Sir,

I read the article by Lajin and Alachkar (2013) with great interest, which appeared in your journal as Br J Cancer. The study aimed to evaluate the impact of the NQO1 gene polymorphism with cancer risk. Although the study provides preliminary evidence to consider NQO1 polymorphism as a risk factor for cancer; however, after careful reading of the article, a few important issues came out that must be addressed for further actions.

First, it appears that the authors somehow missed the statistical power in this study. Sample sizes remain a major issue in genetic case–control studies analysing the association of polymorphism with disease susceptibility. The authors did neither mention the statistical power of individual studies nor their overall meta-analysis. Hence, the study should obtain an adequate statistical power (80%) to estimate significant association accurately, which remains a primary criterion to perform such studies, especially from the venous blood of study subjects. Underpowered studies usually lead to false-positive associations and misinterpretations (Hattersley and McCarthy, 2005). The authors also failed to mention the incidence rate of various cancers in the said study. The individual studies recruited in the present meta-analysis achieved the required statistical power is questionable and did not discuss in the text.

Second, the authors mention the sample size of Malik et al as 107 gastric cancer cases and 195 controls (Malik et al, 2011). But the exact number of gastric cancer cases is 108 in the study by Malik et al. All these points suggest a thorough examination of the association observed in the said study, and must be clarified before concluding that NQO1 gene polymorphism is a potential marker of cancer.