Next Article in Journal
Application of Patient-Derived Cancer Organoids to Personalized Medicine
Previous Article in Journal
Pharmacogenetics of Donepezil and Memantine in Healthy Subjects
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JPM
Volume 12
Issue 5
10.3390/jpm12050790
jpm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Clinical and Genetic Risk Factors for Drug-Induced Liver Injury Associated with Anti-Tuberculosis Treatment—A Study from Patients of Portuguese Health Centers

by
Maria João Cavaco
1,*,†,
Celeste Alcobia
2,3,†,
Bárbara Oliveiros
4,5,
Luís Alcides Mesquita
6,
Aurora Carvalho
7,
Fernando Matos
8,
José Miguel Carvalho
9,
Miguel Villar
10,
Raquel Duarte
11,
João Mendes
6,
Carolina Ribeiro
6,
Carlos Robalo Cordeiro
2,
Fernando Regateiro
6 and
Henriqueta Coimbra Silva
5,6
1
Oeste Hospital Center, 2560-295 Lisboa, Portugal
2
Department of Pneumology, Coimbra Hospital and Universitary Centre, 3004-561 Coimbra, Portugal
3
Pneumological Diagnostic Center of the Centre, 3000-075 Coimbra, Portugal
4
Laboratory of Biostatistics and Medical Informatics, Faculty of Medicine, University of Coimbra, 3000-548 Coimbra, Portugal
5
Coimbra Institute for Clinical and Biomedical Research (iCBR), Faculty of Medicine, University of Coimbra, 3000-548 Coimbra, Portugal
6
Institute of Medical Genetics/UC Genomics, Faculty of Medicine, University of Coimbra, 3000-548 Coimbra, Portugal
7
Department of Pneumology, Vila Nova de Gaia Hospitalar Centre, 4434-502 Vila Nova de Gaia, Portugal
8
Pneumological Diagnostic Center of Aveiro, 3810-042 Aveiro, Portugal
9
Pneumological Diagnostic Center of Santarém, 2005-324 Santarém, Portugal
10
Pneumological Diagnostic Center of Venda Nova, 2700-220 Lisboa, Portugal
11
Pneumological Diagnostic Center of Vila Nova de Gaia, 4400-088 Vila Nova de Gaia, Portugal
 
add Show full affiliation list
 
remove Hide full affiliation list
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
J. Pers. Med. 2022, 12(5), 790; https://doi.org/10.3390/jpm12050790
Submission received: 17 March 2022 / Revised: 8 May 2022 / Accepted: 10 May 2022 / Published: 13 May 2022
Browse Figure
Versions Notes

Abstract

:
Drug-induced liver injury (DILI) is an unpredictable and feared side effect of antituberculosis treatment (AT). The present study aimed to identify clinical and genetic variables associated with susceptibility to AT-associated hepatotoxicity in patients with pulmonary tuberculosis treated with a standard protocol. Of 233 patients enrolled, 90% prospectively, 103 developed liver injury: 37 with mild and 66 with severe phenotype (DILI). All patients with mild hepatitis had a RUCAM score ≥4 and all patients with DILI had a RUCAM score ≥ 6. Eight clinical variables and variants in six candidate genes were assessed. A logistic multivariate regression analysis identified four risk factors for AT-DILI: age ≥ 55 years (OR:3.67; 95% CI:1.82–7.41; p < 0.001), concomitant medication with other hepatotoxic drugs (OR:2.54; 95% CI:1.23–5.26; p = 0.012), NAT2 slow acetylator status (OR:2.46; 95% CI:1.25–4.84; p = 0.009), and carriers of p.Val444Ala variant for ABCB11 gene (OR:2.06; 95%CI:1.02–4.17; p = 0.044). The statistical model explains 24.9% of the susceptibility to AT-DILI, with an 8.9 times difference between patients in the highest and in the lowest quartiles of risk scores. This study sustains the complex architecture of AT-DILI. Prospective studies should evaluate the benefit of NAT2 and ABCB11 genotyping in AT personalization, particularly in patients over 55 years.

1. Introduction

In the 21st century, a quarter of the world’s population is still infected with Mycobacterium tuberculosis and thus at risk of developing active tuberculosis (TB) during their lifetime [1]. Standard antituberculosis treatment (AT) [2] is associated hepatotoxicity in 1–36% of treated patients. The spectrum of presentation is broad, ranging from asymptomatic and frequently transient elevations of liver aminotransferases in 20% of patients, to more severe phenotypes imposing interruption and modification of treatment (currently classified as drug-induced liver injury (DILI)) and even acute liver failure in 1–5% of cases [3]. Despite its high efficacy, isoniazid (INH) is the most frequently associated drug [4], though rifampicin (RIF) and pyrazinamide (PZ) have also been implicated, with the combination of drugs showing an increased risk [5,6].
AT-associated hepatotoxicity is a complex phenotype involving multiple low penetrance genetic and non-genetic variables [7]. The development of integrated risk models to assess individual susceptibility may lead to less adverse reactions, better adherence to therapy, and lower risk of emergence of resistant forms. Age, female gender, race, alcoholism, and pre-existing liver disease are some of the risk factors thought to be involved, although studies show contradictory results [3,8]. The characterization of single nucleotide polymorphisms (SNP) from the gene encoding arylamine N-acetyltransferase 2 (NAT2) enzyme, responsible for 80% of INH’s clearance, allows to classify individuals as slow (SA), intermediate (IA), or rapid acetylators (RA) [9,10]. According to most studies and meta-analyses, SA have an increased susceptibility to the development of INH-induced hepatotoxicity [7,9,10,11,12,13,14,15,16].
The cytochrome P450 2E1 (CYP2E1) is a phase I metabolizing enzyme that converts INH subproducts hydrazine and acetyl hydrazine into reactive hepatotoxic metabolites [17]. The SNP rs2031920, a 2Kb gene upstream variant, has been associated with higher enzymatic activity [18,19]. Glutathione S-Transferases (GSTs) are phase II metabolizing enzymes with an important role in the detoxification of metabolites resulting from the biotransformation of xenobiotics by phase I enzymes [17]. Copy number variations (CNV) consisting in the deletion of Glutathione S-Transferase theta 1 (GSTT1) and Glutathione S-Transferase Mu 1 (GSTM1) have been associated with AT-DILI, although with controversial results [20,21].
Loss of function variants in the gene ATP-binding cassette, subfamily B, member 11 (ABCB11), encoding the bile salt export pump (BSEP), may also be implicated. The SNP rs2287622 (p.Val444Ala) has been related to an increased risk of intrahepatic cholestasis in different clinical situations, including tuberculosis treatment [22]. Moreover, an in vitro study proved that the association of INH with RFP significantly down-regulated the expression of BSEP in liver extracts of mice [23].
IL6 is a pleiotropic cytokine with a complex role, including pro-inflammatory, anti-inflammatory, and regenerative responses [24]. In the liver, IL6 is a major inducer of the acute phase response after infection, simultaneously improving hepatic regeneration and repair, but in more chronic exposure, it can actually contribute to liver damage [25,26,27]. IL6 gene upstream SNPs rs1800796 and rs1800797 have been associated with susceptibility to liver diseases though the impact and risk genotypes varied with ethnicity [28,29]. Although NAT2 function is specifically related to INH-DILI, CYP2E1, GSTs, BSEP, and IL6 have broad-range roles and may also be implicated in the hepatotoxic effect of other antituberculosis drugs.
The aim of the present study was to identify genetic and clinical variables associated with susceptibility to AT-associated hepatotoxicity in patients with pulmonary disease.

