Background: Single nucleotide polymorphisms (SNPs) occurring in Toll-like receptors (TLRs) may contribute to cancer risk. Many polymorphisms of TLR2 have been studied for associations, but the findings are conflicting. Methodology/Principal Findings: We performed a meta-analysis of 14 studies to confirm the association between TLR2+597T>C (rs3804099), +1350C>T (rs3804100) and Arg753Gln (rs5743708) polymorphisms and cancer risk. Odds ratio (OR) and 95% confidence intervals (95% CI) were used to assess the strength of associations. There was no significant association between TLR2+597T>C and cancer risk in the codominant models (CC vs. TT: OR = 1.01, 95%CI = 0.86-1.17, $P_{heterogeneity}=0.148$; CT vs. TT: OR = 0.92, 95%CI = 0.69-1.23, $P_{heterogeneity}$ < 0.001), the recessive model (CC vs. CT+TT: OR = 0.86, 95%CI = 0.67-1.10, $P_{heterogeneity}=0.007$), the dominant model (CC+CT vs. TT: OR = 0.93, 95%CI = 0.76-1.15, $P_{heterogeneity}=0.001$) and the allele model (C vs. T: OR =0.93, 95%CI = 0.81-1.08, $P_{heterogeneity}=0.019$). Similarly, no significant associations between TLR2+1350C>T, Arg753Gln polymorphisms and cancer risk were found. However, in the sub-group analysis of ethnicities, the trend of pooled ORs in Asians was opposite to Caucasians. Conclusions: The present meta-analysis suggests that TLR2+597T>C (rs3804099), +1350C>T (rs3804100) and Arg753Gln (rs5743708) polymorphisms are not associated with cancer risk.