CDKN1B V109G Polymorphism And Cancer Risk:A Systematic Analysis

Currently, cancer has become as one of main threaten to human health. According to a recent estimates from International Agency for Research on Cancer (IARC), the annually increasing percentage for the incidence of cancer will be3%-5%all over the world in the future. By2020, there will be up to20million new cases worldwide each year, and12million deaths. The incidence of cancer in low-income developing countries is much higher than developed countries such as Britain and America. And our country appears to be on a trend of high incidence of cancer. The development of cancer was found to be associated with the environment, the polluted foods, physical damage, deficiency of nutrients, and the consumption of alcohol and tobacco. However, only a subset of individuals exposed to the above listed exogenous risk factors will develop cancer. It suggests a role of endogenous factor in the cancer development.It is the most basic characteristic to proliferate with abnormal cell cycle regulation for the cancer cells, leading to uncontrolled tumor cell growth. Therefore, the tumor is reconized as the cell cycle disturbing disease. Disorder of cell cycle regulation is considered of great importance during tumorgenesis.Protein CDKN1B coden by cyclin-dependent kinase inhibitor (CDKI) belongs to the cell cycle protein dependent protein kinase cyclin-dependent protein kinases (Cdks) inhibitor family. It has the function of the inhibiting tumor cell growth in many tumor developing process, is reconized as one of tumor suppressor gene. Gene polymorphism refers to the two or more than two alleles in the same site at the same time, and all the alleles are of high frequency (>5%). CDKNIB is highly conservative but with single nucleotide polymorphisms (SNP), including V109G which may cause functional variation of CDKNIB, and in turn leads to change of cancer susceptibility.Current research has found that one of SNP in CDKNIB gene occurs at codon109, converting from thymidine to guanine, which resulting in the transition of coding amino acids from valine (V) to a glycine (G). p38proteins encoded by Jabl genes can bind to CDKN1B, and promot CDKN1B phosphorylation, phosphorylated CDKNIB will be translocated from nucleus to cytoplasm, thus further degradated by ubiquitin-proteasome, the codon109is located in the p38binding region of CDKNIB. Therefore CDKN1B gene V109G gene polymorphism may produce a different biological characteristic of CDKNIB.Considering this, many domestic and international case-control studies were conducted to explore the relationship of V109G polymorphism of CDKN1B gene with susceptibility of prostate cancer, breast cancer, lung cancer, esophageal cancer, head and neck squamous cell carcinomas and other common tumor (18-23). Howerver, conflicting or even contradict results confused us on the role of V109G polymorphism on cancer susceptibility.So in this article, based on publicaly published case-control studies, we conducted an systemetic analysis for CDKN1B V109G and its relationship with malignant tumor susceptibility.ObjectsTo explore relationship of CDKNIB V109G gene polymorphism and risk of common cancers, level of epidemiological evidence, which will offer proof for tumorgenesis mechanism associated with CDKNIB V109G polymorphism, favor the interventions and prediction of pertinent cancer, and provide basis for clinical decision making. MethodsSearching the PubMed, CENTRAL, EMBASE, CNKI database, retrieving literature on V109G CDKNIB gene polymorphism and human tumor (breast cancer, prostate cancer, thyroid papillary carcinoma, esophageal cancer). At the same time we refer to article references or seek help from the authors for original data. Collecting data of literature including the first author, published year, race, nationality, tumor type, research type, sample numbers of control and case group and tumor risk ratio. Setting up the including and excluding criteria. To perform meta-analysis when at least two studies were available.Using software Statal1.0and RevMan5.0to research and statistically analyze the data obtained. The primary choice of statistical indicators is the odds ratio (OR), together with95%confidence interal to evaluate associated strength of CDKNIB gene polymorphism and the overall risk of tumor. Stratifiedly analyze different type tumor, gender, ethnicity. For ordinary genetic variations, we also take the dominant model and recessive model to re-evaluate. If some irrelevant variations turn out or lost in the data, then dominant model was utilized to assess. If the data allow, subgroup analysis was performed on ethnicity, age, living-habits and so on.Meta-analysis results of evaluation including heterogeneity analysis, sensitivity analysis and evaluation of publication bias. Heterogenity between studies was assessed by Cochran Q test, and the p value less