2. Materials and Methods

2.1. Patients

A total of 233 unrelated patients diagnosed with pulmonary tuberculosis between 2004 and 2017 were studied. From 2004 to 2008, the study was retrospective and recruited patients (10%) from Coimbra Hospital and University Centre (CHUC) and from the Pneumological Diagnostic Center (CPD) of Coimbra. From 2008, the study was prospective and also included patients from the CPDs of Vila Nova de Gaia, Aveiro, Venda Nova, and Santarém.
Patients were classified as “cases” if they developed AT-associated hepatotoxicity or as “controls” if they did not. For patients that agreed to participate, a blood sample was collected for genetic analysis and a specific clinical record was created. Since 2008, for “cases” genotyping of NAT2 was immediately performed to personalize INH dosing. To limit the number of controls, for every patient becoming a case, the next patient or the next two patients were included as controls, thus assuring a blind selection. Genetic studies other than NAT2 for cases and all genes for controls were only performed after all patients had been enrolled, and only in the selected individuals. Overall, of the initially selected patients, 10% refused to participate and six patients were posteriorly excluded: three had incomplete genotype results, two had hepatotoxicity attributed to other causes, and one patient developed cancer.
Blood samples were referred by code number. All participants were informed about the study and signed an informed consent. The study was approved by the Ethics Committee of Coimbra Hospital and University Centre (Approval Code: CES011, Approval Date: 15 January 2008).
Eligibility criteria: Diagnosis of pulmonary tuberculosis, no contraindication to the standard treatment, age >16 years old, normal baseline laboratory tests including: aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), total bilirubin, and γ glutamyl transpeptidase. Human hepatitis virus (A, B and C) and HIV serology should be negative.
Exclusion criteria: pregnancy, intellectual disability, incomplete data or genotyping results; hepatitis attributed to other causes than AT drugs by causality assessment applying Roussel Uclaf Causality Assessment Method (RUCAM) [30], development of severe co-morbidity during follow-up, previous diagnosis of hepatitis of any cause, and evidence of non-compliance.
World Health Organization (WHO) guidelines for Tuberculosis [2,31,32] were followed for the management of patients, including initial standard treatment with 2 months of isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), and ethambutol (EMB) followed by a continuation phase of 4 months of INH and RIF. With the exceptions of patients assuring a high compliance, directly observed treatment was adopted. Patients had a first follow up appointment 15 days after initiating therapy and monthly thereafter. In each appointment, a clinical record was fulfilled, and blood levels of liver enzymes were evaluated.
AT-associated hepatotoxicity was classified in three grades: (1) mild hepatitis, considering an elevation of serum concentration of ALT exceeding two-fold the upper limit of normal (ULN) [33]; (2) severe/DILI, considering an ALT level of more than five-fold above the ULN, or an ALP level of more than two-fold above the ULN; or ALT level of more than three-fold above the ULN, with the simultaneous elevation of total bilirubin levels to more than two-fold above the ULN [34]; (3) liver failure, considering a grade 2 criteria plus death or liver transplantation. The RUCAM score [30] was used for the adjudication of patients as having AT-associated hepatotoxicity (defined as cases). The liver injury pattern was assessed by the R value [33]. The earliest identified pattern of liver injury was recorded.
Co-morbidities at the time of diagnosis excluded transient acute diseases. Active smoking habits are described as: “not present” (less than 1 cigarette a day in the last month) or “present” [35]. Alcoholic intake refers to average habits in the last three months preceding the diagnosis of tuberculosis. Patients were classified as having “no alcohol” consumption if drinking no more than 20 g of pure alcohol a day for men or half that amount for women [36]. During treatment, patients were instructed to reduce their alcohol drinking habits by at least 50%. Concomitant chronic unspecified medication and concomitant chronic medication, including hepatotoxic drugs, were registered. Drugs were classified as hepatotoxic according to EASL guidelines [37]. These drugs were excluded as the cause of liver injury (RUCAM score < 3).

2.2. Genotyping

Extraction of DNA was performed from peripheral blood samples using the “NZY Blood gDNA Isolation” kit.
For the NAT2 gene, the gold standard 11 SNPs were evaluated by Sanger sequencing [38,39]. The SNPs rs2287622 from ABCB11 gene, rs2031920 form CYP2E1 gene, and rs1800796 and rs1800797 from IL6 gene were also characterized by Sanger sequencing. To identify GSTT1 and GSTM1 deletions, a multiplex PCR reaction was performed using a sequence of the β-globin gene as internal control [40]. The evidence of amplification, described as positive, identifies the presence of at least one allele, not distinguishing homozygotes from heterozygotes, while the absence of amplification, described as null, identifies homozygosity for the deletions.
Primers and annealing temperatures are described in Supplementary Table S1.

