IGFBP3 polymorphisms and risk of cancer: a meta-analysis.

Until now, there were several studies evaluating the association between the polymorphisms in the IGFBP3 gene and cancer risk in diverse populations and in multiple types of cancer, but their outcomes have been contradictory and need to be investigated further. Here, we performed a meta-analysis from all eligible case-control studies to address the association of IGFBP3 A-202C and Gly32Ala polymorphisms to cancer. 20 articles including 41 studies for A-202C variant including 28,322 cancer patients and 36,772 healthy controls and six articles for Gly32Ala variant including 4,477 cases and 5,443 controls were selected in our analysis. Overall, A-202C polymorphism was appeared to be a risk factor of cancer (OR = 0.98, P = 0.05). A allele of IGFBP3 A-202C SNP was significantly less common in the cancer patients than in controls and AA genotype significantly decreased the cancer risk in additive genetic model when comparing to CC genotype (OR = 0.93, P = 0.004). Another SNP, Gly32Ala, seemed to be in linkage equilibrium with A-202C SNP. However, no significance was found when we analyzed the relation of cancer risk and Gly32Ala polymorphism (OR = 0.93, P = 0.36). Further, we compared the distributions of A-202C SNP in different types of cancer, significant association was found in additive genetic model in breast cancer (OR = 0.93, P = 0.01) and prostate cancer (OR = 0.88, P = 0.05). In the analysis of the variants in different population, A-202C variant was significantly associated with cancer risk in Africans (OR = 0.90, P = 0.05), but not in Caucasians (OR = 0.98, P = 0.12) or in Asians (OR = 1.03, P = 0.61). These results indicated that polymorphisms of IGFBP3 might have different effect in different types of cancer and different population. Further large study combining both IGFBP3 A-202C and Gly32Ala SNPs on different types of cancer in different populations were needed to validate former results.