Toxicity in combination immune checkpoint inhibitor and radiation therapy: A systematic review and meta-analysis.

• Few studies characterize combined immunotherapy and radiation. • Toxicities similar after adding radiation to immunotherapy. • Anti-CTLA-4 associated with more toxicity than anti-PD-1/PD-L1. • No significant difference based on timing of immunotherapy and radiation. Immune checkpoint inhibitor with radiation therapy (ICI + RT) is under investigation for improved patient outcome, so we performed a systematic review/meta-analysis of toxicities for ICI + RT compared to immune checkpoint inhibitor (ICI) therapy alone. A PRISMA-compliant systematic review of studies in MEDLINE (PubMed) and in the National Comprehensive Cancer Network guidelines was conducted, with primary outcome grade 3 + toxicity. Criteria for ICI alone were: phase III/IV trials that compared immunotherapy to placebo, chemotherapy, or alternative immunotherapy; and for ICI + RT: prospective/retrospective studies with an arm treated with ICI + RT. Meta-analysis was performed by random effects models using the DerSimonian and Laird method. The I 2 statistic and Cochran's Q test were used to assess heterogeneity, while funnel plots and Egger's test assessed publication bias. This meta-analysis included 51 studies (n = 15,398), with 35 ICI alone (n = 13,956) and 16 ICI + RT studies (n = 1,442). Our models showed comparable grade 3–4 toxicities in ICI + RT (16.3%; 95% CI, 11.1–22.3%) and ICI alone (22.3%; 95% CI, 18.1–26.9%). Stratification by timing of radiation and irradiated site showed no significant differences, but anti-CTLA-4 therapy and melanoma showed increased toxicity. The grade 5 toxicities were 1.1% and 1.9% for ICI alone and ICI + RT respectively. There was significant heterogeneity, but not publication bias. The random effects model showed comparable grade 3–4 toxicity in using ICI + RT compared to ICI alone in CNS melanoma metastases, NSCLC, and prostate cancer. ICI + RT is safe for future clinical trials in these cancers. [ABSTRACT FROM AUTHOR]