2.3. Statistical Analysis

Statistical analysis was performed with the software SPSS Statistics, Version 25 (SPSS, Inc., IBM Company, Armonk, NY, USA). Hardy–Weinberg equilibrium was tested applying a chi-squared test. A p-value below 0.05 was considered statistically significant. Univariate analyses were performed for each risk factor. For age, the Mann–Whitney test was used. To transform age in a binary variable, ROC curve analysis was performed (cut-off value for hepatotoxicity: ≥55 years, sensitivity 56.2%, specificity 73.8%). Given its normal distribution, weight was assessed with an ANOVA test. The remaining variables were analyzed using Fisher’s exact test. Multicollinearity was determined between independent variables using Fisher’s exact test. The identification of predictors of hepatotoxicity was performed through logistic regression. The adequacy of the logistic regression models was evaluated by the Hosmer–Lemeshow test and Nagelkerke R2 was computed as a measure of the variance explained by the model. Quartile analysis of cases and controls based on the range of probability values obtained from the multivariate regression model was also performed.

3. Results

3.1. Patients Characteristics and Univariate Analysis

Of the 233 patients included in the study, 103 (44.2%) developed AT-associated hepatotoxicity (cases): 37 patients were classified as grade 1, 61 patients as grade 2, and five patients as grade 3. Patients with grade 1 (36% of hepatotoxicity group) were described as having mild hepatotoxicity and grade 2 and three patients (64% of hepatotoxicity group) were joined and described as severe hepatotoxicity/DILI. RUCAM scores for causality assessment of AT drugs were as follows: all patients in grade 2 had a score ≥ 6 (probable), with 14 patients (23%) reaching a score ≥ 8; all patients with grade 1 had a score ≥ 4 (possible), 9 (24%) reaching a score ≥ 6.
The hepatocellular pattern was the most common among patients with severe hepatotoxicity/DILI (Table 1). For patients with mild forms, there was a greater heterogeneity, with the mixed pattern being the more frequent.
Clinical characteristics evaluated for cases and controls are described in Table 2.
A univariate analysis was also performed for patients with mild hepatitis and AT-DILI and described in Supplementary Tables S2 and S3.
For alcohol consumption, other medication, and other hepatotoxic drugs, different subgroups were analyzed but only results with statistical significance are described in Table 2. A total of 116 patients (49.8%) had no concomitant medication. The most common hepatotoxic drugs registered were statins and oral contraceptives.
NAT2 genotypes were grouped according to acetylator phenotype (Table 3): 126 patients (54.1%) were SA, 95 (40.8%) were IA, and 12 patients (5.1%) were RA. The genotypic and phenotypic frequencies found in this study were similar to those previously described for the Portuguese population [40,41]. For other analyzed genes, the frequencies of genotypes in the global sample and in patients with and without hepatotoxicity are also described in Table 4. Hardy–Weinberg equilibrium was confirmed for all SNPs (p > 0.05).
For each gene, all other combinations of SNP variants were tested but no statistically significant association was revealed.
Analysis of ethnicity interference showed that the genotype AG of IL6 rs1800797 SNP and presence of at least one allele of GSTT1 are more frequent among Caucasians (p = 0.002 and p = 0.036, respectively, results not shown). However, in a logistic regression analysis involving the interaction between each of these genetic variants and race, none was found to interfere (Race and IL6: OR:1.28, p = 0.692; Race and GSTT1 OR:0.78, p = 0.668). Thus, the inclusion of non-Caucasians does not introduce a bias in the results.

3.2. Multivariate Analysis

A logistic multivariate analysis for all cases (Table 4) was performed, including variables with a p < 0.05 in univariate analysis (Table 2 and Table 3). The same analysis was also performed for DILI cases compared to controls (Table 4).
Finally, a refined analysis (Table 5) was performed including only variables with statistical significance in the previous logistic multivariate analysis.
Each variable increases the susceptibility to AT-associated hepatotoxicity by 2–3 times. For all cases, the presented model explains 24.2% of the susceptibility to AT-associated hepatotoxicity. Considering the refined logistic regression model, the achieved Nagelkerke R Square (0.242), and the four predictors identified at a 5% significance level with a sample of 233 subjects, the attained statistical power for the analysis is 87.2%.
For patients with severe hepatotoxicity/DILI (Table 5), the same variables were identified: age ≥ 55 years (OR:3.67; 95% CI:1.82–7.41; p < 0.001), concomitant medication with other hepatotoxic drugs (OR:2.54; 95% CI:1.23–5.26; p = 0.012), SA status (OR:2.46; 95% CI:1.25–4.84; p = 0.009) and presence of CC genotype for ABCB11 SNP (OR:2.06; 95% CI:1.02–4.17; p = 0.044). The statistical model also explains about 25% of the susceptibility to this more severe phenotype.
A logistic multivariate analysis was also performed for patients with mild hepatotoxicity (Supplementary Table S4). The presence of other hepatotoxic drugs and SA status were the only risk factors identified (p = 0.025 and p = 0.038, respectively), with age ≥ 55 years showing an almost statistically significant effect (p = 0.075).

3.3. Quartile Analysis of Risk Scores

Quartile analysis of risk scores showed a 3.5 times difference on susceptibility to AT-associated hepatotoxicity (both forms) between patients in the highest and the lowest quartile (Figure 1).
For patients with severe forms, AT-DILI, there is an 8.9 times difference between patients in the highest and in the lowest quartiles of risk scores (Figure 1).

