age, ECOG PS, liver metastasis, PD-L1 expression, and OS of both PD-1/PD-L1 inhibitor monotherapy and combination therapy. Conclusions. Both PD-1/PD-L1 inhibitor monotherapy and combination therapy significantly prolonged the OS of patients with advanced malignant tumors. EGFR status for monotherapy and sex and smoking status for combination therapy were important predictors of survival benefits.