Immune checkpoint inhibitors (ICIs) have revolutionized cancer therapy by reinvigorating antitumor immune responses. PD-1/PD-L1 and CTLA-4 inhibitors are among the most promising, with some approved for cancer treatment and others in clinical trials. However, only a fraction of patients respond to ICIs, and immune-related adverse events (irAEs) can occur. This review discusses potential biomarkers for predicting response to ICIs and irAEs, including immune cell infiltration, PD-L1 overexpression, neoantigens, and genetic/epigenetic signatures. Immune inflamed tumors respond better to ICIs due to increased tumor-infiltrating lymphocytes (TILs) and other immune cells. Biomarkers such as increased peripheral blood absolute lymphocyte count (ALC), low serum lactate dehydrogenase (LDH), and specific T-cell subsets can predict response. PD-L1 overexpression is a potential biomarker, but its correlation with response is not universal. Neoantigens, generated by mutations in tumor DNA, are promising biomarkers as they can be targeted by ICIs. Genetic signatures, such as mutations in PD-L1 and PD-L2, and mismatch-repair deficiency, also show potential. Epigenetic modifications, including DNA methylation and miRNA expression, are being explored as biomarkers. Despite progress, challenges remain in developing reliable biomarkers due to the dynamic nature of cancer and individual genetic differences. Combination biomarker sets are needed to improve prediction of ICI outcomes and reduce irAEs. Further research is required to validate these biomarkers and integrate them into clinical practice for personalized cancer immunotherapy.Immune checkpoint inhibitors (ICIs) have revolutionized cancer therapy by reinvigorating antitumor immune responses. PD-1/PD-L1 and CTLA-4 inhibitors are among the most promising, with some approved for cancer treatment and others in clinical trials. However, only a fraction of patients respond to ICIs, and immune-related adverse events (irAEs) can occur. This review discusses potential biomarkers for predicting response to ICIs and irAEs, including immune cell infiltration, PD-L1 overexpression, neoantigens, and genetic/epigenetic signatures. Immune inflamed tumors respond better to ICIs due to increased tumor-infiltrating lymphocytes (TILs) and other immune cells. Biomarkers such as increased peripheral blood absolute lymphocyte count (ALC), low serum lactate dehydrogenase (LDH), and specific T-cell subsets can predict response. PD-L1 overexpression is a potential biomarker, but its correlation with response is not universal. Neoantigens, generated by mutations in tumor DNA, are promising biomarkers as they can be targeted by ICIs. Genetic signatures, such as mutations in PD-L1 and PD-L2, and mismatch-repair deficiency, also show potential. Epigenetic modifications, including DNA methylation and miRNA expression, are being explored as biomarkers. Despite progress, challenges remain in developing reliable biomarkers due to the dynamic nature of cancer and individual genetic differences. Combination biomarker sets are needed to improve prediction of ICI outcomes and reduce irAEs. Further research is required to validate these biomarkers and integrate them into clinical practice for personalized cancer immunotherapy.