All right, thanks everyone for joining on day one of the Oppenheimer Healthcare Conference. My name is Frank Brisbois. I'm one of the biotech analysts at Oppenheimer. Our next company presenting here is Dyne Therapeutics from the company. We have CEO John Cox and CMO Doug Kerr to present. In terms of format, what we'll do is a fireside chat.
Feel free to send questions in the Q&A tab. If not, you can email me but with that, thank you, guys. I know it's been a busy new year, so I appreciate you guys taking the time to do this here and maybe if we could just start, I think a lot of people are familiar, but just a quick overview of the company for those that might not be.
Yeah, thanks, Frank. Good to be here. Happy to do that. So I think, as you know well, this company was founded a little over five years ago. And it was founded on a belief that there are various genetic medicines, oligonucleotides, that could be used to treat diseases, neuromuscular diseases like Duchenne muscular dystrophy, like myotonic dystrophy type 1 . It wasn't a question of whether the oligonucleotides could work so much.
It was about delivery. Could you get enough oligonucleotide payload to the billions of cells and tissues that you had to get to to have an effect? And Dyne had created a platform, which we call the FORCE platform, which was really all about using an antibody fragment to hit a particular epitope on the transferrin receptor, which you find across virtually all muscle cells.
You find it across the blood-brain barrier as well, because the transferrin receptor transfers iron into cells. It's designed for transport. And we wanted to co-opt that receptor just for intermittent periods of time to transfer in particular oligonucleotides for these diseases. And the question was, if you did that and you were able to do that, could you start seeing broad effect, functional impact, and in a reasonable timeframe?
The experiment's been run, Frank. I mean, as you saw the latest data from us that we presented in an event in early January and then presented again at JPMorgan, we had most recently completed a multi ascending dose study, which we call ACHIEVE for DM1. It's a massive, it's a rare disease, but it is a big market, roughly 100,000 patients across the U.S. and Europe. And there's nothing for these patients.
So we did a multi ascending dose study in which we were trying to determine if you got the right oligonucleotide, in this case, an antisense oligonucleotide, to the nucleus of a broad set of tissues that are affected by this disease. Could you get to the or address the fundamental pathobiology, which is a splicing disease or splicing correction that needs to happen? And so we went through the exercise of measuring splicing as we increased the dose of our drug, and we indeed saw splicing correction as we went up in dose.
We also started to see that we were unlocking broad functional benefit. And that really opened up to us, I think uniquely to us, what we call an accelerated approval pathway. And I'll just touch on that for a second, because I think it's important people understand that you're seeing accelerated approval and full approval.
For accelerated approval, the expectation with the FDA is that you have a biomarker that's robust and that is reasonably predictive of functional benefit that will matter to patients. It's not a full approval with full stat significance, but you have to be able to show that your biomarker does hit as a primary endpoint statistically and that you're seeing correlations with functional benefits that matter to patients. So the bar is different than full approval, and you can accelerate.
There was never a biomarker established for DM1. We have demonstrated that the natural history data around splicing can be used as a biomarker to show improvement. And I think we've accomplished that, and now we're in a position to pursue, and I don't think anybody else is in a position to do it, pursue the accelerated approval pathway.
At the same time, the expectation is if the FDA approves you based upon a biomarker and accelerated approval, that you support that with a full approval. And we'll have a full approval protocol with that as well, which is what most people do, but the accelerated is available to us.
And I'll just wrap up by saying at the same time, we have another program for DMD in which we followed a proven accelerated approval pathway in which we are measuring dystrophin. And that data to date, what we had shown was quite profound, potentially showing for the first time functional benefit in a small number of patients and profound levels of a biomarker called dystrophin.
That program is also under an accelerated approval pathway with a registrational cohort that is nearly enrolled, and we're moving forward aggressively to try and get both of these programs on the market roughly in the 2027 timeframe. I'll stop there.
Yeah, no, that's a great overview. And so January comes around, the data comes out, people have a little bit of time after to digest it. Then you come out of JPM, you add some standard deviations. Basically, you add some statistical information to help people feel more comfortable with the data. You're consistent across delivery, DMPK, splicing, functional benefit. You go for a higher dose at 6.8, which you didn't even need to because 5.4 was good enough.
You see CNS and PROs and MDHI, you see good stuff that tells you, regardless of that outlier or not, you're going with the 6.8, which means you clearly like 6.8 more than what was good enough at 5.4. So to me, that seems successful. So to cut right to the chase of it, maybe it's been a long day, what are people missing here? What are people really struggling?
Is it the biotech market has been really tough? It's hard for me to believe people see a cash overhang that's substantial based on your position. So you guys have had time to think about it. I don't know how much fun it was to think about it too much, but what do we think people are missing here?
