Introduction. Hepatitis B virus (HBV) infection is major health problem around the world. Chronic HBV infection is a life-threatening condition when the disease progresses to liver cirrhosis or hepatocellular carcinoma. Yet it is not well understood why some individuals are chronically infected whereas the others can spontaneously resolve the infection. Recent studies analyzed the association between genetic polymorphisms and the chronicity of HBV infection with conflicting results.
Aim. The objective of this study was to summarize from existing literature the association between genetic polymorphisms and the predisposition to chronic HBV infection comparing chronic carriers with spontaneous resolvers.
Methods. Review of the literature on the subject was performed by using three online databases: PubMed, Embase and ISI Web Of Knowledge. Keywords composed of free terms and/or thesaurus were selected with stringent criteria to build search equations. Pertinent materials were further selected by using a three-step process. The first articles were screened by title, the second by abstract and the third by full-text. Study characteristics, allelic counts and genotype distribution were collected from the articles. To make the results uniform and thus comparable, the univariate unconditional dominant model was chosen as reference for genetic associations. A meta-analysis was performed to assess the strength of the association.
Results. In total 126 articles out of 4901 passed selection and were included in the systematic review. After removing duplicates, articles with other language than English and other articles than original papers, 3314 articles were remaining. 2452 of them were removed based on their title, then 360 based on their abstract and finally 376 based on their full text. Most studies included individuals of Asian ethnicity. The median date of publication was 2010. A total number of 396 polymorphisms were analyzed in 117 genes, with 47 polymorphisms analyzed more than once and 349 of them analyzed only once. Among polymorphisms analyzed once, 45 showed significant p-values (p<0.05), with rs2071543 in LMP7 and rs1799988 in CCR5 presenting P values <1E-05. Among polymorphisms analyzed more than once, 15 polymorphisms among 13 genes showed significant association considering the whole population: TNF, a locus near HLA-C, UBE2L3, MCP1, STAT4, CCR5, IL18, TAP1, GNLY, IFNL, IL10, IFNG, and CTLA4. Among them, rs1800630 in TNF (OR=1.38, 95% CI 1.15-1.65), rs3130542 in a locus near the HLA-C gene (OR=1.42, 95% CI 1.23-1.64) and rs482116 in UBE2L3 (OR=0.77, 95% CI 0.66-0.89) showed the most consistent association. Other associations are less robust.
Conclusion. To our knowledge, this is the first such exhaustive systematic review and meta-analysis that has been performed on the subject. Although large number of polymorphisms in immune genes were studied, most of the significant associations were either reported only once without further replication or not equally validated. This may be due to the limitations of the studies, such as different ethnicity of study population, the choice of study subgroups, study design. Further genetic studies are needed to verify those findings.