than0.01indicated significant heterogeneity[24]. I2was used for quantitative expression of heterogeneity, it is commonly believed that I2value less than25%indicates mild heterogeneity, I2between25%to50%indicates moderate heterogeneity, I2value greater than50%shows a greater heterogeneity. If greater heterogeneity exsist between studies (I is more than50%), data were merged with a random effect model; If heterogeneity is small (I2less than50%), fixed effect model was applied. Restricted Maximum Likelihood Estimation of meta regression was applied to evaluate potential heterogeneity between the studies for the real impact on the variable. Sensitivity analysis was used to evaluate the stability of our meta-analysis results and the possible bias. According to the heterogeneity inspection results, choose the fixed effect model or random effects model to combine the OR and95%CI; If the merger effect didn’t changed before and after one literature was removed, then the results in this literature were stable and believable, otherwise indicated potentially bias, and may be the origin of the different conclusion. Using the funnel chart and Egger’s linear regression method put forward by Harbord and et al to determine effects of publication bias on combined results. In addition, we performed analysis by randomly remove one research to estimate the degree of stability, if the results fall outside of the95%confidence interval after removal, then the study may potentially give negative influence.ResultsA total of584references were retrived. Through screening,17articles involving V109G CDKN1B gene polymorphism and tumor susceptibility case-control studies were included, including5230tumor patients and5597health control.1. The characteristics of included studisAll were case-control studies, including four breast cancer research, two lung cancer research, two prostate cancer research,2esophageal cancer research.4researches were based on the census samples, and12samples were from the hospital, one from the community. On the other hand, of the17study,7took white as the research object, the remaining10took Asian yellow as the research object.12of17studies applied PCR-RFLP technique to analysis SNP. Match factors for control were sex, age and the genotype distribution consistent with Hardy-Weinberg equilibrium.2.Quantitative evaluation1) We found no obvious correlation between CDKN1B V109G gene polymorphism with all above tumors as an entirety. When analysis heterogeneity, only the dominant genetic model showed significant heterogeneity (P<0.01), while homozygous model (P=0.46) and the recessive model (P=0.61) showed no obvious heterogeneity. So we using the random effects model to incorporate dominant genetic model, the final results show that the G allele can reduce the susceptibility of cancer (OR=0.84,95%CI=0.611.19, P=0.05). After removal of3studies by Liu F, Francisco G and Daniela Pasquali which do not conform to the HWE or HWE can not be obtained, re-evaluation results show no correlation of G allele with tumor susceptibility (OR=0.90,95%CI=0.82-0.99, P=0.00), GG vs TT (OR=1.09,95%CI=0.861.40, P=0.47), and recessive model GG vs TT+TG (OR=1.13,95%CI=0.891.44, P=0.31). Further stratificated analysis on ethnicity also showed no evidence that CDKN1B V109G gene polymorphism has substantial influence on cancer risk for white and Asian yellow race (Overall OR=1.11,95%CI:0.941.31).2) Meta analysis on single type of tumor showed that CDKN1B V109G gene polymorphism has no influence for risk of breast cancer (OR=0.99,95%CI=0.851.15, P=0.90) and non-small cell lung cancer (OR=1.27,95%CI=0.672.41, P=0.46). Subgroup analysis such as age, gender, also did not showed significant difference.3) T transiton to G can reduce tumor susceptibility for prostate cancer (OR=0.60,95%CI=0.360.98, P=0.04) and esophageal cancer (OR=0.34,95%CI=0.220.55, P<0.01) when appling dominant model (TG+GG genotype compared with wild type TT). Similarity is also mentioned that CDKN1B V109G gene polymorphism can reduce ovarian cancer, malignant melanoma, medullary thyroid carcinoma and carcinoma of gastric cardia susceptibility.4) T transiton to G can increase the susceptibility of head and neck squamous cell carcinoma and hepatocellular carcinoma (HCC).Publication biasWhen analysis the CDKN1B V109G gene polymorphism and the overall prevalence of tumor, the shape of the funnel figure showed asymmetry, indicated potential studies were not publishe for less sample size or no statistical significance. All case-control studies based on for meta-analysis were tested to be in normal distribution, and heterogeneity is in low level, so Egger’s test was appled to inspect publication bias. Egger’s test showed:t=0.65, P=0.527for TT vs TG/GG, t=0.92, P=0.375for t-allele vs. the G allele. The results can still not confirmed the correlation of CDKN1B V109G gene polymorphism and the overall prevalence of cancer.