Review
Metabolomic profiling for drug-induced liver injury with autoantibodies

https://doi.org/10.1016/j.intimp.2022.109084Get rights and content

Highlights

  • The metabolic bias of different DILI with autoantibodies are great differences.

  • The metabolic characteristics of some DILI with autoantibodies are similar to AIH.

  • The difference among DILI with autoantibodies is mainly lies in energy metabolism.

  • DILI with autoantibodies may need differentiated management in clinical practice.

Abstract

Backgrounds

Drug induced liver injury (DILI) is sometimes similar to autoimmune hepatitis (AIH) in serology and histology. Clinicians empirically screened DILI with significant autoimmune characteristics to implement clinical intervention. We tried to characterize DILI with autoantibodies by metabolomics.

Methods

Untargeted metabolomics coupled with pattern recognition approaches were performed on sera samples including AIH (n = 59), DILI with autoantibodies (DILIAb+, n = 68), and DILI without autoantibodies (DILIAb-, n = 75). The differential metabolites and fingerprint metabolites between AIH and DILIAb- were screened by orthogonal partial least squares-discriminant analysis and hierarchical clustering respectively.

Results

Of the 388 annotated differential metabolites between AIH and DILIAb-, 74 fingerprint metabolites were screened. The eigenmetabolite compressed from the fingerprint possessed high discrimination efficacy (AUC:0.891; 95 %CI, 0.838–0.944). In the fingerprint-based PCA model, AIH and DILIAb- were separated into three regions: the “pure region” of AIH (Region 1), the “pure region” of DILIAb- (Region 3), the mixture region of AIH and DILIAb- (Region 2). After incorporated into the PCA model, DILIAb+ samples were distributed into the three regions, indicating that DILIAb+ samples had different etiological tendencies. Moreover, the fingerprint-based radar model verified the results of PCA model characterizing DILIAb+. Notably, the antibody titers of DILIAb+ in the three regions did not differ significantly, while the response rates for glucocorticoids were obviously different. The metabolic difference among DILIAb+ in different regions mainly lies in energy metabolism.

Conclusions

In terms of metabolic signature, DILIAb+ may not be a community of same pathogenesis, including AIH-inclined parts. Which deserves further study.

Introduction

Drug induced liver injury (DILI) refers to the disease caused by drugs or their metabolites and / or liver allergic reaction to drugs and metabolites [1]. DILI is more common in women, and most patients are acute, self-healing, and negative in autoantibodies [2]. However, clinically, some patients with DILI can be accompanied by autoantibody [3]. Some DILI with autoantibodies had serological and histological characteristics of classical autoimmune hepatitis (AIH) [2], [4]. For example, Ynto S. de Boer et al. [3] reported that nearly half of patients with liver injury caused by tetracycline and furantoin had serological characteristics similar to AIH. In DILI induced by clomiphene, serum IgG increased significantly in 80 % of patients, and anti-nucleic antibody (ANA) or anti-smooth muscle antibody (ASMA) was positive in 60 % of patients [5]. DILI induced by immune checkpoint inhibitors is also accompanied by autoantibodies [6]. In addition to DILI with autoantibodies, bleomycin can also cause systemic sclerosis, which is also an autoimmune disease [7]. Besides, the classic histological characteristics of AIH, such as interface hepatitis, infiltration of plasma cells, lymphocytes and eosinophils in portal area, central lobular necrosis and so on, can also appear in DILI patients [8], [9].

To describe these DILI, some blended terms such as “autoimmune-like” DILI (AL-DILI)[2], “AIH DILI” [10] and so on came up. Albert J. Czaja called DILI with autoantibodies as drug-induced liver injury with autoimmune characteristics [11]. The clinical phenotype of AIH can be mimicked by some DILI with autoantibodies, or vice versa [2], [10]. Weiler-Normann and Schramm [12] proposed to divide the relationship between AIH and DILI into three combination types: AIH with DILI, drug induced AIH and immune mediated DILI in 2011. In a long term follow-up study of 685 patients with DILI, 23 patients had been hospitalized for liver disease and 5 of these were diagnosed with AIH after a mean of 5.8 years [13]. These patients need to take immunosuppressant for a long time to maintain the normal function of the body. Although some DILI with autoantibodies have similar serological and histological characteristics of classic AIH, and similar treatment response as AIH, the difference from AIH is that they have no recurrence after stopping immunosuppressive treatment [8], [11], [14], [15]. In addition, some DILI with autoantibodies did not respond to glucocorticoids [16], [17]. Their liver function can gradually recover when the offending drugs are stopped. Based on these situations, some clinicians or scholars tend to carry out glucocorticoid therapy only for those DILI patients with significant autoimmune signature [11], [18]. While for DILI without autoantibodies, it is generally considered that glucocorticoids therapy is not necessary. If glucocorticoids therapy is wrongly used, the risks outweigh the benefits for these patients [19]. Unfortunately, the current screening of those DILI patients with significant autoimmune signature is only empiric fashion of clinicians [18], [20].

Above observation suggest that although these DILI are accompanied by autoantibodies, there may be significant internal variation among them. In other words, we argue that DILI patients with autoantibodies may not be a community of same pathogenesis, and the pathologic mechanism of some patients may be similar to that of AIH patients or DILI patients without autoantibodies. Therefore, it is necessary to systematically characterize DILI with autoantibodies by holistic omics, so as to better understand the internal variation of these patients.

Metabolomics is a phenotypic method for studying metabolites, small molecule substrates, and intermediary metabolites which can directly reflect the potential biochemical activities and states of tissues or cells [21]. It is a powerful method to explore specific biomarkers of different liver diseases [22], and it is also very helpful to the diagnosis, treatment and prognosis of diseases [23]. Therefore, in this study, metabolomics was used to classify the metabolic variation of DILI with autoantibodies, in order to provide reference for clinical screening of DILI with significant autoimmune characteristics, so as to solve the current situation that clinicians can only empirically screen such patients.

Section snippets

Study design

Biobanked serum samples of patients with AIH or DILI hospitalized in the Fifth Medical Center of Chinese PLA General Hospital from January 31, 2015 to May 19, 2017 were collected, and baseline demographic and clinical data were obtained, regardless of age and gender. All cases were reassessed by two intermediate doctors. According to the diagnostic scoring system of AIH revised by the International Autoimmune Hepatitis Group (IAIHG) in 1999, AIH cases without glucocorticoid treatment and score

Demographic and clinical characteristics of the study Cohort

The demography and clinical characteristics among AIH, DILIAb+, and DILIAb- were presented in Table 1. There was no significant difference in the mean ages among AIH (49.3; 95 %CI: 47.6, 51.0), DILIAb+ (47.4; 95 %CI: 47.8, 50.0), and DILIAb-(47.9; 95 %CI: 45.1, 50.6). Female was dominant in all three groups. The levels of AST/ALT, IgG, Pt, γ- Globulin and the proportion of ascites gradually decreased in AIH, DILIAb+, and DILIAb-. The levels of ALT, CHE, TC, TG, ALB, WBC and PLT in DILIAb+ were

Discussion

Some DILIAb+ samples have serological and histological characteristics similar to classical AIH, which is easy to be confused in clinical diagnosis. It has shown that some DILI can progress to AIH with the extension of time [13]. When using glucocorticoids to treat DILI clinically, it is suggested paying attention to the selection of indications and screening DILI with significant autoimmune characteristics for treatment [30]. In this study, we found that DILIAb+ may be was not a community of

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by National Natural Science Foundation of China (Nos. 82074112, 81630100 and 82104702), National Science and Technology Major Project (No. 2015ZX09501-004-001-008) and Project of China PLA General Hospital (Nos. 2019-JQPY-003 and 2019MBD-023)

References (39)

  • G. Lenaz et al.

    Mobility and function of coenzyme Q (ubiquinone) in the mitochondrial respiratory chain

    BBA

    (2009)
  • M.L. Genova et al.

    Functional role of mitochondrial respiratory supercomplexes

    BBA

    (2014)
  • N.S. Chandel et al.

    Reactive oxygen species generated at mitochondrial complex III stabilize hypoxia-inducible factor-1alpha during hypoxia: a mechanism of O2 sensing

    J. Biol. Chem.

    (2000)
  • K.P. Robinson et al.

    Defining intermediates and redundancies in coenzyme Q precursor biosynthesis

    J. Biol. Chem.

    (2021)
  • G.A. Kullak-Ublick et al.

    Drug-induced liver injury: recent advances in diagnosis and risk assessment

    Gut

    (2017)
  • M. Sebode et al.

    “Autoimmune(-Like)” Drug and Herb Induced Liver Injury: New Insights into Molecular Pathogenesis

    Int. J. Mol. Sci.

    (2017)
  • A. Hisamochi et al.

    An analysis of drug-induced liver injury, which showed histological findings similar to autoimmune hepatitis

    J. Gastroenterol.

    (2016)
  • E. Pariente et al.

    Hepatitis caused by clometacin (Dupéran). Retrospective study of 30 cases. A model of autoimmune drug-induced hepatitis?

    Gastroenterologie clinique et biologique

    (1989)
  • M. Au et al.

    Checkpoint inhibitor induced steroid refractory drug-induced liver injury

    Intern Med J

    (2021)
  • Cited by (0)

    1

    These authors contributed equally to this work.

    View full text