Discussion and conclusions
The peak area of metabolites, relative to IVM, was used to estimate the relative abundance of each metabolite. In 60-min microsomes reactions, the five most abundant metabolites were M3 > M1 > M5 > M6 > M9 (Table 1). Metabolites M1, M3, and M6 were found after 24-hour exposure to primary human hepatocytes and in volunteer blood samples taken 24 hours post IVM administration. This is the first report of IVM metabolites identified from human hepatocytes and clinical blood samples.
The IVM demethylation, oxidation, and monosaccharide metabolites identified from microsomes in this study were consistent with those reported previously from human microsomes (Zeng et al., 1998). However, four additional IVM metabolites, including ketone and carboxylic derivatives, were also found in our study. Advancements in UHPLC technology (Churchwell et al., 2005) and state-of-the-art high resolution mass spectrometry (Meyer & Maurer, 2012; Ramanathan et al., 2011; Theodoridis et al., 2012) used here improved the sensitivity for metabolite detection. By using pure CYP enzymes, we observed evidence for two CYP metabolism pathways for IVM; CYP3A5 produced demethylated IVM (M1) and CYP2C8 produced hydroxylated IVM (M13).
The chemical structures of the most abundant metabolites were obtained by NMR for M1 and M3 but not for M6. The metabolic pathway data suggest that M1, M3, and M6 were all produced by CYP3A4, and that M6 is a combination of oxidation and demethylation. Thus, we propose that M6 is 3″-O-demethyl, 4-hydroxymethyl-ivermectin, a further (common) metabolite product of M1 and M3. With two sites of transformation occurring in M6 (demethylation and oxidation), it is more polar and elutes earlier than M1 and M3. Additional reversed phase chromatography data supports the M6 structure based on the elution order. The elution order in this study is also consistent with Zeng et al. 1998.
Interestingly, many low abundance metabolites produced in microsomes were not detected from primary human hepatocytes in culture nor from human volunteer blood after IVM administration. Several factors that could influence the metabolic function of hepatocytes in vitro , including initial cell suspension, confluence density of adherent cells, and drug concentration. The lower number of metabolites found in volunteer blood samples compared to microsomes could be because of phospholipids in blood samples. In vitro systems are also more efficient in producing metabolites and do not have elimination pathways (such as renal elimination) compared to in vivo systems, which could have an impact on detection results. In future studies, we will characterize IVM metabolites produced at later time points from hepatocytes and human blood, when metabolite abundance may possibly be altered compared to the 24-hour time point thus allowing characterization of in vivo metabolism over time. To support these efforts, we have performed a clinical trial (NCT03690453) to asses the pharmacokinetic profile of key IVM metabolites in orally treated volunteers over several weeks, characterized and synthesized these metabolites, and evaluated their potential mosquito-lethal and antimalarial effects. A better understanding of IVM metabolite pharmacokinetics can provide further insight into pharmacodynamics and efficacy for NTDs, especially ones requiring multiple administrations such as scabies and strongyloidiasis. Furthermore, these IVM metabolites may inhibit viral replication and should be evaluated against SARS-CoV-2, the causative agent of COVID-19.
In conclusion, we report for the first time, novel IVM metabolites from human liver microsomes and IVM metabolites from primary human hepatocytes and from human blood after oral IVM dosing. Importantly, we identify that the two major IVM metabolites in humans are 3″-O -demethylation IVM and 4-hydroxymethyl IVM.