Introduction
Colorectal cancer (CRC) has been an public health problem for many years because of increasing incidence and high mortality rate despite some overall improvements in diagnosis and treatment [1, 2]. It is a common form of cancer, with more than 1.5 million new patients diagnosed every year worldwide [3] and more common in developed than developing countries [4]. Previously published studies have indicated that age over 60 years, alcohol consumption and tobacco habits may contribute to sporadic CRC risk [5, 6]. Moreover, the development of CRC is widely considered as a multi-step, multi-factorial process involving gene-gene and gene-environment interactions [7]. It has been suggested that inter-individual differences including single nucleotide polymorphism (SNP) might influence CRC risk [8, 9] .
The cytochrome P450 enzymes (CYPs) take part in the metabolism of carcinogenic compounds in the human body [10]. CYP1A1 is expressed in the large bowel [11]. Because of the involvement of CYP1A1 in the metabolism of PAHs and estrogens, and perhaps cruciferous vegetables and heterocyclic amines [12], it might be expected that CYP1A1 variants might influence the risk of colorectal cancer. A commonly studied SNP in the CYP1A1 gene has been indicated to associate with CRC risk, which was localized on chromosome 15q22 [13]. The commonly reported the 3801T>C (also referred to as 2A, m1, or rs4646903) and A2455G polymorphisms (also known as E7, *2C, m2, rs1048943) with CRC risk [14]. The T3801C and A2455G polymorphisms might change the level of gene expression or messenger RNA stability, resulting in a highly inducible activity of the enzyme [15].
To date, 37 publications (supplemental Table 2) and seven previous meta-analyses [16-22] have been reported to investigate the correlation between the CYP1A1 T3801C and A2455G polymorphisms with CRC risk. However, the results were inconsistent or even contradictory. In addition, a lot of original studies did not involve in previously published meta-analyses (supplemental Table 1). Moreover, previously published meta-analyses also did not evaluate positive results to identify multiple comparisons. Therefore, we performed an updated meta-analysis and systematically evaluated the published meta-analyses on the association between CYP1A1 T3801C and A2455G polymorphism with CRC susceptibility.