4. Discussion

Despite extensive research efforts, the mechanisms beyond AT-associated hepatotoxicity are still not fully understood. In this study, we searched for genetic and non-genetic variants associated with this complex phenotype. Regarding non-genetic risk factors, results show that age ≥ 55 years was associated with a triple increase in the risk of AT-associated hepatotoxicity. Other studies confirm that older age is a risk factor [42] though the mechanisms involved are unclear [43]. Aging is known to associate with the deterioration of renal function and with a decrease of organ volume, blood flow, and cytochrome-mediated metabolism in the liver, all of which might affect drug pharmacokinetics [44]. Despite these changes, renal and liver function are essentially preserved in healthy older humans. It is also important to consider factors, such immunoinflammatory modifications [45] and polypharmacy, which are more common among the elderly [44]. In our patients, chronic diseases and concomitant medication were not risk factors, but the use of other hepatotoxic drugs increased the risk increased the risk by 2.3 times. Given this association, in clinical practice, the suspension of other hepatotoxic medication before starting AT treatment should be evaluated on a case-by-case basis. Contrarily to other authors, gender was not identified as a risk factor [44]. Pre-existing liver disease was not evaluated in this study as patients without normal baseline laboratory tests were excluded. In multivariate analysis, alcohol intake and smoking habits were not identified as risk factors. These are subjective variables as they depend on the reliability of patients’ responses. Though alcohol consumption is frequently referred to as a risk factor and is included in RUCAM score [30], a recent study failed to establish an association with DILI attributed to isoniazid [8]. The fact that no patient with baseline evidence of liver damage was included and that alcohol intake was restricted during treatment may also explain our results. The interference of all these clinical factors in DILI remains a subject of discussion [37].
Our results confirm the role of NAT2 SA genotypes in AT-DILI and show that the impact is extendable to mild forms of AT-associated hepatotoxicity. There is evidence that genotyping-based INH dosing may significantly decrease the incidence of AT-associated hepatotoxicity and early treatment failure, allowing a 31% reduction in absolute risk of unfavorable events [16]. It was also suggested that patients with SA status would likely benefit from closer surveillance [14,15]. Within the SA group, there is heterogeneity in phenotype due to variations in enzyme activity conferred by different alleles, with some alleles associating with “ultra-slow” acetylation, supporting the advantage of NAT2 genotyping [11]. NAT2 genotyping was performed with the 11-SNP panel, considered to be the gold standard method, but more practical approaches of seven- and four-SNP panels show a classification accuracy of 98.4% in USA individuals, including 79.3% Caucasian, 10.55% African-American, and 7.03% Hispanic [39]. We verified that the seven-SNP panel could classify all our patients. Genotype ambiguities were within the same acetylator activity group and did not compromise phenotype classification.
ABCB11 gene SNP rs2287622 also shows impact in AT-DILI but not in mild AT-associated hepatotoxicity. This gene has been implicated in hereditary and acquired forms of liver cholestasis but population studies on the association with AT-DILI are scarce [22]. In our patients, even though the most common pattern was the hepatocellular, the presence of CC genotype doubled the chance of AT-DILI. For the functional genetic polymorphisms studied in CYP2E1, GSTM1, GSTT1, and IL6 genes, no association was found. For these genes, conflicting results have been published. Considering recent meta-analyses, two support the roles of CYP2E1, GSTM1, and GSTT1 in AT-DILI [19,20], one does not implicate GSTT1 [21], and in another CYP2E1 could only be implicated in East Asian populations [18]. In a trans-ethnic meta-analysis from Nicoletti et al., CYP2E1 was not associated with AT-DILI [46]. For IL6 association with AT-DILI, the few results reported are not conclusive [29].
Variation in population genetic background and in environmental exposures, namely to herbals, regional medication policies, and prescription habits, may account for some discrepancies in studies evaluating susceptibility to AT-DILI [44]. Differences in study design, including genotyping methodologies and criteria for AT-DILI, may also contribute. AT-DILI is a complex phenotype better described by multifactorial models, including multiple genetic and non-genetic factors, but most studies do not analyze the impact of variables, such as pre-existing liver diseases or concomitant medication, though not excluding them, as is the case of meta-analyses and studies well designed to explore the association only with genetic variants [46]. The impact of the identified risk factors in the predictability of the pathologic condition is also frequently not referred to.
In this study, when mild and more severe liver injury were separately evaluated, different risk profiles emerged. For mild presentations, characterized by a more heterogeneous hepatitis pattern, including a higher frequency of cholestatic forms, only age ≥ 55 years, and NAT2 SA status, could be implicated. This suggests that mild, transient forms and more severe phenotypes are related and share some risk factors.
For AT-DILI, quartile analysis of risk scores shows an 8.9 times difference between patients in the highest and lowest quartiles. Yet, exploring seven genetic and nine clinical variables, only 25% of the risk of AT-DILI could be predicted, supporting the multifactorial and polygenic architecture of the phenotype. In contrast, INH blood levels behave as a quasi-monogenic trait, mostly depending on NAT2 genotype. A role for immune mediated response in AT-DILI events has been suggested [4] and is supported by a recent association described with HLA-B*52:01 [46].
This study has several limitations. First, as all hepatotoxicity events occurred when patients were submitted to treatment with four drugs, although INH is thought to be the most frequently involved, no specific antituberculosis drug can be implicated. In order to assess INH-DILI susceptibility factors, studies, including genotyping and monitoring of plasma INH toxic metabolites, should be performed on individuals treated with isoniazid alone, as in latent TB [47]. Second, to increase statistical power and to enable variant-interaction analysis, the sample size should be extended. The sample dimension also precluded a genome wide association approach. Information on alcoholic and smoking habits and therapeutic compliance was self-reported and may not be accurate. The possible role of epigenetic factors was not evaluated [48,49].

5. Conclusions

In this study, exploring multiple genetic and clinical variables, four risk factors for AT-DILI were identified: NAT2 SNPs determining SA phenotype, CC genotype for ABCB11 SNP rs2287622, age ≥ 55 years, and concomitant use of hepatotoxic drugs. The overall effect of the studied variables is modest, which suggests a complex interaction of many still unknown genetic, epigenetic, and environmental factors. For clinical translation, prospective studies should evaluate the role of NAT2 and ABCB11 genotyping in the AT personalization, particularly in patients over 55 years. Due consideration should also be given to the interruption of other hepatotoxic medication during AT treatment.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jpm12050790/s1, Table S1: Primers used in amplification and sequencing reactions; Table S2: Characterization of non-genetic risk factors in mild hepatitis and DILI; Table S3: Characterization of genetic risk factors in mild hepatitis and DILI; Table S4: Results of logistic multivariate analysis for mild hepatitis.

Author Contributions

Conceptualization, B.O. and H.C.S.; Data curation, M.J.C. and C.A.; Formal analysis, M.J.C. and B.O.; Funding acquisition, F.R. and H.C.S.; Investigation, M.J.C., C.A., L.A.M., A.C., F.M., J.M.C., M.V., R.D., J.M., C.R., C.R.C. and H.C.S.; Methodology, C.A., L.A.M., J.M., C.R. and H.C.S.; Project administration, H.C.S.; Resou-rces, C.A., A.C., F.M., J.M.C., M.V., R.D., C.R.C. and F.R.; Software, B.O.; Supervision, C.R.C., F.R. and H.C.S.; Validation, L.A.M., J.M. and C.R.; Visualization, M.J.C., C.A., A.C., F.M., J.M.C. and R.D.; Writing—original draft, M.J.C., C.A., B.O. and H.C.S.; Writing—review & editing, L.A.M., A.C., F.M., J.M.C., M.V., R.D., J.M., C.R., C.R.C. and F.R. All authors have read and agreed to the published version of the manuscript.

