Although a number of genetic studies have attempted to link transforming growth factor beta 1 gene (TGFB1) -509 C>T polymorphism to breast cancer, the results were often irreproducible. We therefore aimed to meta-analyze all available case-control studies from the English-published literature to explore the association of this polymorphism with breast cancer. A total of 6 studies with 9 populations involving 10,197 patients and 13,382 controls were identified as of February 20, 2010. A random-effects model was performed irrespective of the between-study heterogeneity. Study quality was assessed in duplicate. The frequencies of TGFB1 -509 T allele in patients and controls ranged from 21.72 to 51.74%, and 24.53 to 52.40%, respectively. The presence of -509 T allele conferred a nonsignificant protective effect on breast cancer [odds ratio (OR) = 0.99; 95% confidence interval (CI) 0.93-1.05; P = 0.72]. This lack of association persisted under co-dominant, dominant, and recessive models. However, exclusion of the initial study significantly strengthened the magnitude of this protective effect. For example, under the dominant assumption, carriers of -509 T allele had a moderate reduced risk for breast cancer compared with the -509 CC homozygous (OR = 0.94; 95% CI 0.88-1.00; P = 0.04). Subgroup analyses by study designs and geographic areas did not substantially affect the present associations. No publication biases were observed by the fail-safe number. Taken together, our results demonstrated that TGFB1 -509 T allele was associated with a reduced risk to develop breast cancer and this allele appeared to act in an additive mode.