Dyne Therapeutics, Inc. Common Stock 2026 Global Healthcare Conference

Yeah, happy to, Brian. Good to be here. Thanks for hosting us. Listen, I think Dyne was really built on the notion that the delivery of genetic medicines, antisense oligonucleotides, in neuromuscular diseases needed to be done in a way that would broadly and fully distribute the oligonucleotides to tissue and across the blood-brain barrier, and to do so in a way that would be considered a safe way to do that. The intentional design of our platform, and it was very intentional, was that we would use the targeting of the transferrin receptor 1, which others have used, which naturally transports iron into cells. And we would target that particular receptor, not with a monoclonal antibody like most, but with a monovalent Fab.

The idea behind that was that what people had seen with mAbs is that you tended to block that transferrin receptor, and you would be limited by anemia. And we always had a concern that we wanted to avoid any cross-linking of multiple transferrin receptors, particularly around the blood-brain barrier. And so we intentionally designed a Fab that would allow us to deliver to the CNS, broadly to tissue, a much smaller molecule than you, about a third of the size of a mAb, so that we could dose escalate in a way without seeing anemia and deliver high quantities of genetic medicines. Everybody knew the genetic basis for things like DMD and DM1. The challenge was being able to deliver enough of the payload.

The scientists at Dyne, this is before I came, but this is the reason I came, is we started to actually see pharmacology and ability to dose higher, that you started to see functional improvement. And last year, in 2025, we had actually, I think, validated this platform in two really tough diseases, DMD and DM1. In both of those diseases, across multiple functional measures, we didn't just show slowing of decline, which is what most people would hope to see at best. We actually saw functional improvement from baseline and from placebo, both in DMD and in DM1. So this year has been all about execution, and we are hoping to launch our first product next year.

Yeah, I am happy to do that. Listen, when we saw the announcement recently that they had failed to hit on their primary endpoint, the video hand opening time for hand myotonia, we were surprised because that is a particular endpoint in DM1 that is, it moves early, it moves easily, and it moves with very little biological activity, meaning drug activity. So we had anticipated that they would hit on that. We have gone back and really taken a look at it, and I think what we should first start with is the fundamental biology. Foundationally, just foundationally, we have a very different molecule, and really, I think in two ways. One is the payload, and one is the distribution. If I start with the payload, that is really important because I think you know, Brian, with this disease, with DM1, the target has to be the nucleus.

The target, more specifically, is the mutant DMPK that is causing the mis-splicing that cannot get out of the nucleus. Our payload is an antisense oligonucleotide because that payload gets to the nucleus. The other drug that you are referring to with Novartis is an siRNA. siRNA acts in the cytoplasm. So getting the first principle of making sure you are getting enough drug to the target to get sufficient splicing is key. That is one point, and we can talk a little bit about the trial differences in a second. The second point is that it is not just getting to the target, but it is about enough payload to the target to drive enough of what we call splicing correction, and there is a way to measure that. It is called a CASI analytic. We measure that in every single patient.

When we did our dosing at a very low dose, the first dose in the multi-centering dose study, 1.8 mg per kg, we saw vHOT moving. It moves with very little splicing correction. But we did not stop. We kept dosing higher. We went all the way up to 6.8 mg per kg. vHOT was taken care of. But then we started seeing potentially CNS effects. We started seeing broad functional improvement across multiple measures, and that is where we selected our dose. So that distribution and what I described before with the Fab, us being able to go to a higher dose without the safety limitations that are referred to, allowed us, we believe, to get a very different level of targeting and amount of material to the nucleus, unlike the other product.

Now, when we go back and just look at some of the early work, I really would like to, I hope Novartis releases some of this data so we can all learn. Maybe there were just some outliers, things we can learn about vHOT. But if you go back to their MARINA study, at their low dose, their first dose, 2 mg per kg, they did not see much in terms of vHOT. They went to 4 mg, and that is when they saw a good vHOT response. But that was dosed every six weeks. Then they went into the phase III, spreading out the dose to every eight weeks. If you were sort of on the cusp of having enough, and their splicing numbers were never very high, unlike what we had seen. Fundamentally, you have got to get splicing. That is where the truth lies.

If they were borderline on that, which would make sense if they are not getting in the nucleus, you would expect with an endpoint like vHOT, which has variability, that you have got to be definitive about getting enough material there.

I remember you making that point before.