Modeling Nonalcoholic Fatty Liver Disease in the Dish Using Human-Specific Platforms: Strategies and Limitations

“

TL;DR

This review evaluates human-specific in vitro platforms for modeling nonalcoholic fatty liver disease (NAFLD), including hiPSC-derived cells, precision-cut liver slices, intrahepatic cholangiocyte organoids, and liver-on-chip systems. No single platform fully recapitulates all NAFLD hallmarks, particularly the chronic fibroinflammatory progression and inter-organ crosstalk that occur over decades in patients. The choice of model depends on the specific research question, with hiPSC platforms offering scalability and genetic studies, while tissue-derived models better preserve native architecture and cellular context.

Full summary

4,763 CHARS

**Background:** Nonalcoholic fatty liver disease (NAFLD) is a chronic liver disease affecting multiple cell types, with up to one-fourth of the global population presenting with some degree of NAFLD. Despite well-understood pathomechanisms in early NAFLD development, there is currently no approved treatment. Significant inter-species differences have led to lost-in-translation cases where therapeutic candidates validated in animals did not effectively ameliorate NAFLD in clinical trials. This review evaluates human-specific in vitro modeling strategies, their promises, progress, and limitations.

**Methods:** The authors review four major human-specific in vitro platforms: (1) human induced pluripotent stem cell (hiPSC)-based platforms, (2) precision-cut liver slices (PCLSs), (3) primary tissue-derived intrahepatic cholangiocyte organoids (ICOs), and (4) perfused liver-on-chip models. They assess each platform through a lens of how effectively they model clinical hallmarks of the NAFLD spectrum, including steatosis, lipotoxicity, oxidative stress, endoplasmic reticulum stress, mitochondrial dysfunction, inflammatory response, hepatic stellate cell activation, and fibrosis. The review covers methodological overviews, implementation hurdles (reproducibility, costs), and suitability for specific research applications.

**Key Results:** For hiPSC-based platforms, directed differentiation protocols produce hepatocyte-like cells expressing fetal markers (alpha-fetoprotein) with low adult cytochrome P450 (CyP3A4) expression, described as 'fetal-like' cells. Forward programming offers less time-consuming and more robust protocols but still requires specialized culture media. Kumar et al. developed a 3D co-culture system combining hepatoblast-like cells, HSCs, macrophages, and endothelial cells in polyethylene glycol hydrogel, demonstrating that oleic acid treatment resulted in lipid accumulation, increased collagen and alpha-SMA expression, and surge in proinflammatory cytokines (interleukin-6, tumor necrosis factor-alpha). The PPARα/δ activator elafibranor showed positive effects on fibrosis in vitro. Takebe's group developed human liver organoids (HLOs) via hiPSC co-differentiation into endodermal and mesodermal progenitors, which can model steatosis and proinflammatory response after oleic acid treatment. HLOs were tested under oleic acid and insulin-high growth conditions to link the GCKR gene risk allele to mitochondrial activity. For PCLSs, viability may be maintained up to 6 hours in simple culture conditions or up to a few days in controlled conditions. One study claimed maintenance for more than 2 weeks, though this requires further validation. Liver slicing activates inflammatory responses during the first 4 days of culture, and experiments should be conducted after 7 days when the system stabilizes. For ICOs, McCarron et al. studied organoids from NASH-related cirrhosis patients (n=6) versus controls (n=3). NASH organoids displayed 4.5 times higher oleic acid-induced lipid uptake compared with non-NASH controls, while low-density lipoprotein accumulation was about 2-fold reduced. NASH cirrhosis organoids replicated faster but underwent growth arrest earlier, limiting expandability. NASH cirrhosis-derived organoids were also more susceptible to infection with SARS-CoV-2 pseudovirus. For liver-on-chip models, Raasch et al. developed a perfusable biochip with two chambers separated by a porous membrane. Another microfluidic NASH-on-a-chip model using primary human hepatocytes, KCs, HSCs, and LSECs in collagen-based hydrogel demonstrated that inflammatory cytokine secretion increased upon lipotoxic conditions and was reduced by elafibranor. Chips seeded with PNPLA3 mutant cells had increased NASH phenotypes (inflammation, lipotoxicity, fibrogenesis). The system remained viable for at least 2 weeks.

**Clinical Implications:** The review emphasizes that no single platform recapitulates all nuances of NAFLD/NASH pathobiology. HiPSC-based platforms allow scalable, parallel differentiation of multiple cell types but at the expense of cellular maturity and liver tissue-specific context. PCLSs preserve spatial tissue architecture for a limited period but are not scalable. ICOs help understand epithelial injury and hepatocyte-biliary cell plasticity but lack multicellular context. Liver-on-chip models address multicellularity with dedicated cell compartments but face challenges in deriving primary human liver cells. The metabolic and immune crosstalk between liver, muscle, fat, and other tissues is only rudimentarily captured by current models. The authors conclude that results generated in vitro need to be validated using animal models, patient data, and ultimately clinical trials.