Funding

This article was funded by project “GenomePT—National Laboratory for Genome Sequencing and Analysis” with the reference POCI-01-0145-FEDER-022184, supported by Lisboa Portugal Regional Operational Programme (Lisboa2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF) and project “Central Region Training Project for Personalized/Precision Medicine, with a genomic basis”, financed by the program CENTRO2020 and with the reference CENTRO-08-5864-FSE-000039 (PEP IN1194)This research was also supported by the Institut Mérieux and the GlaxoSmithKline Foundation for health sciences.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by Ethics Committee of Coimbra Hospital and Universitary Centre (Approval Code: CES011, Approval Date: 15 January 2008).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. WHO. Global Tubeculosis Report 2020; WHO: Geneva, Switzerland, 2020; ISBN 9789240013131. [Google Scholar]
  2. WHO. Guidelines for Treatment of Drug-Susceptible Tuberculosis and Patient Care; WHO: Geneva, Switzerland, 2017; Volume 62, ISBN 9789241550000. [Google Scholar]
  3. Saukkonen, J.J.; Cohn, D.L.; Jasmer, R.M.; Schenker, S.; Jereb, J.A.; Nolan, C.M.; Peloquin, C.A.; Gordin, F.M.; Nunes, D.; Strader, D.B.; et al. An official ATS statement: Hepatotoxicity of antituberculosis therapy. Am. J. Respir. Crit. Care Med. 2006, 174, 935–952. [Google Scholar] [CrossRef] [PubMed]
  4. Metushi, I.; Uetrecht, J.; Phillips, E. Mechanism of isoniazid-induced hepatotoxicity: Then and now. Br. J. Clin. Pharmacol. 2016, 81, 1030–1036. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  5. Steele, M.A.; Burk, R.F.; DesPrez, R.M. Toxic Hepatitis with Isoniazid and Rifampin. Chest 1991, 99, 465–471. [Google Scholar] [CrossRef] [PubMed]
  6. Durand, F.; Bernuau, J.; Pessayre, D.; Samuel, D.; Belaiche, J.; Degott, C.; Bismuth, H.; Belghiti, J.; Erlinger, S.; Rueff, B.; et al. Deleterious influence of pyrazinamide on the outcome of patients with fulminant or subfulminant liver failure during antituberculous treatment including isoniazid. Hepatology 1995, 21, 929–932. [Google Scholar] [CrossRef]
  7. Daly, A.K. Are Polymorphisms in Genes Relevant to Drug Disposition Predictors of Susceptibility to Drug-Induced Liver Injury? Pharm. Res. 2017, 34, 1564–1569. [Google Scholar] [CrossRef] [Green Version]
  8. Dakhoul, L.; Ghabril, M.; Gu, J.; Navarro, V.; Chalasani, N.; Serrano, J. Heavy Consumption of Alcohol is Not Associated With Worse Outcomes in Patients With Idiosyncratic Drug-induced Liver Injury Compared to Non-Drinkers. Clin. Gastroenterol. Hepatol. 2018, 16, 722–729.e2. [Google Scholar] [CrossRef] [Green Version]
  9. Zhu, Y.; Hein, D.W. Functional effects of single nucleotide polymorphisms in the coding region of human N-acetyltransferase 1. Pharm. J. 2008, 8, 339–348. [Google Scholar] [CrossRef]
  10. Selinski, S.; Blaszkewicz, M.; Ickstadt, K.; Hengstler, J.G.; Golka, K. Refinement of the prediction of N-acetyltransferase 2 (NAT2) phenotypes with respect to enzyme activity and urinary bladder cancer risk. Arch. Toxicol. 2013, 87, 2129–2139. [Google Scholar] [CrossRef]
  11. Suvichapanich, S.; Fukunaga, K.; Zahroh, H.; Mushiroda, T.; Mahasirimongkol, S.; Toyo-Oka, L.; Chaikledkaew, U.; Jittikoon, J.; Yuliwulandari, R.; Yanai, H.; et al. NAT2 ultra-slow acetylator and risk of anti-tuberculosis drug-induced liver injury: A genotype-based meta-analysis. Pharm. Genom. 2018, 28, 167–176. [Google Scholar] [CrossRef]
  12. Zhang, M.; Wang, S.; Wilffert, B.; Tong, R.; van Soolingen, D.; van den Hof, S.; Alffenaar, J.W. The association between the NAT2 genetic polymorphisms and risk of DILI during anti-TB treatment: A systematic review and meta-analysis. Br. J. Clin. Pharmacol. 2018, 84, 2747–2760. [Google Scholar] [CrossRef] [Green Version]
  13. Richardson, M.; Kirkham, J.; Dwan, K.; Sloan, D.J.; Davies, G.; Jorgensen, A.L. NAT2 variants and toxicity related to anti-tuberculosis agents: A systematic review and meta-analysis. Int. J. Tuberc. Lung Dis. 2019, 23, 293–305. [Google Scholar] [CrossRef] [PubMed]
  14. Kinzig-schippers, M.; Tomalik-scharte, D.; Jetter, A.; Scheidel, B.; Jakob, V.; Rodamer, M.; Cascorbi, I.; Doroshyenko, O.; So, F.; Fuhr, U. Should We Use N-Acetyltransferase Type 2 Genotyping To Personalize Isoniazid Doses? Society 2005, 49, 1733–1738. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Donald, P.R.; Parkin, D.P.; Seifart, H.I.; Schaaf, H.S.; van Helden, P.D.; Werely, C.J.; Sirgel, F.A.; Venter, A.; Maritz, J.S. The influence of dose and N-acetyltransferase-2 (NAT2) genotype and phenotype on the pharmacokinetics and pharmacodynamics of isoniazid. Eur. J. Clin. Pharmacol. 2007, 63, 633–639. [Google Scholar] [CrossRef] [PubMed]
