The Business of Biofabrication

A recent study by Quris Technologies highlights how drug metabolism in the liver can significantly alter a compound’s neurotoxicity, providing new insights into limitations of traditional drug testing methods. Published in Biomedicine & Pharmacotherapy, the findings suggest a greater role for multi-organ, human-relevant testing platforms in preclinical development.

Using a proprietary platform called BioAI, Quris researchers connected human-derived liver spheroids and brain organoids to simulate drug interactions across organ systems. Their investigation focused on troglitazone, a diabetes drug withdrawn from the market due to hepatotoxicity. In the BioAI model, the team observed that liver-metabolized troglitazone induced substantially higher levels of brain cell toxicity than direct exposure to the drug alone.

Specifically, the study found that liver-processed troglitazone altered expression in 3,941 genes within brain organoids, compared to 685 genes in a brain-only model — a nearly sevenfold difference. It also affected ten times more biological pathways, particularly those related to neural growth and differentiation.

The 3D liver spheroids used in the study consisted of approximately 2,000 human primary liver cells sourced from ten donors, while the brain organoids were derived from human-induced pluripotent stem cells (iPSCs) and matured over four weeks. The organoids demonstrated functional characteristics, including electrical activity and marker expression.

Notably, the researchers found that not all drugs followed this pattern. Valproic acid, a commonly used antiepileptic, showed reduced toxicity after liver metabolism at high doses — illustrating the complexity of drug-organ interactions.

These results align with ongoing efforts to implement non-animal New Approach Methodologies (NAMs) in preclinical research, including those supported by the FDA Modernization Act 2.0, which encourages the use of more human-relevant models in drug safety testing.

Limitations
The study acknowledged several limitations:

• The absence of a blood-brain barrier component
• Lack of microfluidic flow between liver and brain compartments
• Potential variability due to limited compound selection (six drugs tested)
• Partial cell death observed in liver spheroids during experimentation

The authors noted that incorporating a functional blood-brain barrier and flow-based system would further enhance model fidelity.

Funding and Conflict Disclosure
The research was conducted and funded by Quris Technologies. All authors are affiliated with the company and may hold equity. Additional support was provided by InSphero AG, Axion Biosystems, Azenta Life Sciences, Tharkka Corporation, and microscopy cores supported by NIH grants.

About Quris Technologies
Quris Technologies is a biotechnology company headquartered in Boston and Tel Aviv, specializing in AI-driven drug development platforms that predict clinical safety and efficacy without relying on animal models. Its flagship platform, BioAI, employs patient-on-a-chip systems that combine machine learning with human-derived tissue models. The company has raised a total of $37 million in seed funding, with investments from SoftBank Vision Fund 2, Welltech Ventures, iAngels, GlenRock Capital, and others.

Citation
Wu W., Anas F., Koc B., et al. “Multi-organ model assessment of neurotoxicity following exposure of liver spheroids to drugs.” Biomedicine & Pharmacotherapy, Vol. 186, April 2025, Article 118021. Open access under CC BY-NC-ND license.