Background
Evidence indicates RFC1 G80A polymorphism as a risk factor for a number of cancers. Increasing studies have been conducted on the association of RFC1 G80A polymorphism with acute lymphoblastic leukemia (ALL) risk. However, the results were  controversial. The aim of the present study was to derive a more precise estimation of the relationship. 

Materials and Method
PubMed, Embase, Web of Science, Cochrane database, and Google Scholar were searched to get the genetic association studies between RFC1 G80A polymorphism and ALL. All eligible studies for the period up to February 2016 were identified. Subgroup analyses regarding ethnicity were also implemented. All statistical analyses were done with CMA 2.0.
Results
A total of ten studies comprising of 2,168 ALL cases and 2,693 healthy controls were included in this meta-analysis. Overall, no significant association was detected for allelic model (OR = 1.029, 95 % CI 0.754- 1.405, P=0.000), Dominant model (OR = 1.619, 95 % CI 0.847-3.094, P=0.145), recessive model (OR = 1.169, 95 % CI 10.764-1.790, P=0.429), and homozygote model (OR = 1.288, 95 % CI 0.928-1.788, P=0.130). However, there was an obvious association under the heterozygote model (OR = 1.368, 95 % CI 1.056- 1.772, P=0.018). Also, in the stratified analysis by ethnicity, no significant association of this polymorphism with risk of OC was found in the Asian and Caucasian populations. However, there was not significant
heterogeneity between heterozygote genetic model (P = 0.15, I2 = 33%) in Caucasian. Therefore, we utilized the fixed-effect model to merge OR value.
Conclusion
Based on the available evidence, no association between RFC1 G80A Polymorphism and ALL risk was observed, even in the subanalysis by ethnicity. The direction of further research should focus not only on the simple relationship of RFC1 G80A Polymorphism and ALL risk, but also on gene–gene and gene-environment interaction.