Note of clarification of data in the meta-analysis of evidences on XPC polymorphisms and lung cancer susceptibility.

Abstract

Dear Editor, We read with great interest the paper [1]. The author conducted a meta-analysis of 13 case–control studies to estimate the association of Ala499Val, Lys939Gln, and PAT polymorphisms in XPC gene with lung cancer. His meta-analysis think that there is no significant association of these polymorphisms with lung cancer. It is a valuable study. Nevertheless, careful examination of the data provided by the author (Table 1 in the original text) revealed some issues that are worth noticing. First, the ethnicity is inconsistent with the original studies. One study was a population-based case–control study of lung cancer in Xuan Wei, China [2]. And the other study examined variants in six NER genes (include XPC) in association with primary lung cancer risk among 113 Latino and 255 African American [3]. But they were used as Caucasian for data analysis. Second, the same text data referenced twice. There are two articles that have the same author, same writing time, exactly same case–control samples. After checking carefully the original article, we found they are two versions of an article, a publication in Chinese journal [4], the other is included in the foreign journal [5]. Therefore, we reassessed the association between XPC Ala499Val, Lys939Gln polymorphisms, and lung cancer (result in Table 2). In addition, we update with a new article [6] about the association between Lys939Gln polymorphisms and lung cancer, and another hotspot PAT polymorphism which is in XPC gene to our meta-analysis (Table 3) [7–14]. Though the Table 4 showed that there was no significant difference for association of XPC Ala499Val polymorphisms with lung cancer under the additive model, dominant model, and recessive model. However, there was an association of PAT polymorphisms with lung cancer in Caucasian which was under recessive model (OR=1.307; 95 % confidence interval (CI), 1.037~1.648; p=0.023) (Fig. 1). The people with Glu/Glu had more inflexibility compared with the Lys-carriers, according to the recessive model (OR=1.257; 95 % CI, 1.038~1.522; p=0.013) (Fig. 2).