Idiosyncratic Drug-Induced Liver Injury: Mechanistic and Clinical Challenges

Abstract

Idiosyncratic drug-induced liver injury (IDILI) remains a significant problem for patients and drug development. The idiosyncratic nature of IDILI makes mechanistic studies difficult, and little is known of its pathogenesis for certain. Circumstantial evidence suggests that most, but not all, IDILI is caused by reactive metabolites of drugs that are bioactivated by cytochromes P450 and other enzymes in the liver. Additionally, there is overwhelming evidence that most IDILI is mediated by the adaptive immune system; one example being the association of IDILI caused by specific drugs with specific human leukocyte antigen (HLA) haplotypes, and this may in part explain the idiosyncratic nature of these reactions. The T cell receptor repertoire likely also contributes to the idiosyncratic nature. Although most of the liver injury is likely mediated by the adaptive immune system, specifically cytotoxic CD8+ T cells, adaptive immune activation first requires an innate immune response to activate antigen presenting cells and produce cytokines required for T cell proliferation. This innate response is likely caused by either a reactive metabolite or some form of cell stress that is clinically silent but not idiosyncratic. If this is true it would make it possible to study the early steps in the immune response that in some patients can lead to IDILI. Other hypotheses have been proposed, such as mitochondrial injury, inhibition of the bile salt export pump, unfolded protein response, and oxidative stress although, in most cases, it is likely that they are also involved in the initiation of an immune response rather than representing a completely separate mechanism. Using the clinical manifestations of liver injury from a number of examples of IDILI-associated drugs, this review aims to summarize and illustrate these mechanistic hypotheses.

Keywords: adverse drug reactions; cytochromes P450; damage-associated molecular pattern molecules; immunotoxicity; innate immune response; liver injury; reactive metabolites.

Figure 1

Proposed general mechanism of idiosyncratic drug-induced liver injury (IDILI). The following are the proposed sequence of steps in the mechanism of IDILI: In order to produce an immune response, the drug must interact with MHC-II to produce signal 1 (the human form of MHC is HLA). (1) In most cases this involves the formation of a reactive metabolite by hepatocytes that covalently binds to proteins. These modified proteins can act as neoantigens that are presented by antigen presenting cells (APCs) that express MHC-II. These neoantigens are released by hepatocytes, probably packaged in exosomes. However, there are examples in which the drug does not appear to form a reactive metabolite and may interact non-covalently with MHC-II or the MHC-T cell receptor complex. In order to produce an immune response, the drug must also activate antigen presenting cells. (2) This involves the release of damage-associated molecular pattern molecules (DAMPs) such as heat shock proteins and HMGB1. The production of DAMPs can also be caused by a reactive metabolite that causes cellular dysfunction, although in some cases the release of DAMPs may involve some other mechanism such as inhibition of the bile salt export pump by the parent drug. The DAMPs are released from hepatocytes, most likely in exosomes, and activate antigen presenting cells through pattern recognition receptors. The DAMPs also lead to the recruitment of other innate immune cells. (3) The neoantigens are taken up by antigen present cells to produce signal 1, and activation of these cells by DAMPs leads to expression of costimulatory molecules such as CD80 and CD40, which provide signal 2 to CD4+ helper T cells. (4) The helper T cells are activated by the combination of signal 1 and signal 2 provided by the antigen presenting cell. The helper T cells produce cytokines that facilitate and shape the immune response. Th1 helper T cells promote a cell-mediated adaptive immune response, and Th2 helper T cells promote an antibody-mediated adaptive immune response. Most adaptive immune responses are a combination of both cell and antibody immune responses. (5a) However, because the dominant presentation of intracellular antigens such as the neoantigens produced by reactive metabolites is through MHC-I, which binds to CD8 on cytotoxic T cells, the dominant adaptive immune response in IDILI is usually a cell-mediated immune response. The first steps in this mechanism likely occur in most patients; however, unless the drug, or more likely drug-modified peptides, are recognized by MHC-II, MHC-I, and T cell receptors, there will be no adaptive immune response and no liver injury. The activated antigen presenting cells can also activate CD4+ Treg that dampen the immune response as well as release of cytokines such as IL-10 that also dampen the immune response. (5b) Therefore, unless the binding of the drug, or more likely a drug-modified peptide is very strong, the adaptive immune response will end in immune tolerance, which prevents or limits liver injury.