New insights show “later is better” for some LHON interventions

Neuro-ophthalmologist Alfredo Sadun, MD, PhD, has helped uncover surprising results from gene therapy research for Leber hereditary optic neuropathy (LHON), revealing effects far beyond the initially treated eye. LHON is a rare genetic disorder that causes sudden, severe vision loss in young adults. Sadun holds the Flora L. Thornton Endowed Chair in Vision Research, is chief of ophthalmology at Doheny Eye Institute, University of California, Los Angeles (UCLA); and is a vice chair and professor of ophthalmology at the David Geffen School of Medicine at UCLA.

Using an adeno-associated virus type 2 (AAV2) viral vector to deliver corrective genes into 1 eye, researchers unexpectedly observed improvement in both eyes.¹ Postmortem analysis confirmed bilateral gene presence, challenging assumptions about how gene therapy spreads and functions.

These findings not only reshape clinicians’ understanding of vector behavior but also suggest broader implications for treating other inherited retinal diseases.

Sadun shared highlights of the research in this interview with the Eye Care Network.

Note: The below Q&A has been lightly edited for clarity.

Can you walk us through the unexpected findings from this research for LHON?

Alfredo Sadun, MD, PhD: I’m a neuro-ophthalmologist with a great interest in 1 particular disease, which is LHON. This is a genetic disease where you do fine until you turn 20 [years old] or so, and then you suddenly go blind—first in one [eye] and then the other eye—very dramatically and very unfortunately. We’ve been doing research together with GenSight Biologics from Paris, France, for gene therapy for this disease. In doing so, we have found that, at least in some patients, an injection of the gene via AAV2 intravitreally can lead to some improvement.

This research both succeeded and failed enormously because of the same phenomenon. An injection in 1 eye somehow mysteriously caused improvement of vision in both eyes. This was a failure in the sense that the FDA said, ‘You get approval if you can show the injected eye does much better than the other eye.’ But they both got much better, so they are not on [track] for approval as of right now.

The amazing thing is that they got more bang for [their] buck: Both eyes got better. The agency, GenSight, thought this [was] because the virus went into the brain and came back down via the other optic nerve. I have a PhD in ocular anatomy, and [I] said that didn’t sound very likely.

We’ve been doing these investigations where…not 1 but 2 patients who had undergone gene therapy died. We had an opportunity [to analyze] the eye. In looking for products of the virus and of the transfer gene, we were able to find that it was in the other eye. In this sense, the company was correct.

But the concentrations of the virus did not get weaker. As you went back from the uninvolved eye, they got greater, meaning it got to the eye and then diffused back to the brain—not that it got to the injected eye, to the brain, and diffused forward into the eye. We’re still not sure how it got from one eye to the other, but it does not seem to be via the brain.

Do these bilateral results from a unilateral injection indicate that we’ve misunderstood something about gene therapy as an approach for inherited retinal diseases?

Sadun: No. 1 is that we have misunderstood the directions of these vectors and how the infection leads to transfer. In LHON, not only does this tell us a lot about treatment of that disease, but it [also] tells us about the AAV2 virus being a different story than we thought, which will apply to Leber congenital amaurosis [LCA] and other diseases that might be using gene therapy for treatment.

It’s going to be a fantastic opportunity when we finally understand how it works, but there are a lot of limitations. In LCA, they found that it works well in the short term, but there’s some decay of the effect. Perhaps [this is] because we’re putting in a good variant of the gene, but the old variant is still around, and they’re in competition with each other. Somehow, when you have competition, usually the native protein wins. There are a lot of other issues at play here.

What can you tell us about the ages of the patients and how the age ranges of the cohorts impacted treatment outcomes?

Sadun: LHON tends to affect people in their late teens and mid-20s. It’s very sad because people [had] plans [for] their lives that have been interrupted enormously, shockingly, and quickly. Most of our patients for the gene therapy research were in their early 20s. We only looked at patients who had lost vision within the [past] year, hoping that it was still salvageable because the longer you wait after that, the optic atrophy would have put a ceiling on how much improvement we could have gotten. That’s the nature of the disease, the nature of our experiments, [and] the nature of when the gene therapy would work. But there is something that comes out of that, which is [also] unexpected.

We initially started with 2 different groups: zero to 6 months after loss of vision and 6 to 12 months. The 6- to 12-month group was plan B. Our team was thinking [we were] going to miss some of these patients because the diagnosis doesn’t occur that quickly, [as] there are not too many neuro-ophthalmologists around.

We expected the results to be much better in the [group with] 0 to 6 months [since their diagnosis] because the optic atrophy had not really developed. There were more salvageable cells. To our shock, the patients in the 6- to 12-month group did better than the 0- to 6-[month] group. Furthermore, when we did a subgroup analysis among the 0- to 6-[month] group, the only ones who did well were those who lost vision 5 or 6 months before the injection. Among the 6- to 12-[month] group, the patients who did best were the ones around 11 to 12 months.

Later is better. That’s [so] shocking that we went back and looked at the literature for a medical treatment with an agent called idebenone among the same patients. It’s not published yet, but I’m working on a paper that’s entitled “Later is Better.”

What are the next steps to better understand the bilateral results of this therapy?

Sadun: What we’re looking at [next] is other tissues. We’re looking at vascular tissues, lymphatic tissues—it’s just a wild guess. I don’t mean to suggest that I have a deeper insight than anyone else, but I think that maybe the glymphatic [system], the fluid, [or] the cerebrospinal fluid [CSF] around the brain…might have been the conduit for this. In which case, you would expect it to sort of mix throughout the brain to concentrate in areas that are the cul-de-sac of the CSF flow. The biggest cul-de-sac is the optic nerve head, just at the juncture of the optic nerve in the retina. That gradient we saw, which was highest there and going down further back, would make sense.

Reference

1. Carelli V, Newman N, Biousse V, et al. New insights on AAV2-ND4 transfer between eyes: ocular post-mortem analyses in Leber hereditary optic neuropathy following AAV2 gene therapy. Presented at: Association for Research in Vision and Ophthalmology 2025 Annual Meeting; May 4-8, 2025; Salt Lake City, UT.