Sir,

We have strictly followed the published general guidelines for conducting meta-analyses (Stroup et al, 2000; Minelli et al, 2009). Therefore we tested for deviations from HWE and reported in our paper the deviation of five studies from HWE. Deviation from HWE may imply genotyping error and/or possible heterogeneity in the control population. Most of the excluded studies strongly violated HWE with P-values=0. Nevertheless, when the five studies were included in the meta-analysis the association remained statistically significant with almost unaltered odds ratios. For example, for the TT vs CC model OR=1.17 (1.06–1.30), P=0.002, compared to the reported results with the exclusion of the five studies, OR=1.18 (1.07–1.31), P=0.002. This is not unexpected given the large number of studies included and the fact that the majority of studies were in compliance with the HWE principle. This confirms that no bias has been introduced by the exclusion of the five studies and that the concluded positive association between total cancer risk and the investigated polymorphism is unaffected.

The distinction between the source of controls was not made unambiguously in all published papers. Therefore, we avoided stratification according to the source of controls. In addition, we believe that such stratification is irrelevant within the context of our meta-analysis and that there is no difference in the genotypic distribution between hospital-based and population-based controls since it is assumed that the control subjects in all studies were not diagnosed with any type of cancer or any other condition commonly associated with the studied polymorphism.

Finally, we believe that it is unlikely that the inadvertently missed single-case sample had an impact on the results and conclusions of the meta-analysis that included 21178 case samples.