Well, first of all, you were on a roll there, Frank. I was going to just sit back, man. I loved it. Everything was perfect.
You nailed it.
You nailed it. So listen, I think the investors, when we first presented it, people had never seen an accelerated approval approach for DM1. They've seen it for DMD, and we're not getting questions about that. But people hadn't seen it for DM1. And I think people had thought because the disease is fairly slowly progressing and so heterogeneous that you might need hundreds of patients in a registrational trial. And then the question was, well, how robust is this splicing index?
The analyte is called a CASI, as you know. How robust is it? And if you're going to use it as a surrogate endpoint, which is your primary endpoint, and you're using small patient numbers, 32 up to say 48, is it robust enough? And I think people just naturally had not seen that before, and they questioned it. And so that's why, as you mentioned, after we kind of presented that data and we got the questions, we basically went back to the JPMorgan presentation. And I gave a presentation, but Doug really helped.
And our biostats team is part of Doug's organization said, let's give people things like our power and calculations. Let's give people the standard deviation around splicing. Let's talk about what's meaningful in terms of correlation and so on. Let's provide them the stats, essentially, that they need to go do their own determination of whether or not 40 plus patients is sufficient.
And then we also said, you know what, guys, we need to be really clear about what an accelerated approval package is going to look like in terms of the number of participants. I mentioned to you, Frank, that we had the multi ascending dose portion. That portion does not disappear.
There's 56 subjects in that portion of the study that now get dosed up to the 6.8 dose that you talked about, plus another 40, say 48, but 40 plus patients or subjects in the registrational cohort. Now we're talking about a 100-subject trial moving forward with a surrogate endpoint plus meaningful data and correlations on all of these other functional endpoints that would be part of the package.
So I think we just had to really lay it out for people. It's not a simple story, but it's a deep story. And I think it's a robust one. And as people are getting their arms around it, they get more and more comfortable with how differentiated and unique it is to us.
It does seem like a tough market because I have had a lot of investors come back and say, well, this is a drug, right? Is there anything here that came out that tells you there's no drug here? And absolutely not. So on that standard, I saw that standard error on the splicing at the JPM presentation. How does that work on the blind? So you still don't have the individuals, just your biostats guys that kind of like you guys are still blinded, right? Is the trial sort of.
The 6.8 mg/kg cohort is still blinded so that there's an external party that does just this, so knows which patients in the 6.8 mg/kg got drug and which patients got placebo. All of the other cohorts are completely unblinded. What happens now over the next several weeks or so is we unblind the 6.8 mg/kg fully so we can start to kind of look deeper.
That will be the basis of a regulatory engagement this year. So now you've got the totality of the data, 56 patients, various doses, open- label extension, long-term extension, up dose to 6.8. And that's the basis of an engagement with regulators in the U.S. and ex-U.S. for a couple of reasons, really, for further feedback on the accelerated approval pathway.
Now, we were with them and discussed this at length and received this kind of very clear roadmap of what we would need to do and show to get accelerated approval. This reflects that roadmap. We are executing on that roadmap. But one of the overhangs, I think, when we came out with the data earlier in January was that we hadn't actually gone before them explicitly with this protocol. Well, of course not. We got it over the holidays and really made the decision.
We've got the roadmap. We know what to do. It's time to go. It's time to get these patients enrolled and move as quickly as we can. And as John showed, this wasn't a shortcut. Once people got under the hood and looked at the statistics, this is a very, very robust package for accelerated approval. But this year, we will go get regulatory feedback.
They will get all of the data unblinded. We'll talk about the accelerated approval pathway. Most importantly, we're going to talk about the phase III study, which will begin this year and will be roughly 200 patients, but we'll get explicit feedback on what that is. The importance of that is that the phase III will get us some of the functional aspects and strength aspects and other things that we can't get in the registration expansion cohort. It just takes too many patients and too long a period of time. But also that that trial will be up and running when we go for accelerated approval.
If there is a request from regulators for more safety data or for even some efficacy data, there is the possibility to support this registration expansion cohort with the ongoing phaseIII in order to increase the likelihood of that accelerated approval while still being able to get the confirmatory full approval based on the phase III .
No, that's great. And it's interesting because it's such a market that's looking to nitpick right now that I bet you you've gotten the question of why not go to a higher dose from 6.8. And it used to be so clear, 20%-25% splicing is all you need. So why would you go higher? Let's get going here, sort of thing. And so it's interesting to me where the problem is even if you had gone to a higher dose and your splicing stayed the same, but your function actually was better. Well, then you lose the correlation between splicing and function. And so that could get nitpicked too. So it's a tough one to please, but.
And the other important point on that, Frank, is just that between 5.4 and 6.8, obviously, we did the muscle biopsy. We did the functional outcomes. We did the strength. They were better. The strength and function were better at 6.8, but they're not that far apart. But most importantly, and we had predicted this going in, that above some dose, you will access the more difficult tissues that are really important in myotonic dystrophy.
And that below that dose, kind of the muscle acts as an on-target sink. All of the drug is taken up into the biggest organ in the body, muscle, not leaving any drug left for other tissues like the CNS. And that may be what we're seeing. And it's such an important part of myotonic dystrophy that when you see the 6.8, now with a 44% reduction in MDHI, that's pretty profound.
We want to get that. Now, these were small numbers, but we want to get it. It made us really feel like not only are you fully safe at all doses, including 6.8, you got better strength and functional measures at 6.8, especially when you normalize for baseline imbalances, and you get CNS. It's time to go. Let's get this to patients.
Yep. Did you unblind? So your 5.4 is totally unblinded, but have you guys thought about sharing the standard errors on the splicing for 5.4? Or is that just that you haven't shared that? I'll put it that way, right?
Yeah, I don't know that we've shared it. Yeah, I mean, I think at some point we will. We'll gather all of those data from ACHIEVE. I mean, there's no reason not to.
Yeah. Okay. Awesome. And then I think I want to, obviously, I didn't want to start with DMD because everyone's been talking about DM1 recently here, but there's a lot going on on DMD. Before we move on to that, is there anything else that we kind of jumped over? It's been top of mind for a lot of people, so we kind of went right into it, but anything else on the DM1 side that you guys would like to discuss?
I mean, maybe the only one thing, Frank, would be that we've gone back and looked at the rigor of CASI itself, so I think some people kind of viewed that. Is it too variable? Is it methodologically challenging such that you will lose data or have discordant data, and the answer to that is no.
We've gone back and looked at the variance in this, the robustness of this, the number of samples that we lose to kind of QC reasons like RNA degradation. The answer to that is like 4% of samples, and I mean, that is really, really robust, really rigorous. There are not many biomarkers that are that robust, and so let's not spend time talking about these few outliers that we talked about, but they are few, like four out of 150, really, really good robustness.
Going forward in this registration expansion cohort, we've established analytical methods to kind of increase the rigor of that, such as imputation for missing data, ways to handle data that is a bit discordant. All of that was not in place. It now is. So we walk away saying, you know, look, it's really good that we did this very deep assessment of CASI. We came away feeling like, oh my gosh, it is so good. It captures the pathologic heterogeneity of myotonic dystrophy, and it says exactly what we wanted to say, and we can do this in a very robust, reproducible way.
On that note, can you explain to people what that 0.4 to 0.6 predictability of CASI means and why that's so important?
It was part of the roadmap that we got from the FDA, which said, yes, this accelerated approval pathway is open to you. CASI is a very good biomarker that captures the disease heterogeneity, but you must show statistical significance on it, A, B, that it predicts functional change, which is why we showed that at JPMorgan, and three, you have to show directionality on something that matters to a patient, not myotonia, must be strength or function or a PRO, and so we've gotten all of those, which is, that is, the roadmap that we got, so when you think about a surrogate endpoint for accelerated approval, it must be reasonably likely to predict function.
We've shown that with very good R- values, taking a CASI at one time period and looking at function as an offset three months later. Provenzano, by the way, did that in the JCI paper that came out exactly the same thing. We had nothing to do with that. So independent demonstrations that CASI does predict function further supporting the accelerated approval pathway.
Perfect, and then I guess the last thing on my side, you guys haven't shared timing on this end-of-phase II with the FDA. Is that fair?
We have not yet. Clearly, getting in front of the FDA with the kind of data package that Doug's talking about, we want to do that as urgently as we can. There's a process, and the FDA dictates a bit of timing around that, but we will make that happen as fast as possible.
This is a little wild, not to get political, but any discussions about changes politically and the FDA's speed or anything like that, or has it changed nothing for you guys?
So I don't think we've seen anything. I know people have some concerns. There's only so many things we have control over. And one we have control over is to put a good package together. I know there were some questions about, I think, with the FDA and with CBER and potential of people leaving. We're working with CDER on this. We've got really clear direction from them of what's required for the accelerated approval path. So I think we'll keep our nose to the grindstone and execute.
Awesome. All right, DMD, what's going on there and why is the data? Why people seem to be forgetting a little bit about this data?
Yeah, I think there's a bit of underestimation about the potential with DMD. And you were asking a minute ago, really, about what to think about with DM1, and I think with DMD as well. I mentioned at the beginning, both of these programs have the potential to be on the market commercially in 2027. The dialogue of kind of CASI scoring and all this other stuff, and also the question about even dystrophin, everything's moving to function and ultimately to a label and to launch.
And we'll be getting there in the not-too-distant future. And that's how we think about the company. So DMD is, man, if you look at DMD, the accelerated approval pathway has been done. It's been demonstrated and with questionable functional benefit. You look at our data at 20 mg/kg, and I know you've seen it.
Doug can go on this in particular, and that is the Stride Velocity 95th Centile. It's not effort-dependent. It's an accelerometer over a period of, what, a couple of weeks, Doug, that they're measuring with these boys. Our data had shown a clinically meaningful difference in improvement, not just relative to placebo or decline, improvement off baseline, so now that was small patient numbers, but we've got 50-plus patients moving forward at that 20 mg/kg.
The registrational cohort is nearly fully enrolled and will be this quarter, and then you've got a dystrophin endpoint of six months from now, and our dystrophin levels as a biomarker are so far behind what anybody has seen with exon 51, and then potential of some kind of functional benefit.
I think if you talk to clinicians, whether it's DM1 or DMD, or you talk to patients, it's like you said, there's no question about in people's minds that we have a couple of drugs here. And there is incredible enthusiasm that the world needs a next-generation exon skipper in the DMD space. There's a lot of noise, not noise, just a lot of confusion because there's gene therapy and microdystrophin and all this stuff. I don't think there's any confusion in the community. You need an exon, a next-generation exon skipper, and we are really, I think, in the pole position to have the next drug to be there.
That's great. And in terms of different exons, how are you guys thinking about that?
Yeah, I mean, we've got a number of other candidates lined up. I'm preclinical, and I love the idea of us thinking about this as a franchise. I mean, instead of just thinking exon by exon, the reality is our platform, if you read the FDA guidance on platform clinical trials, you would think it's like ready-made for what we do.
We're using the same Fab, the same linker, and essentially a few nucleotide changes on the payload. So if we can figure out some ways with the FDA, let's see this exon 51 data hold up, do well with function, and then let's start talking about novel regulatory pathways or clinical pathways. The opportunity for us to expand this into a franchise, very exciting.
And it's de-risking across the board, whether it's DM1, Pompe, or the FSHD.
No, exactly. As you know, we've got Pompe, preclinical, and that one is actually a recombinant protein from beginning to end. It's an enzyme attached again to our Fab, and the preclinical data was amazing. I mean, the preclinical data was showing glycogen clearance, not just in the heart like you see now with certain drugs, heart, various muscles, and the brain. And then we've also got similar work going on with FSHD. So the platform's profile across humans as well as in animals with different payloads, very, very unique.
It's tough because it's still a relatively small company, but is anything paused right now, or are we still full steam ahead on everything?
No, we're full steam ahead. I mean, clearly, the priority around DM1, moving that forward, DMD, we're talking about the accelerated approval pathways for both. Both of them, we want to move forward with the phase III. But the pipeline, we love what we have in the pipeline.
Awesome. And then just lastly, because it seems like everyone asked this question at the end here, what's the balance sheet looking like?
Well, I mean, we reported over $600 million in cash and that we have enough cash to take us out to the second half of 2026, so keep in mind, we fully own the assets. We'll see how the data is shaken out through the year and how we're doing, and we'll evaluate opportunities, but we're well funded as a company in very good condition.
Excellent. Well, there's a little bit of a venting session on my end at the start. I've thought about this data quite a bit, not as much as you guys. But what should I have asked? Anything we missed here just to clear the air for investors?
I think the fact that you pointed out the sense that we have based upon functional data to two medicines that really look differentiated is kind of a, I mean, we can't lose sight of that. When we look at the functional data, I get caught up on if we have what the stats look like, and I really love the splicing work we're doing, and I love the fact that we've established biomarker in DM1, and we're following a proven path with DMD.
I love all of that. At the end of the day, as I said at the beginning, our platform was designed to get broad distribution of medicines, genetic medicines in this case, oligonucleotides, so that you could see functional benefit sooner and more broadly and deeper than you would otherwise expect. The data we've been seeing, Frank, as you know, is in six months. These are slowly progressing diseases, and we see functional benefit, small data sets of functional benefit in six months. It's really profound.
Yeah. Excellent. Well, I think we're coming up on the time here, but thank you very much. I know how busy you guys are. I appreciate the time.
Feel better, buddy.
Feel better.
All right. Thanks, guys.
See you.
See you. Thank you.