RDMPK (drug metabolism and pharmacokinetics) studies are a required part of any preclinical drug discovery program. Independent of the modality or disease area, it is essential to understand the systemic and local effects of a therapy. A typical DMPK study evaluates the absorption, distribution, metabolism and excretion (ADME) characteristics of a given therapy1. Several in vitro and in vivo models are routinely used for DMPK studies. Cell-based in vitro models are simple and cost-effective to evaluate specific PK endpoints including solubility and lipophilicity, microsomal and plasma stability, binding to plasma proteins, hepatocyte toxicity and drug-drug interactions2. Well-defined in vitro models are well suited to answer specific properties of the drug such as aqueous vs lipid solubility and degradation of the drug by plasma proteins. Hepatocyte microsomes are a valuable tool to understand the metabolic path for a therapy as the microsomes include several enzymes associated with drug metabolism such as esterases, cytochrome P450 enzymes etc.2 Additionally, early tox studies in hepatocytes are a rapid way to evaluate if a given drug has extremely high toxicity2. Essentially, in vitro DMPK models are a good initial screen to evaluate biophysical characteristics and early toxicity of new therapies. However, in vivo models are required for comprehensive and systemic PK analysis.
Several species are used in DMPK studies including mice, rats, rabbits and large animals including non-human primates, dogs and chickens. A snapshot of the PK properties of novel small molecule drugs can be obtained using the RACE (rapid assessment of compound exposure) where a truncated study is performed on a small cohort of animals using one formulation, dose, sampling and route of administration3. The RACE study analyzes drug levels in plasma to prioritize compounds to move forward to a comprehensive PK study. Comprehensive PK studies include a wide range of endpoints so a study design will depend on the modality type, route of administration and formulation. However, it is important to note that there are species specific difference in drug metabolism and distribution and one of the most well characterized differences is that rodent eliminate drugs much faster than humans4. Additionally, the expression of specific CYP450 drug metabolizing enzymes vary between species. In order to account for the differences in human vs rodent ADME, humanized mouse models are increasingly being used to generate human relevant DMPK data.
One well characterized humanized mouse model is the PXB-mouse model that has a humanized liver which recapitulates human liver function including expression of critical drug metabolizing enzymes5. The PXB-mouse is generated by engrafting human hepatocytes after the ablation of endogenous mouse hepatocytes on a SCID background6. Humanized mouse models for DMPK studies that replace mouse CYP450 enzymes with human orthologs have been reported but have faced challenges as there is significant redundancy in the CYP450 enzyme family leading to confounding data. Several transgenic mouse models have been reported where specific human CYP enzymes replaced the murine version including CYP1A1/A3, CYP2A6 and CYP3A but these models have significant limitations as the murine enzyme expression can be altered upon introduction of the human protein and compensatory changes in expression of other CYP enzymes were noted7. Apart from the CYP450 transgenic models, UGT humanized mice strains were also developed but these have limited application to specific xenobiotics7. Researchers have developed a specific humanized mouse model – the 8HUM mouse to overcome the limitations of partial CYP humanization. The 8HUM mouse has 33 CYP450 human enzymes along with Car and Pxr that are receptors which regulate expression of the CYP450 family8. This large-scale replacement of murine drug metabolism with human proteins has been shown to be representative of human drug metabolism8, 9. The 8HUM mice have been shown to generate human specific drug metabolite profiles and have good microsomal stability9.
While DMPK will always be a critical component of drug discovery, it is clear that more advanced next-gen in vivo models need to be developed to improve the translation from rodent to human thus reducing the attrition of the preclinical drug discovery pipeline due to insufficient or poorly translatable PK data.
References:
2https://www.ncbi.nlm.nih.gov/books/NBK326710/
3https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3694733/
4https://www.sciencedirect.com/science/article/abs/pii/S0924857902000225
5https://www.transpharmation.com/pxb-mouse-studies
6https://www.phoenixbio.com/products/pxb-mouse
7https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6310623/
