Genetic Polymorphisms Of Drug-metabolizing Enzymes And The Susceptibility To Antituberculosis Drug-induced Liver Injury

Objective: Tuberculosis (TB) is a chronic respiratory disease caused byMycobacterium tuberculosis. China has a high incidence of tuberculosis (TB), rankingsecond in the world in terms of number of cases. Every year more than1milliontuberculosis patients are receiving the standard short-course anti-TB treatment providedby China National Tuberculosis Prevention and Control Scheme (CNTS) in China. The3first-line drugs can cause diverse adverse reactions (ADRs). Anti-TB drug-induced liverinjury (ATLI) is the most prevalent and serious ADR encountered in the course of TBtreatment. ATLI would increase the patient suffering and incur substantial additionaldisease burdens, and may eventually contribute to treatment failure, hence affecting theimplementation of CNTS. To explore the pathogenic mechanism of ATLI would bemeaningful for individual treatment and the implement of the national anti-tuberculosistreatment project. In Xinjiang, the incidence of TB, the prevalence of smear positive TBand the TB mortality rate are higher than the national average.The Directly ObservedTreatment Short-course (DOTS), recommended by World Health Organization (WHO)covers the entire population of Xinjiang. The Xinjiang population are particularly at risk,but the prevalence of drug-induced liver injury, and associated risk factors, in thispopulation have not been previously reported. We aimed to determine the incidence ofATLI and the relevant factors inducing ATLI, and to investigate the association of geneticpolymorphisms of drug-metabolizing enzymes (DMEs) with hepatotoxicity, and theirpotential use for identifying patients at increased risk of liver injury. For ATLI as acomplex disease which arises from genetic and environmental factors, further study ongene-environmental factors and gene-gene interaction would be important forilluminating the pathogenic mechanism of ATLI, and the interaction among DMEs genetic polymorphisms and other possible environmental factors on ATLI was alsoassessed. The present study would provide a scientific basis for illuminating thepathogenic mechanism of ATLI. Methods: From January2010to August2012, A total of2811pulmonary TB patients (more than14years old) who fit the criteria and werereceiving standard short-course chemotherapy were consecutively enrolled at the TBinstitution of center for disease control and prevention. After the subjects signinginformed consent, patients were interviewed using a structured questionnaire anddemographic information (gender, age, ethnicity), body weight and height, TB treatmenthistory, anti-tuberculosis treatment induced ADRs, clinical symptoms, and risk factors forliver disease, was recorded. At the end of two months of anti-tuberculosis treatment,serum ALT, AST and total bilirubin levels were measured and a sample of each patient’sblood (including the patients of discomfort symptoms during the two months) was storedfor genotypic analysis. ATLI was measured and scaled according to the criteriaformulated by Chinese State Food and Drug Administration (SFDA). All of thedemographic, clinical, environmental data would be merged for a comprehensive analysis.The incidence of ATLI and the relevant factors inducing ATLI were analyzed. Of2811pulmonary TB patients,2244Uyghur patients were assessed.The genes of NAT2,CYP2E1RsaI, GSTM1, GSTT1from89TB cases with ATLI and1858TB cases withoutATLI were amplified and sequenced.Their genotypes were determined by PCR-LDR andthe genotype frequencies were compared between cases and controls.The relationshipbetween gene polymorphisms and ATLI was analyzed by applying single factor analysisand multivariable logistic regression, and their potential use for identifying patients atincreased risk of liver injury were predicted with screening analysis.The interactionamong DMEs genetic polymorphisms and other possible environmental factors on ATLIwas also assessed by case only analysis, crossover analysis and generalizedmulti-dimensionality reduction analysis. Results:1) This study included a total of2842pulmonary TB patients. Of the2842patients,31patient’s blood samples were lost, thefinal number of study population was2811.The mean age of subjects was43.09years(SD:18.32). In all the2811patients, male accounted for57.5%, women for42.5%; newtreatment for91.4%, retreatment for8.6%; Uyghur for79.8%, Han for10.9%, Kazak for2.8%, the other for6.5%; the South for61.3%, the north for38.7%.2) Of all the2811TBpatients,359patients had abnormal levels for liver function tests and the incidence ofabnormal liver function was12.7%;110patients were diagnosed with ATLI, and theoverall incidence of ATLI was3.9%. Among the incidence of ATLI, male was4.7%, female2.8%; Han3.9%, Uyghur4.0%, Kazak2.5%, the other3.8%;<40years4.5%,40-years3.2%,60-years3.6%; the South4.5%, the north3.0%; new treatment4.1%,retreatment2.1%.3) In terms of the severity of110ATLI cases,65(59.1%) cases hadmild hepatotoxicity (serum transaminase levels two to three times of the upper limit ofnormal),31(28.2%) had moderate hepatotoxicity (serum transaminase levels three tofive times of the upper limit of normal) and14(12.7%) had severe hepatotoxicity (serumtransaminase levels more than five times of ULN).4) In2811TB patients, single factoranalysis results demonstrated that difference of indices such as gender, whether usehepatoprotective drugs, therapeutic regimen, and body weight, was statisticallysignificant between the ATLI case group and the control group.Male’s ATLI incidencerate was higher than female’s, incidence rate of patients with using hepatoprotectivedrugs was higher, incidence rate of patients with taking FDC drugs was not higher thananother therapeutic regimen (also called ‘blister pack drug’), incidence rate of patientswith body weight more than50kg was higher than patients with body weight less than50kg, and their corresponding OR values and95%CI