Abstract
To the Editor:
Thank you very much for forwarding me the letter to the editor concerning our work. After reviewing their comments and checking the original paper (Li et al., 2011) and the raw data carefully, we are glad to reply to their questions.
First of all, in the Identification and eligibility of relevant studies section, the searched databases should be changed according to Mi et al.'s (2012) suggestion. The China National Knowledge Infrastructure, WanFang, and WeiPu databases in China should be added to this section.
Second, there were four different case-control studies involving different types of cancers (cervical, gastric, ovarian, and breast), all performed by Araújo et al. (2009a, 2009b, 2011a, 2011b), the same as Kang et al. (2007a, 2007b), so the overlapping data of cases did not exist. However, the number of controls in the six studies just mentioned may have some overlapping data, which was one of the limitations in our meta-analysis.
Third, in the part of the statistical analysis section, we referred to one publication by Handoll (2006), which considered that a p-value>0.05 rather than >0.10 for the Q-test indicated a lack of heterogeneity among studies.
Fourth, the distribution of genotypes in all the controls was in agreement with Hardy–Weinberg equilibrium (p>0.05), except for four studies in this meta-analysis (Shahbazi et al., 2002; Vauleon et al., 2007; Abu Dayyeh et al., 2011; Araújo et al., 2011b). We included these four studies in our meta-analysis, which was also a limitation. If we deleted these four studies, the overall association between epidermal growth factor (EGF)+61A/G polymorphism and cancer risk was similar.
Finally, the Egger's test (Egger et al., 1997) was used to provide statistical evidence of funnel plot symmetry. When we performed the Egger's test, some significant publication bias was detected. Further studies based on larger sample size are still needed to assess the association of EGF gene polymorphisms with susceptibility to cancer risk, especially for −1380A/G, −1744G/A, and rs698326T/G polymorphism.