What is it about?
Many medicines already approved for one illness might also work for another. Computers and AI can now suggest thousands of such matches — far more than anyone could ever test in patients. The hard part is deciding which suggestions are real. This paper argues that organoids — small, simplified versions of human organs grown in the lab, often from a patient’s own cells — can serve as the missing middle step. Used in the right way, with the success criteria agreed before the experiment starts, they can show which candidate drugs deserve a clinical trial and which should be set aside. The paper describes how to run that step, and which rare diseases it suits best, using ataxia-telangiectasia — a childhood neurological disease — as the worked example.
Featured Image
Photo by National Institute of Allergy and Infectious Diseases on Unsplash
Why is it important?
Rare diseases affect few patients each, so trials are difficult to run and there is little commercial reason to attempt them. That makes it essential to get the preclinical decision right the first time. Animal models often cannot deliver this: the mouse version of ataxia-telangiectasia never develops the progressive loss of balance and coordination that defines the disease in children. The timing matters too. In April 2025 the FDA announced a plan to reduce reliance on animal testing, and in March 2026 it published draft guidance on human-relevant New Approach Methodologies; the EMA is moving in the same direction. This paper offers a concrete proposal for how organoid evidence can be used, and judged, inside those new frameworks.
Perspectives
I have worked with organoids long enough to be wary of overselling them. The argument here is deliberately narrow: an organoid result is decision-grade only for a defined question, in a defined model, against criteria set in advance. That restraint is precisely what makes the evidence usable by a regulator. What pushed me to write the paper was the growing mismatch between how quickly computational candidates now arrive and how little we can honestly say about them before someone has to commit money — and patients.
Massimo Gulisano
Universita degli Studi di Catania
Read the Original
This page is a summary of: Bridging the validation gap: organoids as functional decision systems in drug repurposing, Pharmacological Research, September 2026, Elsevier,
DOI: 10.1016/j.phrs.2026.108360.
You can read the full text:
Contributors
The following have contributed to this page