  16. Azuma, J.; Ohno, M.; Kubota, R.; Yokota, S.; Nagai, T.; Tsuyuguchi, K.; Okuda, Y.; Takashima, T.; Kamimura, S.; Fujio, Y.; et al. NAT2 genotype guided regimen reduces isoniazid-induced liver injury and early treatment failure in the 6-month four-drug standard treatment of tuberculosis: A randomized controlled trial for pharmacogenetics-based therapy. Eur. J. Clin. Pharmacol. 2013, 69, 1091–1101. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  17. Wang, P.; Pradhan, K.; Zhong, X.-B.; Ma, X. Isoniazid metabolism and hepatotoxicity. Acta Pharm. Sin. B 2016, 6, 384–392. [Google Scholar] [CrossRef] [Green Version]
  18. Liu, X.; Ren, S.; Zhang, J.; Xu, D.; Jiang, F.; Jiang, P.; Feng, J.; Deng, F. The association between cytochrome P450 polymorphisms and anti-tuberculosis drug-induced liver injury: A systematic review and meta-analysis. Ann. Palliat. Med. 2021, 10, 6518–6534. [Google Scholar] [CrossRef]
  19. Richardson, M.; Kirkham, J.; Dwan, K.; Sloan, D.J.; Davies, G.; Jorgensen, A.L. CYP genetic variants and toxicity related to anti-tubercular agents: A systematic review and meta-analysis. Syst. Rev. 2018, 7, 204. [Google Scholar] [CrossRef]
  20. Zhang, M.; Wu, S.; He, J. Are genetic variations in glutathione S-transferases involved in anti-tuberculosis drug-induced liver injury? A meta-analysis. J. Clin. Pharm. Ther. 2019, 44, 844–857. [Google Scholar] [CrossRef]
  21. Yang, S.; Hwang, S.J.; Park, J.Y.; Chung, E.K.; Lee, J.I. Association of genetic polymorphisms of CYP2E1, NAT2, GST and SLCO1B1 with the risk of anti-tuberculosis drug-induced liver injury: A systematic review and meta-analysis. BMJ Open 2019, 9, e027940. [Google Scholar] [CrossRef] [Green Version]
  22. Chen, R.; Wang, J.; Tang, S.; Zhang, Y.; Lv, X.; Wu, S.; Yang, Z.; Xia, Y.; Chen, D.; Zhan, S. Role of polymorphic bile salt export pump (BSEP, ABCB11) transporters in anti-tuberculosis drug-induced liver injury in a Chinese cohort. Sci. Rep. 2016, 6, 27750. [Google Scholar] [CrossRef] [Green Version]
  23. Guo, Y.X.; Xu, X.F.; Zhang, Q.Z.; Li, C.; Deng, Y.; Jiang, P.; He, L.Y.; Peng, W.X. The inhibition of hepatic bile acids transporters Ntcp and Bsep is involved in the pathogenesis of isoniazid/rifampicin-induced hepatotoxicity. Toxicol. Mech. Methods 2015, 25, 382–387. [Google Scholar] [CrossRef] [PubMed]
  24. Reeh, H.; Rudolph, N.; Billing, U.; Christen, H.; Streif, S.; Bullinger, E.; Schliemann-Bullinger, M.; Findeisen, R.; Schaper, F.; Huber, H.J.; et al. Response to IL-6 trans- A nd IL-6 classic signalling is determined by the ratio of the IL-6 receptor α to gp130 expression: Fusing experimental insights and dynamic modelling. Cell Commun. Signal. 2019, 17, 46. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  25. Jin, X.; Zimmers, T.A.; Perez, E.A.; Pierce, R.H.; Zhang, Z.; Koniaris, L.G. Paradoxical effects of short- and long-term interleukin-6 exposure on liver injury and repair. Hepatology 2006, 43, 474–484. [Google Scholar] [CrossRef] [PubMed]
  26. Wang, Z.; Wu, S.; Liao, J.; Zhong, L.; Xing, T.; Fan, J.; Peng, Z. Interleukin-6 and rs1800796 locus single nucleotide polymorphisms in response to hypoxia/reoxygenation in hepatocytes. Int. J. Mol. Med. 2016, 38, 192–200. [Google Scholar] [CrossRef] [Green Version]
  27. Singh, M.; Mastana, S.; Singh, S.; Juneja, P.K.; Kaur, T.; Singh, P. Promoter polymorphisms in IL-6 gene influence pro-inflammatory cytokines for the risk of osteoarthritis. Cytokine 2020, 127, 154985. [Google Scholar] [CrossRef]
  28. Wang, X.; Yan, Z.; Ye, Q. Interleukin-6 gene polymorphisms and susceptibility to liver diseases: A meta-analysis. Medicine 2019, 98, e18408. [Google Scholar] [CrossRef]
  29. Li, Y.; Tang, H.; Qi, H.; Shen, C.; Sun, L.; Li, J.; Xu, F.; Jiao, W.; Yang, X.; Shen, A. rs1800796 of the IL6 gene is associated with increased risk for anti-tuberculosis drug-induced hepatotoxicity in Chinese Han children. Tuberculosis 2018, 111, 71–77. [Google Scholar] [CrossRef]
  30. Danan, G.; Teschke, R. RUCAM in drug and herb induced liver injury: The update. Int. J. Mol. Sci. 2016, 17, 10014. [Google Scholar] [CrossRef]
  31. WHO. Treatment of Tuberculosis: Guidelines for National Programmes, 3rd ed.; WHO: Geneva, Switzerland, 2003; Volume 93, p. 72. [Google Scholar] [CrossRef]
  32. World Health Organization. Treatment of Tuberculosis: Guidelines, 4th ed.; WHO: Geneva, Switzerland, 2010. [Google Scholar]
  33. Bénichou, C. Criteria of drug-induced liver disorders: Report of an International Consensus Meeting. J. Hepatol. 1990, 11, 272–276. [Google Scholar] [CrossRef]
  34. Aithal, G.P.; Watkins, P.B.; Andrade, R.J.; Larrey, D.; Molokhia, M.; Takikawa, H.; Hunt, C.M.; Wilke, R.A.; Avigan, M.; Kaplowitz, N.; et al. Case Definition and Phenotype Standardization in Drug-Induced Liver Injury. Clin. Pharmacol. Ther. 2011, 89, 806–815. [Google Scholar] [CrossRef]
  35. Schane, R.E.; Ling, P.M.; Glantz, S.A. Glantz Health Effects of Light and Intermittent Smoking: A Review. Circulation 2010, 121, 1518–1522. [Google Scholar] [CrossRef]
  36. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 2015–2020 Dietary Guidelines for Americans, 8th ed. Available online: https://health.gov/our-work/food-nutrition/previous-dietary-guidelines/2015 (accessed on 25 April 2022).