were0.568(0.376～0.859),2.280(1.484～3.502),1.484(1.223～2.678),3.140(1.355～7.278), respectively.Multi-factorlogistic regression results demonstrated that gender, using hepatoprotective drugs andbody weight were related to ATLI, OR values and95%CI were0.461(0.239～0.888),2.491(1.329～4.670) and3.423(1.174～9.979), respectively; age was also associatedwith ATLI, compared with the young, and the OR values and95%CI of the middle-agedgroups and the old groups were0.255(0.105～0.616),0.639(0.324～0.324), respectively.5) In2811TB patients, logistic regression was applied to establish a model to predict thetendency of patient taking preventative liver protection drugs, using gender, age, ethnicity,drinking alcohol hobby et.al as covariates.The prediction probability was used astendency score and the population was weighted by Standardized Mortality RatioWeighting (SMRW) method. After weighting, statistical significances were not found forthe association between whether using liver protection drugs, taking Glucurolactone,Creatinine and liver tablets with ATLI occurrence. The ATLI incidence rate in using livertablets group was higher than not using protective liver drugs’, and the difference wasstatistically significant (P=0.002).6) In2244Uyghur TB patients, both single factoranalysis and multi-factor logistic regression results showed that the NAT2*5(481CT)genotype was associated with ATLI, with an increased prevalence in persons with the CTgenotype, compared to CC, with an adjusted OR of1.780,95%CI1.086～2.917, existingin the dominant model correlation with ATLI. Individuals may be rapid, intermediate or slow acetylators, according to their activity of NAT2. Both single factor analysis andmulti-factor logistic regression results showed that an increased prevalence in personswith the intermediate acetylators was higher than the rapid acetylators, the slowacetylators higher than the intermediate acetylators, and the difference was notstatistically significant.Among the slow-acetylator genotypes, the NAT2*5/6genotypewas most strongly associated with ATLI, the adjusted OR of3.417,95%CI1.411～8.276.Stratification analysis also showed that: the NAT2*5(481CT) genotype was associatedwith ATLI, with an increased prevalence in persons with the CT genotype, compared toCC among females, age less than60or more than35years old and the south population;in the north populatin, the homozygous mutations AA genotype of NAT2*7(857) geneticlocus was strongly associated with ATLI, compared to the wild genotype GG, an adjustedOR of23.280,95%CI1.651～328.344. For the haplotypes composed of481(rs1799929),590(rs1799930) and857(rs1799931), the haplotype C-A-G with590(rs1799930)mutation allele had higher risk for ATLI than their wild allele combination (C-G-G), anadjusted OR of2.174,95%CI1.365～3.461.7) In2244Uyghur TB patients, both singlefactor analysis and multi-factor logistic regression results showed that: there were nosignificant association between the genotype of CYP2E1RsaI and ATLI, and no statisticalassociation between the genotype of GSTM1null, GSTT1null and ATLI. Stratificationanalysis also showed no any associations between CYP2E1RsaI, GSTM1, GSTT1andATLI.8) In the89cases of the2244Uyghur TB patients, there were no significantassociation between NAT2acetylator, CYP2E1RsaI genotype, GSTM1genotype andGSTT1genotype and ATLI’s order of severity.9) In2244Uyghur TB patients, there wasno evidence showing that NAT2acetylator, CYP2E1RsaI genotype, GSTM1genotypeand GSTT1genotype would be useful for predicting ATLI.10) In2244Uyghur TBpatients, case-case analysis, crossover analysis and generalized multi-dimensionalityreduction analysis showed there was gene-environmental factor interaction betweenNAT2acetylator and drinking alcohol hobby; non-conditional logistic regression analysisand crossover analysis showed there was gene-gene interaction between CYP2E1RsaIgenotype and GSTT1genotype on ATLI. Conclusion:1) In this study population whowere assessed at the end of two months of anti-tuberculosis treatment, the incidence ofATLI was3.9%, and the incidence of severe ATLI was0.50%. Patients withasymptomatic ATLI accounted for33.6%. It is suggested that the monitoring of liverfunction in patients receiving TB treatment should be strengthened.2) This study foundthat ATLI may be associated with sex, age and body weight. Especially male TB patients with body weight more than50kg, and the smaller the age, the greater the risk of ATLI,should be closely observed the changes of liver function.3) In this study population,46.5%patients took different kinds of liver protection drugs, mainly includingGlucurolactone, Creatinine and liver tablets.The effect of liver protection drugs was notobserved, even might be a risk in patients using preventive liver protection drugs.Whether using preventive liver protection drugs in patients with normal liver function isappropriate and is indeed effective, should be confirmed by large-scale randomizedcontrolled trials.4) In Uyghur TB patients, liver injury was associated with the geneticvariant NAT2*5in the dominant model, the NAT2*5/6genotype among theslow-acetylator genotypes firstly reported by this present study.In Uyghur female patientsin the south ranged in age from35to60years old, carrying the NAT2*5*CT genotypewas significantly higher at risk of ATLI than the wild genotype CC; in Uyghur patients inthe north, carrying the NAT2*7*AA genotype was significantly higher at risk of ATLIthan the wild genotype GG.The haplotype C-A-G with590(rs1799930) mutation allelemay be risk factor of ATLI. There was no evidence that ATLI was associated with NAT2acetylator, CYP2E1RsaI genotype, or GSTM1/GSTT1null genotypes, and the geneticmarkers studied are unlikely to be useful for screening patients for identifying persons atrisk of liver injury.5) In Uyghur TB patients, there were interactions between NAT2acetylator and drinking alcohol hobby. Still there were interactions between CYP2E1RsaIgene and GSTT1gene, which should be further confirmed in large sample sizepopulation.