NeuroVoices: Ahmet Hoke, MD, PhD, on Why Animal Models Still Have a Role in Neuropathy Research

The rapid advancement of induced pluripotent stem cell technology has prompted a growing conversation in the neuromuscular community about whether animal models remain necessary in neuropathy research. iPSC-derived neurons now allow researchers to model disease using human cells, conduct mechanistic studies, and screen drugs in ways that were not previously possible, leading some regulatory authorities and scientists to question the continued reliance on animal systems. The debate came to the forefront at the 2026 Peripheral Nerve Society (PNS) Annual Meeting in Maastricht, where the question was posed directly: should animal models still have a place in neuropathy research?

Ahmet Hoke, MD, PhD, is a Professor of Neurology and Neuroscience at Johns Hopkins University School of Medicine and Director of both the Neuromuscular Division and the Merkin Perfect Neuropathy and Nerve Regeneration Center. At the meeting, Hoke represented the case for preserving animal models in the research pipeline, arguing that the three-dimensional complexity they offer cannot yet be replicated by in vitro systems alone.

In this iteration of NeuroVoices, Hoke explained why he believes it is too early to move away from animal models, what the data around iPSC systems actually shows, and what the downstream consequences of a full transition away from animal research could mean for the clinical community. He also addressed the promise and limitations of organoids and shares where he thinks research efforts need to be focused to bring more clarity to this ongoing debate.

NeurologyLive: Why is the use of animal models in neuropathy research an evolving debate in the clinical community right now?

Ahmet Hoke, MD, PhD: I think the main reason we decided to have this debate is that the use of induced pluripotent stem cells has really changed how we do translational research. We can now take patients' own cells, whether blood or skin samples, convert them into induced stem cells, and then convert those into neurons that allow you to model the disease using human cells. You can study disease mechanisms, better understand protein dysfunction, and use these cells for drug screening purposes. That is really pushing some regulatory authorities, as well as scientists, to consider abandoning the use of animal models before going into clinical trials. I will argue that it is too early to make that jump. Animal models still have a role to play in both understanding disease mechanisms and in the drug development pipeline. They bring a three-dimensional complexity and interaction with other cell types and tissues that you simply cannot replicate in a dish with stem cell-derived models.

What has the data actually shown that has evolved this conversation, and how has that shaped people's perceptions on both sides?

The iPSC models have evolved to the point where we can now reproducibly grow them and use them for drug screening and mechanistic studies. That is one of the reasons people started thinking of them as a true human cell system. I remember being at a meeting years ago where a clinician said about a mouse neuropathy model, "humans are not mice." That is true, but even though mouse cells may behave slightly differently or express certain receptors or proteins differently, they are still 98% similar to humans.

The big thing we would lose if we jumped directly from iPSC drug screening to clinical trials is the ability to see whether drugs actually work in a complex organism. Nerve cells are not in isolation. Even the nerves themselves are surrounded by Schwann cells, macrophages, and fibroblasts, encased in a much more complex three-dimensional structure that in vitro systems cannot replicate. There are also interactions with immune cells and blood flow that matter. If you are testing a drug, you need to see how those other cells are affected before putting it into a clinical trial.

If the field were to transition more fully toward iPSC models, what downstream impacts do you think that would have on the clinical community, both positive and negative?

In defense of the iPSC work, one of the real advances has been organoids. Instead of converting iPSC cells into a single cell type, you can create organoids, essentially mini organs in a dish, grown in a three-dimensional structure. You get small structures that resemble a brain or spinal cord, and in our field, researchers have started making complex spinal cord and muscle mixtures where you have neuromuscular junctions, where some cells become muscles and can be innervated.

The big problem in that area, at least for the neuromuscular system, is reproducibility. More progress has been made in the CNS, but not as much in the neuromuscular system, because of the added complexity. Every cell type in the brain shares the same embryonic developmental origin, but in the neuromuscular system, motor neurons derive from the central nervous system, Schwann cells that support the axons come from a different origin, and muscle is very different again. That brings a level of complexity to organoids that we have not yet fully solved.

Where do you think research efforts need to be focused so that we can have clearer answers about which model works best, or about how to incorporate both in a more balanced way going forward?

The future is definitely incorporating both models into our studies, both for understanding disease mechanisms and for drug development. Now that I have seen how well iPSC lines can work, if you do a drug screen with rat cells, I would want to see that replicated in a human cell system before committing the resources to a mouse, rat, or larger animal study, depending on the clinical indication. The development pipeline needs to incorporate both.

The other important point is that our mouse and rat models work best when, for genetic diseases like different forms of Charcot-Marie-Tooth for example, instead of simply knocking out a gene, you actually introduce the mutated gene into the mice. Knock-in models are much better than knockout models for replicating human disease. They are not always perfect either, and lifespan and size remain challenges with mice, but those are issues that exist in dish-based systems as well.

Transcript was edited for clarity. Click here for more PNS 2026 coverage.