Artikel
A Systematic Review on the association between cervical cancer and TP53 codon 72 polymorphism
Suche in Medline nach
Autoren
Veröffentlicht: | 14. September 2004 |
---|
Gliederung
Text
Introduction
Cervical cancer is still the second most common cancer related cause of death in women worldwide. An infection with human papilloma virus (HPV) has been identified as a necessary cause of cervical cancer. However, the majority of HPV infected women never develop cervical cancer. Other factors, such as genetic disposition, could be of importance. In 1998, Storey and colleagues found a 7-fold increased risk for cervical cancer in women with arginine homozygosity at codon 72 on the tumorsuppressor gene TP53. This polymorphism results in translation of either the amino acid arginine (CGC) or proline (CCC) in the p53 tumorsuppressor protein. The two protein variants have different configurations with different properties. It has been shown that the p53 arginine protein variant is more sensitive for degradation by the HPV E6 tumor protein. Since 1998 several studies have been published with controversial results about the association of TP53 polymorphism and cervical cancer. We carried out a systematic review to find out if there is an increased risk for developing cervical cancer in women homozygous for arginine at codon 72 of the TP53 gene.
Methods
A systematic literature search was performed in different databases as well as in 'grey' literature to find all studies published on the association of TP53 polymorphism and cervical cancer between 1998 to 2003.Publication language was no exclusion criteria. Data were extraction systematically from all studies which investigated the association betweenTP53 polymorphism and cervical cancer.For each study, odds ratios were recalculated for homozygous arginine versus homozygous proline genotype, since in the original publications different reference groups were used making direct comparison impossible. Studies with unsuitable controls and studies in which cases and/or controls homozygous for proline were lacking were excluded. Departure from Hardy-Weinberg equilibrium in each study was tested by Chi-square. The frequency of the arginine encoding allele was investigatedin different ethnic groups. Possible publication bias was investigated. Pooled estimates and test for heterogeneity were calculated based on our recalculated odds ratios by the data included in the review.
Results
64 publications from 60 authors were identified: two publications were in Spanish, one in Chinese, the remaining were written in English. Overall 15 publications were excluded for different reasons, 48 studies were included in the review. In 73% of the included studies, the study populations was in Hardy-Weinberg equilibrium. Publication bias seems to have occur only for publications with a study size smaller than 100 women. The frequency of the arginine allele was highest in Caucasian (71%) and lowest in African women (36%). Meta-analysis with random effect model was performed for arginine homozygous versus proline homozygous women. The pooled estimate was 0.93 [0.83-1.03]. Quality of individual studies was different, however results of sensitivity analysis did not reveal major deviation.
Discussion
Our review shows that TP53 codon 72 polymorphism is not involved in the development of cervical cancer. However, such a comparison of published molecular epidemiological studies has several limitations. Quality of the individual studies, specially in respect to epidemiological criteria vary greatly. Systematical bias in some studies can not be excluded. The main problems were selection of controls, comparison of different ethnic groups and use of different biological tissues and molecular detection methods. All these issues lead to a certain degree of heterogeneity between the studies.