AUTHOR=Petros Zelalem , Kishikawa Junko , Makonnen Eyasu , Yimer Getnet , Habtewold Abiy , Aklillu Eleni 

TITLE=HLA-B*57 Allele Is Associated with Concomitant Anti-tuberculosis and Antiretroviral Drugs Induced Liver Toxicity in Ethiopians

JOURNAL=Frontiers in Pharmacology

VOLUME=8

YEAR=2017

URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2017.00090

DOI=10.3389/fphar.2017.00090

ISSN=1663-9812

ABSTRACT=<p>Drug-induced liver injury (DILI) is a known adverse effect of both anti-tuberculosis (anti-TB) and antiretroviral (ARV) drugs. Recent studies highlight the implications of genetic predispositions to DILI. We performed a case-control study to identify Human Leukocyte Antigen-B (HLA-B) variant alleles associated with anti-TB and ARV co-treatment induced liver toxicity in Ethiopian TB and HIV co-infected patients. A total of 495 newly diagnosed TB and HIV co-infected patients were enrolled and received rifampicin based anti-TB and efavirenz based ARV therapy. Change in liver enzyme level from baseline was monitored 1st, 2nd, 4th, 8th, 12th, and 24th weeks after treatment initiation to identify patients who developed DILI (cases) and those who did not (treatment tolerants). Genomic DNA from 46 cases and 46 sex and age matched treatment tolerants were genotyped for HLA-B variant alleles using Olerup SSP®HLA-B DNA Typing Kits. The proportion of <italic>HLA-B</italic><sup>*</sup><italic>57</italic> allele carriers in DILI cases (37.0%), particularly in those who developed cholestatic type of DILI (44.8%) was significantly higher compared with those who tolerated the treatment (2.2%). The <italic>HLA-B</italic><sup>*</sup><italic>57</italic> allele frequency was significantly higher in cases (25%) than treatment tolerants (1.1%). In a multivariate logistic analysis, the proportion of patients carrying <italic>HLA-B</italic><sup>*</sup><italic>57</italic> (<italic>P</italic> = 0.002) and <italic>HLA-B</italic><sup>*</sup><italic>14</italic> (<italic>P</italic> = 0.014) alleles were significantly higher in DILI cases compared with treatment tolerants. <italic>HLA-B</italic><sup>*</sup><italic>57</italic> was significantly associated with cholestatic (<italic>P</italic> = 0.001) and mixed (<italic>P</italic> = 0.017) types of liver toxicity, and mild-to-moderate severity (<italic>P</italic> = 0.001). Of all <italic>HLA-B</italic><sup>*</sup><italic>57</italic> alleles detected, <italic>HLA-B</italic><sup>*</sup><italic>57:03</italic> accounted 58.3% and <italic>HLA-B</italic><sup>*</sup><italic>57:02</italic> accounted 41.7%. <italic>HLA-B</italic><sup>*</sup><italic>57:01</italic> was not detected. The variant allele frequencies of <italic>HLA-B</italic><sup>*</sup><italic>57:03</italic> (15.2 vs. 0%) and <italic>HLA-B</italic><sup>*</sup><italic>57:02</italic> (9.8 vs. 1.1%) were significantly higher in the DILI cases than treatment tolerants (<italic>P</italic> &lt; 0.03). We conclude that <italic>HLA-B</italic><sup>*</sup><italic>57</italic> alleles (<italic>B</italic><sup>*</sup><italic>57:03</italic> and <italic>B</italic><sup>*</sup><italic>57:02</italic>) confer susceptibility to the development of anti-TB and ARV drugs co-treatment induced liver toxicity, which is mainly of cholestatic type. The possible association of <italic>HLA-B</italic><sup>*</sup><italic>14</italic> with anti-TB and ARV drugs co-treatment induced liver toxicity requires further investigations.</p>