  37. Andrade, R.J.; Aithal, G.P.; Björnsson, E.S.; Kaplowitz, N.; Kullak-Ublick, G.A.; Larrey, D.; Karlsen, T.H. EASL Clinical Practice Guidelines: Drug-induced liver injury. J. Hepatol. 2019, 70, 1222–1261. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  38. Deitz, A.C.; Rothman, N.; Rebbeck, T.R.; Hayes, R.B.; Chow, W.H.; Zheng, W.; Hein, D.W.; García-Closas, M. Impact of misclassification in genotype-exposure interaction studies: Example of N-acetyltransferase 2 (NAT2), smoking, and bladder cancer. Cancer Epidemiol. Biomark. Prev. 2004, 13, 1543–1546. [Google Scholar]
  39. Hein, D.W.; Doll, M.A. Accuracy of various human NAT2 SNP genotyping panels to infer rapid, intermediate and slow acetylator phenotypes. Pharmacogenomics 2012, 13, 31–41. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  40. Lemos, M.C.; Coutinho, E.; Gomes, L.; Carrilho, F.; Rodrigues, F.; Regateiro, F.J.; Carvalheiro, M. Combined GSTM1 and GSTT1 null genotypes are associated with a lower risk of papillary thyroid cancer. J. Endocrinol. Investig. 2008, 31, 542–545. [Google Scholar] [CrossRef]
  41. Lemos, M.C.; Regateiro, F.J. N-acetyltransferase genotypes in the Portuguese population. Pharmacogenetics 1998, 8, 561–564. [Google Scholar] [CrossRef]
  42. Fountain, F.F.; Tolley, E.; Chrisman, C.R.; Self, T.H. Isoniazid hepatotoxicity associated with treatment of latent tuberculosis infection: A 7-year evaluation from a public health tuberculosis clinic. Chest 2005, 128, 116–123. [Google Scholar] [CrossRef] [Green Version]
  43. Onji, M.; Fujioka, S.; Takeuchi, Y.; Takaki, T.; Osawa, T.; Yamamoto, K.; Itoshima, T. Clinical characteristics of drug-induced liver injury in the elderly. Hepatol. Res. 2009, 39, 546–552. [Google Scholar] [CrossRef]
  44. Ortega-Alonso, A.; Stephens, C.; Lucena, M.I.; Andrade, R.J. Case characterization, clinical features and risk factors in drug-induced liver injury. Int. J. Mol. Sci. 2016, 17, 50714. [Google Scholar] [CrossRef] [Green Version]
  45. John, P.; Kale, P.P. Prominence of Oxidative Stress in the Management of Anti-tuberculosis Drugs Related Hepatotoxicity. Drug Metab. Lett. 2019, 13, 95–101. [Google Scholar] [CrossRef]
  46. Nicoletti, P.; Devarbhavi, H.; Goel, A.; Venkatesan, R.; Eapen, C.E.; Grove, J.I.; Zafer, S.; Bjornsson, E.; Lucena, M.I.; Andrade, R.J.; et al. Genetic Risk Factors in Drug-Induced Liver Injury Due to Isoniazid-Containing Antituberculosis Drug Regimens. Clin. Pharmacol. Ther. 2021, 109, 1125–1135. [Google Scholar] [CrossRef] [PubMed]
  47. Ng, C.S.; Hasnat, A.; Al Maruf, A.; Ahmed, M.U.; Pirmohamed, M.; Day, C.P.; Aithal, G.P.; Daly, A.K. N-acetyltransferase 2 (NAT2) genotype as a risk factor for development of drug-induced liver injury relating to antituberculosis drug treatment in a mixed-ethnicity patient group. Eur. J. Clin. Pharmacol. 2014, 70, 1079–1086. [Google Scholar] [CrossRef] [PubMed]
  48. Urban, T.J.; Goldstein, D.B.; Watkins, P.B. Genetic basis of susceptibility to drug-induced liver injury: What have we learned and where do we go from here? Pharmacogenomics 2012, 13, 735–738. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  49. Murata, K.; Hamada, M.; Sugimoto, K.; Nakano, T. A novel mechanism for drug-induced liver failure: Inhibition of histone acetylation by hydralazine derivatives. J. Hepatol. 2007, 46, 322–329. [Google Scholar] [CrossRef] [PubMed]
Jpm 12 00790 g001 550
Figure 1. Proportion of cases and controls in each quartile using the best predictive model for AT-associated hepatotoxicity and AT-DILI, respectively. The dark bars represent the proportion of cases which fall into each quartile and the light bars represent the proportion of controls. The number of cases and controls who fall within each quartile are indicted in each block.
Figure 1. Proportion of cases and controls in each quartile using the best predictive model for AT-associated hepatotoxicity and AT-DILI, respectively. The dark bars represent the proportion of cases which fall into each quartile and the light bars represent the proportion of controls. The number of cases and controls who fall within each quartile are indicted in each block.
Jpm 12 00790 g001
Table
Table 1. Distribution of hepatitis pattern according to the grade of hepatotoxicity.
Table 1. Distribution of hepatitis pattern according to the grade of hepatotoxicity.
Hepatitis PatternGrade of Hepatotoxicity
Mild n (%)DILI n (%)Total n (%)
Hepatocellular10 (33.3%)60 (90.9%)70 (72.9%)
Cholestatic7 (23.3%)4 (6.1%)11 (11.5%)
Mixed13 (43.4%)2 (3.0%)15 (15.6%)
Total30 (100%)66 (100%)96 (100%)
n—number of patients.
Table
Table 2. Characterization of non-genetic risk factors.
Table 2. Characterization of non-genetic risk factors.
Clinical VariablesControls
n (%)		Cases
n (%)		Global
n (%)		pOR (95% IC)
Age (17–87 years)
<55 years96 (73.8%)46 (44.7%)142 (60.9%)<0.001
≥55 years34 (26.2%)57 (55.3%)91 (39.1%)3.49 (2.02–6.07)
Mean (sd)45.1 (16.1)55.2 (19.0)49.6 (18.1)
