Narrowing the translational gap in metabolic dysfunction-associated steatotic liver disease (MASLD) drug development

<p dir="ltr">Metabolic dysfunction-associated steatotic liver disease (MASLD) currently poses a significant health burden, with a prevalence of 38% in the general population. With the parallel increase of obesity and type 2 diabetes prevalence, MASLD is currently one of the top indications for liver transplantation. Its progressive form, termed metabolic dyfunction-associated steatohepatitis (MASH), remains a major risk factor for liver-related morbidity and mortality. Despite its clinical relevance, the development of effective treatments has been hindered by the lack of physiologically relevant in vitro systems capable of capturing human-relevant disease mechanisms, pharmacological responses, and enabling compound screening for novel target discovery. This thesis aims to improve the fidelity and pathophysiological relevance of biological liver model systems for preclinical research in MASLD and MASH.</p><p dir="ltr">In the first study, we showed that primary human hepatocyte (PHH) spheroids can recapitulate insulin resistance when challenged with nutrient-excess media and free fatty acids (FFA), mimicking a key hallmark of MASLD onset. Basal expression of genes involved in gluconeogenesis and de novo lipogenesis was elevated in insulin-resistant PHH spheroids, but their expression was not markedly affected by acute insulin challenge, corroborating the insulin resistance phenotype after chronic exposure to excess nutrients and FFA. Additionally, the development of a sensitive glucose sensor enabled longitudinal measurement of glucose utilization in PHH spheroids, allowing us to quantify both insulin-dependent and -independent glucose uptake, reproducing values observed in human clamp studies and previous simulations.</p><p dir="ltr">The second study showed that incorporating non-parenchymal cells (NPC) from MASH patients with biopsy-proven fibrosis allows for the phenocopying of MASH hallmarks, including steatosis, inflammation, and fibrosis. Not only could we aggravate the fibrosing phenotype by adding FFA, but we also demonstrated that the model is amenable to pharmacological intervention. We benchmarked the model against several compounds in clinical development, including resmetirom and obeticholic acid, and observed delayed, dose-dependent reductions in pro-collagen I secretion as a proxy for fibrosing liver microtissues, consistent with clinical outcomes. Chemogenomic screening further revealed the CHRM1-TRPM8 axis as a novel regulatory pathway in hepatic fibrosis, demonstrating the platform's utility in target discovery.</p><p dir="ltr">In the final study, we systematically examined how different polymers influence compound bioavailability through absorption. Using a panel of eight polymers and hepatotoxic drugs with varying physicochemical properties, we found that hydrophobic compounds were substantially lost in highly absorptive materials such as poly(dimethyl siloxane) (PDMS), leading to false-negative toxicity results. In contrast, low-absorption materials like thiol-ene epoxy (TEE) and poly(tetrafluoroethylene) (PTFE) retained bioactive concentrations and enabled accurate toxicity profiling. This work establishes key design rules for polymer selection in drug-screening platforms and informs future assay standardization strategies.</p><p dir="ltr">Altogether, this thesis provides a coherent framework for improving the physiological relevance and translational reliability of in vitro human liver models. By integrating biological fidelity with engineering considerations, these contributions support the development of more predictive platforms for MASLD research and compound evaluation.</p><h3>List of scientific papers</h3><p dir="ltr">I. Insulin-dependent glucose consumption dynamics in 3D primary human liver cultures measured by a sensitive and specific glucose sensor with nanoliter input volume. <b>Kemas AM</b>, Youhanna S, Zandi SR, Lauschke VM. FASEB Journal. 2021 35;3 e21305. <a href="https://doi.org/10.1096/fj.202001989rr" rel="noreferrer" target="_blank">https://doi.org/10.1096/fj.202001989rr</a></p><p dir="ltr">II. Chemogenomic Screening in a Patient-Derived 3D Fatty Liver Disease Model Reveals the CHRM1-TRPM8 Axis as a Novel Module for Targeted Intervention. Youhanna S*, <b>Kemas AM</b>*, Wright S, Zhong Y, Klumpp B, Klein K, Motso A., Michel M, Ziegler N, Shang M, Sabatier P, Kannt A, Sheng H, Vilarnau N, Büttner FA, Seashore-Ludlow B, Windbergs M, Hülsmeier AJ, Hornemann T, Olsen JV, Wang Y, Gramignoli R, Sundström M, Lauschke VM. Advanced Science. 2025, 12;3 e2407572. *SY and AMK contributed equally. <a href="https://doi.org/10.1002/advs.202407572" rel="noreferrer" target="_blank">https://doi.org/10.1002/advs.202407572</a></p><p dir="ltr">III. Compound Absorption in Polymer Devices Impairs the Translatability of Preclinical Safety Assessments. <b>Kemas AM</b>*, Zandi SR*, Taebnia N, Michel M, Preiss L, Hofmann U, Lauschke VM. Advanced Healthcare Materials. 2024, 13;11 e2303561. *AMK and SRZ shared first co-authorship. <a href="https://doi.org/10.1002/adhm.202303561" rel="noreferrer" target="_blank">https://doi.org/10.1002/adhm.202303561</a></p>