Merrie Mosedale, Ph.D. - US grants

Project Summary The high failure rate of drugs late in clinical development is an indication that nonclinical in vitro models and animal models are not accurately predicting compound performance in humans. Only 1 out of 10,000 molecules identified by pharmaceutical companies as a potential drug candidate is successfully advanced through FDA approval. Drug toxicity in general and drug-induced liver injury (DILI) in particular, represent the major cause of drug attrition or removal from the marketplace. There is a critical gap between the need for predictive in vitro platforms to evaluate the effects of compounds on human health and the availability of solutions which meet the throughput and accuracy required. To meet this need for increased biological relevance, the next generation of toxicity testing is beginning to incorporate fluidics and more complex cellular models into in vitro compound safety and efficacy testing. Non-fluidic (static) tissue culture plates have been a staple of the in vitro pharmaceutical testing market for decades and are the major model system used for liver safety testing. However, static cell culture systems fail to generate biologically relevant gradient drug exposures and may lack the viability and metabolic competency essential for safety testing. Additionally, these static systems do not allow for discrimination between primary drug effects and those mediated by metabolic breakdown products or cellular responses. While these static systems afford a picture of the acute toxicity of the compound itself, this picture is incomplete, at best, and potentially leads to the progression of compounds with serious safety issues into animal studies and clinical trials. These late stage failures come at huge financial costs to pharmaceutical companies, creating a significant need in the marketplace We have developed the SciFlowTM 1000 Fluidic Culture System to address existing shortcomings in drug safety testing. SciFlow 1000 is an innovative, gravity driven, fluidic tissue culture system providing highly biologically relevant compound exposures, and an innate ability to distinguish between parent drug and metabolite effects. The SciFlow System is based on a standard, SBS compliant, 96-well plate format with the addition of fluidic pathways connecting the wells along each row of the plate. This enables the evaluation of gradient compound concentrations on cells, under dynamic one way fluidic conditions that are more representative of the in vivo environment. SciFlow is designed as an open platform, supporting the culture of many cell types, in both two-dimensional (2D) and three-dimensional (3D) formats, and in a more biologically relevant fashion. To validate its diverse culture capabilities, SciFlow?s compartments have been populated with cells representing a wide variety of phenotypes including primary liver cells (hepatocytes: human, dog, rat, mouse, etc.) and many diverse cell lines (HepG2, HepaRG, Caco2, etc.). Preliminary compound toxicity studies have been completed, utilizing many biochemical and high content imaging assays to assess cellular outcomes. This proposal describes the development of an SOP to leverage the benefits of the SciFlow 1000 to provide improved drug-induced liver injury prediction. The aims of this project are to: 1) Optimize existing exposure and assay protocols in the SciFlow 1000. 2) Use optimized assays to test a library of 21 compounds of known and varying DILI on 3 liver cell models (primary human hepatocytes, co-cultures of primary human hepatocytes and non-parenchymal cells, and HepaRG cells). Utilize that dataset to drive selection of a panel assays to include in a final SciFlow DILI predictive SOP. 3) Demonstrate the capabilities of the SciFlow DILI predictive SOP in two blinded studies. This work will be completed in collaboration with the UNC Eshelman School of Pharmacy Institute for Drug Safety Sciences. The outcomes of this study will be a complete solution for conducting predictive toxicology analyses on new drug candidates with previously unattainable levels of sensitivity and specificity. These accurate in vitro to in vivo extrapolations (IVIVE) will decrease the number of compounds with serious toxicity liabilities proceeding into the later pre-clinical and early clinical drug development, saving pharmaceutical companies both time and money while enabling the ultimate goal of more rapidly providing safe and effective therapeutics.

PROJECT SUMMARY/ABSTRACT Many drugs pass through preclinical and early clinical studies before safety concerns are realized, putting patients at risk and creating a bottleneck in the drug development process. The long-term objective of our research is to improve human risk assessment in drug safety testing. Genetic susceptibility is an important feature of adverse drug reactions not currently represented in preclinical toxicology models. Therefore, we hypothesize that controlled incorporation of genetic diversity in preclinical safety studies would improve prediction and understanding of adverse drug reactions in humans. Furthermore, the identification of specific genes and pathways contributing to toxicity susceptibility would allow us to better understand the relationship between preclinical toxicology findings and patient response. We have previously demonstrated the utility of the Collaborative Cross (CC) mouse population to model toxicity responses that require genetic susceptibility factors. Currently, the CC approach requires large in vivo studies that are time consuming, expensive, and limited in scope. We are developing a novel in vitro CC platform containing primary cells isolated from CC lines and cultured on multi-well plates to allow for multiple concentrations, treatment regimens, and endpoints to be assayed across replicate wells in a single experiment. Our platform will enable the rapid and cost-effective identification of gene-by-treatment interactions associated with adverse drug response at all stages of drug development. We are beginning platform development with cultured CC hepatocytes. This will support an initial focus on drug- induced liver injury (DILI), which is one of the main adverse responses leading to the termination of clinical drug development programs and withdrawal of approved drugs from the market, and an area in which we have well- established expertise. The platform will include cryopreserved hepatocytes isolated from CC lines and cultured in 3D spheroids which will increase the physiological relevance of the in vitro model while decreasing the number of cells (and animals) needed overall. We will evaluate the utility of the in vitro CC platform to screen new drug candidates for DILI liability and aid in the improved estimation of maximum safe starting dose for first-in-human clinical trials (Aim 1); provide new understanding of the mechanisms of DILI and inform precision medicine risk mitigation strategies to improve patient safety and reduce the cost of drug development (Aim 2); and validate causal associations and inform species differences in genetic factors contributing to drug response (Aim 3). Collectively, our proposed research will improve preclinical drug safety screening and support the identification of genetic risk factors, mechanisms, and interspecies differences contributing to drug toxicity in humans. Together, these insights will further reduce the potential for patient harm and the cost of drug development.