Studies on associations between ataxia telangiectasia-mutated (ATM) polymorphisms and late radiotherapy-induced adverse events vary in clinical settings, and the results are inconsistent.We conducted the first meta-analysis following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to investigate the role of the ATM polymorphism rs1801516 in the development of radiotherapy-induced late fibrosis.We searched PubMed, Embase, Web of Science, and Chinese National Knowledge Infrastructure databases to identify studies that investigated the effect of the ATM polymorphism rs1801516 on radiotherapy-induced late fibrosis before September 8, 2015. Summary odds ratios (ORs) and the corresponding 95% confidence intervals (CIs) were used to assess the association between late fibrosis and the rs1801516 polymorphism. Subgroup analyses were conducted to evaluate the influence of clinical features on the genetic association. Tests of interaction were used to compare differences in the effect estimates between subgroups.The overall meta-analysis of 2000 patients from 9 studies showed that the minor allele of the rs1801516 polymorphism was associated with a significantly increased risk of developing late fibrosis (OR = 1.78, 95% CI: 1.07, 2.94), with high between-study heterogeneity (I2 = 66.6%, P = 0.002). In subgroup analyses, we identified that the incidence of late fibrosis was a major source of heterogeneity across studies. The OR for patients with a high incidence of late fibrosis was 3.19 (95% CI: 1.86, 5.47), in contrast to 1.09 (95% CI: 1.01, 1.17) for those with a low incidence. There was a significant difference in the effect estimates between the 2 subgroups (ratio of OR = 2.94, 95% CI 1.70, 5.08, P = 0.031).This meta-analysis supported previously reported effect of the ATM polymorphism rs1801516 on radiotherapy-induced late fibrosis. This finding encouraged further researches to identify more genetic polymorphisms that were predictive for radiotherapy-induced adverse events. In addition, we showed that the inconsistency of the associations seen in these studies might be related to variations in the incidence of late fibrosis in the patients. This suggested that future studies should consider the incidence of radiotherapy-induced adverse events when investigating radiosensitivity signature genes.