Gender
Female39 (30.0%)36 (35.0%)7 (32.2%)0.4221.25 (0.72–2.18)
Male91 (70.0%)67 (65.0%)158 (67.8%)
Race
Caucasian120 (92.3%)95 (92.42%)215 (92.3%)0.983
Non-Caucasian10 (7.7%)8 (7.8%)18 (7.7%)1.01 (0.38–2.66)
Weight (33–103 kg)
Mean (sd)61.5 (11.3)62.2 (11.8)61.8 (11.5)0.419
Chronic Diseases
No63 (48.5%)36 (35.0%)99 (42.5%)0.039
Yes67 (51.5%)67 (65.0%)134 (57.5%)1.75 (1.03–2.98)
Smoking Habits
No79 (60.8%)82 (79.6%)161 (69.1%)0.002
Yes51 (39.2%)21 (20.4%)72 (30.9%)0.39 (0.22–0.72)
Alcohol Consumption
No85 (65.4%)82 (79.6%)167 (71.7%)0.018
Yes45 (34.6%)21 (20.4%)66 (28.3%)0.48 (0.26–0.88)
Other Medication
<3 drugs114 (87.7%)76 (73.8%)190 (81.5%)0.008
≥3 drugs16 (12.3%)29 (26.2%)43 (18.5%)2.53 (1.28–5.01)
Other Hepatotoxic Drugs
No106 (81.5%)57 (55.3%)163 (70.0%)<0.001
Yes24 (18.5%)46 (44.7%)70 (30.0%)3.56 (1.98–6.42)
n—number of patients; p—p value; OR—odds ratio; 95% IC—confidence interval; sd—standard deviation. Statistically significant results are highlighted in bold.
Table
Table 3. Characterization of genetic risk factors.
Table 3. Characterization of genetic risk factors.
GenotypesControls
n (%)		Cases
n (%)		Global
n (%)		pOR (95% IC)
NAT2/Acetylation status
SA58 (44.6%)68 (66.0%)126 (54.1%)
IA65 (50.0%)30 (29.1%)95 (40.8%)0.0010.39 (0.23–0.69)
RA7 (5.4%)5 (4.9%)12 (5.1%)0.4180.61 (0.18–2.02)
RA + IA vs. SA72 (55.4%)35 (34.0%)107 (45.9%)0.0010.42 (0.24–0.71)
ABCB11
TT + TC21 + 76 (74.6%)13 + 48 (59.2%)158 (67.8%)0.013
CC33 (25.4%)42 (40.8%)75 (32.2%)2.024 (1.16–3.53)
GSTM1
Null69 (53.1%)49 (47.6%)118 (50.6%)0.404
Positive61 (46.9%)54 (52.4%)115 (49.4%)1.247 (0.74–2.09)
GSTT1
Null17 (13.1%)9 (8.7%)26 (11.2%)0.299
Positive113 (85.9%)94 (91.3%)207 (88.8%)1.571 (0.67–3.69)
IL6 rs1800797
AA + AG16 + 54 (53.8%)8 + 54 (60.2%)132 (56.7%)0.332
GG60 (46.2%)41 (39.8%)101 (43.3%)0.772 (0.46–1.30)
IL6 rs1800796
CC + CG2 + 16 (13.8%)1 + 17 (17.5%)36 (15.5%)0.447
GG112 (86.2%)85 (82.5%)197 (84.5%)0.759 (0.37–1.55)
CYP2E1
CC114 (87.7%)93 (90.3%)207 (88.8%)0.532
TT + CT1 + 15 (12.3%)0 + 10 (9.7%)26 (11.2%)1.305 (0.57–3.01)
n—number of patients; p—p value; OR—odds ratio; 95% IC—confidence interval; sd—standard deviation. Statistically significant results are highlighted in bold.
Table
Table 4. Results of logistic multivariate analysis for all cases and DILI.
Table 4. Results of logistic multivariate analysis for all cases and DILI.
All Cases vs. ControlsDILI Cases vs. Controls
VariablesORIC 95%pORIC 95%p
Age ≥ 55 years2.781.45–5.310.0023.651.74–7.620.001
Chronic Diseases0.940.49–1.810.8540.840.42–1.770.654
Alcohol intake0.720.34–1.510.3820.660.27–1.580.350
Smoking habits0.610.29–1.250.1750.660.29–1.510.325
Other medication ≥3 drugs1.190.49–2.870.6911.000.37–2.690.993
Other hepatotoxic drugs2.351.15–4.810.0192.311.01–5.280.048
Slow Acetylator2.521.39–4.570.0022.551.28–5.070.008
ABCB11–CC1.911.03–3.560.0441.960.96–4.000.064
OR—odds ratio; CI—confidence interval; p—p value. Nagelkerke R Square: 26.4%and 26.9%, respectively. Statistically significant results are highlighted in bold.
Table
Table 5. Results of refined logistic multivariate analysis for all cases and DILI.
Table 5. Results of refined logistic multivariate analysis for all cases and DILI.
All Cases vs. ControlsDILI Cases vs. Controls
VariablesORIC 95%pORIC 95%p
Age ≥ 55 years2.971.62–5.43<0.0013.671.82–7.41<0.001
Other hepatotoxic drugs2.741.44–5.210.0022.541.23–5.260.012
Slow Acetylator2.401.34–4.310.0032.461.25–4.840.009
ABCB11-CC1.981.07–3.670.0302.061.02–4.170.044
OR—odds ratio; CI—confidence interval; p—p value. Nagelkerke R Square: 24.2% and 24.9%, respectively. Statistically significant results are highlighted in bold.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Cavaco, M.J.; Alcobia, C.; Oliveiros, B.; Mesquita, L.A.; Carvalho, A.; Matos, F.; Carvalho, J.M.; Villar, M.; Duarte, R.; Mendes, J.; et al. Clinical and Genetic Risk Factors for Drug-Induced Liver Injury Associated with Anti-Tuberculosis Treatment—A Study from Patients of Portuguese Health Centers. J. Pers. Med. 2022, 12, 790. https://doi.org/10.3390/jpm12050790

AMA Style

Cavaco MJ, Alcobia C, Oliveiros B, Mesquita LA, Carvalho A, Matos F, Carvalho JM, Villar M, Duarte R, Mendes J, et al. Clinical and Genetic Risk Factors for Drug-Induced Liver Injury Associated with Anti-Tuberculosis Treatment—A Study from Patients of Portuguese Health Centers. Journal of Personalized Medicine. 2022; 12(5):790. https://doi.org/10.3390/jpm12050790

Chicago/Turabian Style

Cavaco, Maria João, Celeste Alcobia, Bárbara Oliveiros, Luís Alcides Mesquita, Aurora Carvalho, Fernando Matos, José Miguel Carvalho, Miguel Villar, Raquel Duarte, João Mendes, and et al. 2022. "Clinical and Genetic Risk Factors for Drug-Induced Liver Injury Associated with Anti-Tuberculosis Treatment—A Study from Patients of Portuguese Health Centers" Journal of Personalized Medicine 12, no. 5: 790. https://doi.org/10.3390/jpm12050790

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop