Secondary Logo

Journal Logo

Original Articles: Hepatology

Analysis of Immunogenetic Factors in Idiosyncratic Drug-induced Liver Injury in the Pediatric Population

Ocete-Hita, Esther; Salmerón-Fernández, MaJ.; Urrutia-Maldonado, Emilia; Muñoz-de-Rueda, Paloma; Salmerón-Ruiz, María§; Martinez-Padilla, MaC.||; Ruiz-Extremera, Angela

Author Information
Journal of Pediatric Gastroenterology and Nutrition: May 2017 - Volume 64 - Issue 5 - p 742-747
doi: 10.1097/MPG.0000000000001502
  • Free


What Is Known

  • Idiosyncratic drug-induced liver injury in children is a rare disease.
  • It is multifactorial and complex.

What Is New

  • HLAC0401 and HLADQB0603 may provide a hepatoprotective mechanism.
  • HLADQA0102 and HLA-DR12 may be risk factors.

An idiosyncratic reaction to a drug is defined as an adverse reaction that meets the following criteria: it affects only a small proportion of those exposed to the drug (<5%); the reaction is not related to the pharmacological effect of the drug per se; there is no obvious dose-reaction relation; and it occurs with inconsistent and short-lived patterns in relation to exposure to the drug (1).

Idiosyncratic drug–induced liver injury (DILI) is a multifactorial complex disease, in which the toxic potential of the drug, together with genetic and acquired factors, can determine susceptibility and make individuals unique in their development of DILI (2,3).

A fundamental aspect of this area of research is the search for genetic traits that may explain susceptibility and enable the above phenomenon to be reliably predicted (4).

Drugs are metabolized preferentially in the endoplasmic reticulum. Here, too, the molecules of the major histocompatibility complex (or human leukocyte antigen—HLA) with their antigens are synthesized and assembled. Because HLA molecules exhibit a high degree of polymorphism and participate in the activation of the specific immune system, studies have been made of certain allelic variants and haplotypes that may be related to immunohepatotoxic reactions in adults (5–8).

Once idiosyncratic liver injury has initiated, its extent and intensity would depend on the integrity of the adaptation pathways, especially those of the innate immune system, whereas the polymorphisms producing low levels of interleukin-10 (IL-10) appear to be associated with the risk of fulminant outcome once the damage has initiated (9).

Another important function of HLA class I molecules is their regulation of natural killer (NK) cells, which form part of the innate immune system (9,10) and which effectively lyse virus-infected cells, and thus performing an important defensive role. The main functions of these cells are to present cytotoxicity and to secrete cytokines to mediate immune processes. Among various receptors known to modulate the activation and inhibition of NK cells are killer cell immunoglobulin-like receptors (KIRs). These receptors recognize HLA class I molecules and are directly related to the activation or inhibition of NK cells. The combination of HLA class I molecules and their corresponding KIRs is known to be highly polymorphic (11,12), and so KIR/HLA class I interactions have considerable potential as biomarkers. The greater specificity of NK cells in recognizing the target cell arises from the inhibition receptors, a question that has been amply studied (13). The aim of the present study is to analyse the genetic factors (HLA, cytokine polymorphisms, and KIR genotype) of children who experience an episode of DILI.



Prospective multicentre case-control study in which 9 Spanish hospitals were participated.

Subjects and Scope of Study

The subjects included in the study were 30 paediatric patients ages between 0 and 15 years selected by the attending physicians at the participating hospitals and who presented possible liver disease associated with the intake of medicines, herbal products, drugs, or toxins. These children were all registered on the Spanish record of hepatotoxic reactions among the paediatric population, which includes children who had experienced an episode of DILI since 2008. Any children who had previously experienced a hepatotoxicity event during chemotherapy were excluded from the study. As a control group, 62 subjects were selected from the Paediatric Emergency Unit at the Virgen de las Nieves University Hospital, Granada, Spain. These control subjects had experienced no previous episodes of hepatotoxicity after drug intake and were attended for processes in which there was no suspicion of DILI (60% respiratory disorders, 15% trauma, 3% urinary tract infections, 5% dermatological processes, 17% gastroenteritis).

The following inclusion criteria were applied:

Children ages 0 to 15 years

One or more of the following alert signals for possible drug-induced hepatotoxicity:

  • 1. >76 U/L aspartate transaminase;
  • 2. >0.6 m/dL conjugated bilirubin;
  • 3. >80 U/L aspartate transaminase, >2 mg/dL total bilirubin, and >516 U/L alkaline phosphatase (ALP) (simultaneously).

All cases were reported via a structured protocol, compiling information about the patients’ demographic characteristics, medical history and current status, concurrent diseases, and exposure to other drugs. The report form inquired about drugs consumed by the patient in the 3 months before hospitalization or consultation and about the use of any other therapeutic remedies, such as herbal medicines. For each drug consumed, information (provided by the patient or family member) on the start and end date of use, the dose, and the indication were noted.

To rule out possible alternative causes, a detailed clinical history was obtained for all patients with respect to biliary or liver disease, and information on risk factors associated with liver disease such as drug addiction, alcohol consumption, blood transfusion, or surgery during the 6 months before the DILI episode provoking inclusion in the study. Serological markers of acute viral hepatitis, and serum ceruloplasmin and autoantibodies related to autoimmune liver disease, were determined in all patients before diagnosis of hepatotoxicity. In every case, before DILI was diagnosed, a hepatobiliary ultrasound examination was performed. Other data compiled were the duration of use of the drug in question, the moment of onset of symptoms, and other data of clinical interest, especially those related to hypersensitivity.

The reactions were classified as follows:

  • 1. Hepatocellular injury: isolated alanine transaminase (ALT) elevation, to twice normal limits, or ALT/ALP activity, expressed as a multiple of the upper normal limit, >5.
  • 2. Cholestatic injury: isolated ALP elevation to twice normal limits, or ALT/ALP <2.
  • 3. Mixed injury: elevated ALT, elevated ALP, and ALT/ALP >2 but <5.

Establishing Causality

The diagnosis of DILI was first evaluated by the patients’ own doctors and subsequently by the coordinator of the multicentre study. The evaluation criteria used included confirmation of an appropriate temporal sequence from administration of the drug to the onset of symptoms; a clinical course that did not correspond to the effects of any concomitant disease, other drug used, or other nondrug therapy; improvement with drug withdrawal; and the existence of previous cases of hepatotoxicity to the drug in question.

Hepatotoxicity cases that met the above clinical criteria and whose causality to the drug (s) was estimated as possible, probable, or so defined by the CIOMS/RUCAM scale (14,15) were included in the study group. This scale contains a number of preselected questions that are scored according to the information available. The total score obtained is then transferred to one of the following categories of suspicion: defined or highly probable, probable, possible, improbable, and excluded.

A more detailed description of the method is presented elsewhere (16,17).

Liver Biopsies

According to standard criteria, the pathological pattern of injury was categorized into acute hepatitis (predominantly lobular inflammation, with or without confluent or bridging necrosis; absence of cholestasis), chronic hepatitis (portal inflammation with interface hepatitis, with or without portal-based fibrosis; no cholestasis), acute cholestasis (hepatocellular and/or canalicular cholestasis; minimal inflammation), chronic cholestasis (duct sclerosis and loss; periportal cholate stasis; portal-based fibrosis; copper accumulation), cholestatic hepatitis (acute or chronic hepatitis pattern and cholestasis), and other patterns (18).

Human Leukocyte Antigen Genotyping

HLA class I and HLA class II genotyping. Class I HLA-A, B and C loci and class II HLA-DRB1, DQB1, DQA1, and DP loci genotyping was performed in 92 children. Genotyping was carried out by SSO LABType (One Lambda, Canoga Park, CA). LABType applies Luminex technology to the reverse SSO DNA typing method. First, target DNA is polymerase chain reaction (PCR) amplified using a group-specific primer. The PCR product is biotinylated, which allows it to be detected using R-phycoerythrin-conjugated streptavidin (SAPE). The PCR product is denatured and allowed to rehybridize to complementary DNA probes conjugated to fluorescently coded microspheres. A flow analyser, Luminex 200 system (Luminex xMAP), identifies the fluorescent intensity of PE (phycoerythrin) on each microsphere. The assignment of the HLA typing is based on the reaction pattern compared with patterns associated with published HLA gene sequences. HLA allele assignment was performed with HLA Fusion Version 4.0 software (One Lambda) following the manufacturer's instructions.

Killer Cell Immunoglobulin-like Receptor Genotyping

We genotyped KIR in all the subjects (14 KIR genes and for 2 pseudo-KIR genes) using KIR SSO Genotyping Test (One Lambda, Inc), applies Luminex technology to the reverse SSO DNA typing method. First, target DNA is PCR amplified using 3 separate group-specific primer sets targeting Exons 3+4, 5, and 7–9. Each PCR product is biotinylated, which allows later detection using R-phycoerythrin-conjugated streptavidin (SAPE). Each PCR product is denatured and allowed to hybridize to complementary DNA probes conjugated to fluorescently coded microspheres. After washing the beads, bound amplified DNA from the test sample is tagged with SAPE. A flow analyser, the Luminex 200 system (Luminex ×MAP), identifies the fluorescent intensity of PE (phycoerythrin) on each microsphere. The assignment of genotypes is based on the reaction pattern compared to patterns associated with published KIR gene sequences.

Data analysis was performed by Luminex ×PONENT Software and HLA Fusion Version 4.0 software (One Lambda) following the manufacturer's instructions. All analyses were performed at the research unit of the Complejo Hospitalario Universitario de Granada.


Identification and analysis of genetic polymorphisms of the cytokine IL-10, tumour necrosis factor alpha (TNF-α), transforming growth factor beta (TGF-β1), IL-6, and interferon gamma (IFN-γ). Genotyping: cytokine polymorphisms were obtained from DNA samples collected with ethylenediaminetetraacetic acid and extracted from whole blood using the High Pure PCR Template Preparation Kit (Roche Diagnostic, Mannheim, Germany), in accordance with the manufacturer's instructions. Cytokine genetic polymorphisms were analysed using the Cytokine Genotyping Tray (One Lambda, Inc), following the manufacturer's protocol, by PCR-SSP. After the PCR process, the amplified DNA fragments were separated by agarose gel electrophoresis and visualized by staining with ethidium bromide and exposure to ultraviolet light. Interpretation of the PCR-SSP results was based on the presence or absence of a specific amplified DNA fragment. The following polymorphisms were studied: TNF-α (-308G/A), TGF-β1(-10T/A; 25C/G), IL-10 (−1082G/A; −819T/C; −592 C/A), IL-6 (−174G/C), and IFN-γ (+874T/A).

Statistical Analysis

Data management and analysis were performed using SPSS19.0 for Windows (SPSS Inc, Chicago, IL). The criterion for statistical significance was P ≤ 0.05. Qualitative variables are expressed as absolute numbers with percentages, and quantitative variables are expressed as mean values ± standard deviation. Intergroup comparisons were made by chi-square for categorical variables and by Student t test for quantitative variables. The Bonferroni correction for multiple tests was applied.

Ethical Considerations

The present study was carried out in accordance with the latest version of the Declaration of Helsinki and with applicable local requirements. The study was approved by the local Ethics Committee. All patients gave their signed informed consent for inclusion in the study.


Table 1 shows the demographic, clinical, and laboratory findings concerning the children who had experienced DILI. Seven biopsies were performed (23.4%) (Table 1). The drugs in question are listed in Table 2.

Demographics, clinical, laboratory, and biopsy parameters of hepatotoxicity patients (n = 30)
Drugs responsible for the episode of drug-induced liver injury

Human Leukocyte Antigen Study

No statistically significant differences were observed between any HLA-A or HLA-B between controls and patients.

The children with HLA-C0401 were less likely to present hepatotoxicity (odds ratio [OR] = 0.3; P = 0.045; 95% confidence interval [CI] = 0.09–0.097). Otherwise, in the HLA-C group there were no differences between controls and patients (Fig. 1).

Distribution of human leukocyte antigen C (HLA-C): patients versus control. No = absent; Yes = present.

HLA-DRB12 and HLA-DQA0102 were more frequent among the children with DILI (OR = 9.3; P = 0.05; 95% CI = 1–88.1; OR = 2.51; P = 0.058; 95% CI = 0.9–6.5, respectively), whereas the presence of the HLA-DQA0603 allele was greater among the control group (OR = 0.16; P = 0.087; 95% CI = 0.2–1.3).

HLA-DPA was not associated with the disease in any case.

Killer Cell Immunoglobulin-like Receptor Study

K2DP1, K3DP1, K3DL3, K3DL2, and K2DL4 were present in 100% of the children with DILI (P = 0.002, 0.028, 0.008, and 0.05, respectively).

The children with KIR K2DS5 were less likely to present hepatotoxicity (OR = 0.4, P = 0.051, 95% CI = 0.1–1.003).

Although some of the children presented 100% activator or inhibitor KIRs, there were no differences in this respect between the patients and the controls.

The study of KIR haplotypes showed that 80.8% of the children with the BB haplotype (activator of NK cells) were controls, although the differences were not statistically significant. Neither were there significant differences when the alleles were considered separately (AA + AB vs BB and BB + AB vs AA).

When the ligands were studied, first separately and then by KIR-ligand, no differences were found between the patients and the controls.

When 2 KIR alleles were from the same gene, a comparison was made between homozygous children for 1 allele and heterozygous children (KIR2DL2/KIR2DL3 and KIR3DL1/KIR3DDS1). No differences were found between the groups.

Cytokines Study

Fifty-three percent of the children who had experienced a hepatotoxic reaction had a low producer phenotype for IL-10, whereas 53% of controls had an intermediate phenotype, and tended to produce more IL-10 than the patients, although this difference did not reach statistical significance (P = 0.1).

The comparison of TNF-α, TGF-β, IL-6, and IFN-γ revealed no differences between patients and (19) controls.


Numerous immunogenetic studies have been conducted in adult patients with DILI (13,19–28), but few prior references exist to comparable studies in children, especially to specific populations such as those presenting DILI after being given antituberculosis drugs (29), and almost all concern anecdotal cases or limited series (16,17,20–22).

We have considered pathological transaminase value 2LSN. Since the publication of Navarro et al (23) proposals were made for this limit to be increased to 3× ULN and, since 2011, to 5× ULN. The reasons for this progressive increase have, however, mainly concerned diseases affecting the adult population (hepatic steatosis, in particular). Nevertheless, descriptions of DILI among the paediatric population have increased significantly in recent years. We believe it will soon be possible to establish more appropriate criteria for a child population, and that the diagnosis and potential causes of DILI will be much better suited to children (eg, by eliminating factors such as alcoholism, age older than 65 years, etc from the list of possible causes).

In the study population, DILI was more common among women, as reported by Molleston et al (20). In another important series concerning Chinese children, DILI was, however, more frequent among male patients (31).

Conducting genetic studies on hepatotoxicity in adults presents a number of difficulties and challenges, including the fact that the drugs used are heterogeneous in their chemical properties, therapeutic effects, and biological targets. In consequence, most of the genetic associations in DILI that have been established are in populations that have used specific drugs, presumably under the assumption that these associations will be specific to each drug. Although this is a valid assumption, it is also true that genetic susceptibility may be common to a diverse group of drugs or to all drugs in general.

The “signature” with which a drug induces hepatotoxicity (relatively constant latency period and type of injury) should be viewed with caution, because these manifestations often vary from one subject to another. In addition, the most severe forms of liver toxicity (provoking acute liver failure) are relatively infrequent (24). Because the occurrence of hepatotoxicity is low and it is underdiagnosed in the paediatric population, the above-mentioned circumstances limit our ability to identify a sufficient number of cases produced by a single drug or drug group, which is why this study seeks genetic markers that may be common to all the drugs that cause idiosyncratic hepatotoxicity in children.

In adults, various associations have been described between HLA alleles and the probability of hepatotoxicity induced by certain drugs (25–30,32). A question of current interest is whether genetic variations in molecules of the major histocompatibility complex (HLA II) determine a certain predisposition to adverse hepatic reactions due to the role played by these molecules in the immune response.

Our comparison of patients and controls revealed no differences between any HLA class I, A or B. HLA-C0401 was more commonly found in the control group, and seemed to exert a protective effect against liver damage.

With respect to HLA class II, there was a greater presence of HLA-DQA0102 in the group of children with hepatotoxicity and of HLA-DQA0603 in the control group (protective factor). Another association observed was that of HLA DRW12. Although this allele was present in only 5 children, the possibility of a hepatotoxic drug reaction when the child was positive for HLA DRW12 was 9.2 times higher than that in the control group. These associations have not been described in previous studies with adults. Our findings show that some drugs can cause liver damage by activating the adaptive immune system.

As some genetic effects may have only small-scale clinical consequences, we believe that researchers should consider large sample sizes to discover more genetic associations. In any case, the associations found would need to be reproduced to confirm their real significance.

According to a recent report, the balance between pro and anti-inflammatory mediators may determine the susceptibility and severity of liver injury in experimental acetaminophen-induced hepatotoxicity. The activation of innate immunity of the liver and the resulting inflammation may also play an important role in idiosyncratic hepatotoxicity (24). Consequently, IL-10, IL-4, and TNF-α are strong candidates as the susceptibility genes responsible for liver damage secondary to toxicity.

IL-10 is a potent anti-inflammatory agent and an immunoregulator of Th2 cytokines. It acts directly on the macrophages, reducing their production of IL-12, and limits Th1 differentiation and the production of harmful cytokines such as TNF-α, IFN-γ, and IL-8 (4–6). In an experimental model, IL-10 has been found to protect against acetaminophen-induced hepatotoxicity in mice (4).

In our study, 53% of the children with DILI had a low phenotype, tending to produce small quantities of IL-10. Fifty-three percent of the controls presented an intermediate phenotype, and so were more likely than the patients to produce IL-10. These differences were, however, not statistically significant. In studies of adults, low IL-10-producing polymorphisms were not found more frequently in patients with DILI than in controls. In one study (33), such a low IL-10-producing polymorphism was, however, significantly associated with the absence of eosinophils in peripheral blood, and with a severe or fulminant evolution, which suggests that once the injury has initiated, a decreased Th2 response could favour the progression of liver inflammation and necrosis.

These findings suggest that the mediator involved in paediatric hepatotoxicity is IL-10, because no intergroup differences were found in our analyses of TNF-α, TGF-β, IL-6, and IFN-γ.

With respect to the KIR levels, it is noteworthy that in the control group no child was positive for KIR3DL1, which is an inhibitor of NK cells. Similarly, in the control group, 80.8% of the children had a BB KIR haplotype, that is, they tend to be activators of NK cells. This result, however, did not reach statistical significance. These 2 findings suggest that NK cells have a greater capacity for action in the control group; in other words, there is a greater inhibition of NK cells in children with DILI. No previous studies have been conducted on the possible relation between KIR genotypes and DILI, in adults or children.

Some research groups have proposed models for specific KIR-HLA combinations that determine the activating or inhibitory nature of NK cell activity, and which may also determine the patient's susceptibility or resistance to certain pathologies, or modify their clinical course (31). In liver transplant patients, an association between KIRs and their ligands and the progression of hepatic fibrosis has been described (34). Further research is, however, needed into the exact role of KIRs and their ligands with respect to hepatotoxicity. This project will not be easy, because no mouse models exist that would facilitate the study of these interaction systems, due to the absence of a KIR system in the mouse. Accordingly, many of the questions that remain outstanding must be addressed by means of in vitro studies with human cells or those of higher primates.

The aim of the present study was to achieve a better understanding of the relation between genotype and individual susceptibility or resistance to developing idiosyncratic liver disease induced by drugs or herbal products, in a paediatric population. Its main limitation is the size of the sample; few patients were analysed, because of the small number of cases diagnosed with this disease. Drug-induced idiosyncratic hepatotoxicity is a challenge for clinical researchers because its clinical manifestations may be nonspecific, because diagnosis is by exclusion, because specific markers are not available, and because our understanding of the mechanism involved in the damaged tissue remains incomplete.


Although HLAC0401 and HLADQB0603 may provide a hepatoprotective mechanism in the paediatric population, HLADQA0102 and HLA-DR12 are more commonly found in sick children and their presence may be related to liver damage directly induced by a drug or herbal compound.

Polymorphisms that are low producers of IL-10 occur more frequently in children who have experienced hepatotoxicity. The genetic polymorphisms of TNF-α, TGF-β, IL-6, and IFN-γ are not related to the risk of liver damage. The KIR inhibitor KIR3DL1 was not present in any child in the control group. Its positivity in children who have experienced liver damage from drugs or herbal products could be related to its inhibition of NK cells.


1. Uetrecht J, Naisbitt DJ. Idiosyncratic adverse drug reactions: current concepts. Pharmacol Rev 2013; 65:779–808.
2. Sarges P, Steinberg JM, Lewis JH. Drug-induced liver injury: highlights from a review of the 2015 literature. Drug Saf 2016; 39:801–821.
3. Weiler S, Merz M, Kullak-Ublick GA. Drug-induced liver injury: the dawn of biomarkers? F1000Prime Rep 2015; 7:34.
4. Shah RR. Pharmacogenetics in drug regulation: promise, potential and pitfalls. Philos Trans R Soc Lond B Biol Sci 2005; 360:1617–1638.
5. Grove JI, Aithal GP. Human leukocyte antigen genetic risk factors of drug-induced liver toxicology. Expert Opin Drug Metab Toxicol 2015; 11:395–409.
6. Aithal GP, Grove JI. Genome-wide association studies in drug-induced liver injury: step change in understanding the pathogenesis. Semin Liver Dis 2015; 35:421–431.
7. Urban TJ, Daly AK, Aithal GP. Genetic basis of drug-induced liver injury: present and future. Semin Liver Dis 2014; 34:123–133.
8. Stephens C, Lopez-Nevot MA, Ruiz-Cabello F, et al. HLA alleles influence the clinical signature of amoxicillin-clavulanate hepatotoxicity. PLoS One 2013; 8:e68111.
9. Vivier E, Raulet DH, Moretta A, et al. Innate or adaptive immunity? The example of natural killer cells. Science 2011; 331:44–49.
10. McQueen KL, Parham P. Variable receptors controlling activation and inhibition of NK cells. Curr Opin Immunol 2002; 14:615–621.
11. Middleton D, Gonzelez F. The extensive polymorphism of KIR genes. Immunology 2010; 129:8–19.
12. Yawata M, Yawata N, Draghi M, et al. Roles for HLA and KIR polymorphisms in natural killer cell repertoire selection and modulation of effector function. J Exp Med 2006; 203:633–645.
13. Rajalingam R. Human diversity of killer cell immunoglobulin-like receptors and disease. Korean J Hematol 2011; 46:216–228.
14. Danan G, Benichou C. Causality assessment of adverse reactions to drugs—I. A novel method based on the conclusions of international consensus meetings: application to drug-induced liver injuries. J Clin Epidemiol 1993; 46:1323–1330.
15. Benichou C, Danan G, Flahault A. Causality assessment of adverse reactions to drugs—II. An original model for validation of drug causality assessment methods: case reports with positive rechallenge. J Clin Epidemiol 1993; 46:1331–1336.
16. Ocete-Hita E, Sameron-Fernandez MJ, Gonzalez Gomez JM, et al. Hepatotoxicity from the consumption of herbalist products by a paediatric population. Nutr Hosp 2015; 32:652–655.
17. Hita EO, Garcia JA, Gonzalez JC, et al. Amoxicillin-clavulanic acid hepatotoxicity in children. J Pediatr Gastroenterol Nutr 2012; 55:663–667.
18. Kleiner DE. The pathology of drug-induced liver injury. Semin Liver Dis 2009; 29:364–372.
19. Lucena MI, Andrade RJ, Kaplowitz N, et al. Phenotypic characterization of idiosyncratic drug-induced liver injury: the influence of age and sex. Hepatology 2009; 49:2001–2009.
20. Molleston JP, Fontana RJ, Lopez MJ, et al. Characteristics of idiosyncratic drug-induced liver injury in children: results from the DILIN prospective study. J Pediatr Gastroenterol Nutr 2011; 53:182–189.
21. Lim JR, Faught PR, Chalasani NP, et al. Severe liver injury after initiating therapy with atomoxetine in two children. J Pediatr 2006; 148:831–834.
22. Anderson GD. Children versus adults: pharmacokinetic and adverse-effect differences. Epilepsia 2002; 43 (suppl 3):53–59.
23. Navarro VJ, Senior JR. Drug-related hepatotoxicity. N Engl J Med 2006; 354:731–739.
24. Kaplowitz N. Idiosyncratic drug hepatotoxicity. Nat Rev Drug Discov 2005; 4:489–499.
25. Daly AK, Donaldson PT, Bhatnagar P, et al. HLA-B5701 genotype is a major determinant of drug-induced liver injury due to flucloxacillin. Nat Genet 2009; 4:816–819.
26. Kindmark A, Jawaid A, Harbron CG, et al. Genome-wide pharmacogenetic investigation of a hepatic adverse event without clinical signs of immunopathology suggests an underlying immune pathogenesis. Pharmacogenomics J 2008; 8:186–195.
27. Martin AM, Nolan D, James I, et al. Predisposition to nevirapine hypersensitivity associated with HLA-DRB10101 and abrogated by low CD4 T-cell counts. AIDS 2005; 19:97–99.
28. Singer JB, Lewitzky S, Leroy E, et al. A genome-wide study identifies HLA alleles associated with lumiracoxib-related liver injury. Nat Genet 2010; 42:711–714.
29. Spraggs CF, Budde LR, Briley LP, et al. HLA-DQA102:01 is a major risk factor for lapatinib-induced hepatotoxicity in women with advanced breast cancer. J Clin Oncol 2011; 29:667–673.
30. Yuan J, Guo S, Hall D, et al. Toxicogenomics of nevirapine-associated cutaneous and hepatic adverse events among populations of African, Asian, and European descent. AIDS 2011; 25:1271–1280.
31. Rajagopalan S, Long EO. Understanding how combinations of HLA and KIR genes influence disease. J Exp Med 2005; 201:1025–1029.
32. Hirata K, Takagi H, Yamamoto M, et al. Ticlopidine-induced hepatotoxicity is associated with specific human leukocyte antigen genomic subtypes in Japanese patients: a preliminary case-control study. Pharmacogenomics J 2008; 8:29–33.
33. Pachkoria K, Lucena MI, Crespo E, et al. Analysis of IL-10, IL-4 and TNF-alpha polymorphisms in drug-induced liver injury (DILI) and its outcome. J Hepatol 2008; 49:107–114.
34. Buhler S, Giostra E, Gbame C, et al. A significant effect of the KIR ligand HLA-C on fibrosis progression in chronic C hepatitis with or without liver transplantation. Liver Int 2016; 36:1331–1339.

children; genetic factors; hepatotoxicity; liver injury

Copyright © 2017 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition