Numerous studies investigated the associations of VDR polymorphisms with various types of cancer, suggesting an influence on cancer risk. FokI is one of the most frequently analysed polymorphisms but the results from single studies are contradictory. We performed a meta-analysis looking at the association between the FokI and all cancer sites and investigating sources of heterogeneity. We identified 77 independent studies up to April 2014. We presented the summary odds ratios (SORs) by cancer sites, ethnicity and study features. We found a significant association between FokI and ovarian cancer for ff genotype versus (vs) FF with no heterogeneity: SOR=1.20 (95%CI: 1.02-1.41, I(2)=0%). Moreover, we found a significant increased risk of any cancer: SOR=1.08 (95%CI: 1.01-1.16, I(2)=58%). A significant increased risk of any cancer is confirmed among Caucasian, among studies in Hardy-Weinberg (HW) equilibrium and among nested case-control studies. Furthermore among studies in HW equilibrium, skin cancer was found significantly associated with FokI: SOR=1.24 (95%CI: 1.01-1.54; I(2)=24%) for ff vs FF. The estimated number of cases attributable to ff genotype is 4221 for ovarian cancer and 52858 for skin cancer worldwide each year. No indication for publication bias was found for any cancer site. In conclusion, we found an overall significant association of FokI polymorphism with any cancer, with differential effect by ethnicity. In particular the summary estimates indicate an increase risk for ovarian and skin cancer for ff vs FF. However other factors may act modifying the association, and further studies are needed to clarify the impact on cancer risk.