Hepatic responses as potential biomarkers of idiosyncratic drug reaction risk.

Idiosyncratic drug reactions (IDRs) and, in particular, idiosyncratic drug-induced liver injury (iDILI) represent a significant risk for drug development. There are multiple lines of evidence that such reactions are mediated by the adaptive immune system. The idiosyncratic nature of adaptive immune responses makes it difficult to develop biomarkers to predict the risk that drug candidates will cause such adverse reactions. However, an adaptive immune response requires an innate immune response to activate antigen-presenting cells such as macrophages, and the innate immune response to drugs should not be idiosyncratic. We had previously shown that clozapine and nevirapine produce a marked innate immune response in rodents. We sought to extend these studies to additional drugs that cause IDRs; namely, amodiaquine, carbamazepine, isoniazid, and ximelagatran. Ximelagatran was specifically chosen because, unlike the others, it does not appear to form a reactive metabolite. Amodiaquine and carbamazepine also produced changes associated with an innate immune response, such as an increase in serum corticosterone and GDF-15 and changes in hepatic mRNA for Dusp1, Gadd45b, and Cebpd. These changes resolved within hours. Not surprisingly, ximelagatran did not produce significant changes. However, it was surprising that isoniazid produced few such changes. There is evidence that ximelagatran can directly activate macrophages, and it is plausible that isoniazid can also directly activate macrophages. It is likely that changes in hepatic mRNA consistent with a stress/innate immune response would represents biomarkers of iDILI risk; however, some drugs may directly activate macrophages without the need of factors released from the liver.