PepGen Inc. ($PEPG)

Welcome to the PepGen conference call to discuss top line results from lowest dose MAD cohort in the Phase II FREEDOM2 study, demonstrating promising safety, splicing and vHOT data. [Operator Instructions] As a reminder, this call is being recorded today, Monday, March 30, 2026. I would now like to hand the conference call over to Laurence Watts of New Street Investor Relations. Please go ahead.

Thank you, operator. Joining us on the call today from PepGen are President and Chief Executive Officer, James McArthur; Executive Vice President, Head of Research and Development, Paul Streck; and Chief Financial Officer, Noel Donnelly. During this call, management will make forward-looking statements, including statements related to its Phase II trial of PGN-EDODM1 as well as the timing of additional or future data from the studies and the company's development plans. Our actual results and the timing of events could differ materially from those anticipated in such forward-looking statements as a result of risks and uncertainties. Factors that could cause these results to be different from these statements include factors the company describes in its securities filings, including its annual report on Form 10-K and our quarterly reports on Form 10-Q. PepGen undertakes no duty or obligation to update any forward-looking statements as a result of new information, future events or changes in its expectations. With that, I will turn the call over to James McArthur, President and Chief Executive Officer of PepGen. James?

Thank you, Laurence. Good afternoon, everyone. I will initially provide a high-level summary of the results we will review today as well as the mechanism of action of EDODM1. I'm pleased to report that at 5 mg/kg, the lowest dose of the FREEDOM2 multiple ascending dose study, we've observed a promising emerging safety profile and are encouraged by the splicing and vHOT data it generated. The data gives us confidence as we continue to dose the next cohort of 10 mg/kg of the FREEDOM2 multiple ascending dose study. In terms of safety, at 5 mg/kg of PGN-EDODM1, all adverse events were mild or moderate with no serious adverse events, no treatment-related adverse events related to the kidney and no cumulative dose toxicity through the 4 doses. The mean splicing correction observed in the active group was 7.3% versus 6.8% in the placebo group. However, there was one individual splicing outlier that when excluded from the group analysis resulted in a mean splicing correction of 22.9% in the EDODM1 treatment group. We also observed promising trends in hand myotonia as measured using the middle finger video hand opening time or vHOT measurement in the PGN-EDODM1 treatment group. I'm also pleased to report that recruitment for the 10 mg/kg cohort is proceeding well, and we expect to report clinical data from this next cohort in the second half of 2026. To remind everyone, in myotonic dystrophy type 1, 2 DMPK RNAs are produced. One, the normal transcript present in the cytoplasm and the nucleus indicated here schematically with straight lines does not include an expanded CUG repeat and is not pathogenic. A second pathogenic transcript present in the nucleus has an expanded CUG repeat that forms stem loop structures, which trap MBNL1 protein, a critical protein required for the splicing of multiple transcripts. PGN-EDODM1 is designed specifically to bind to the pathogenic DMPK RNA with the expanded CUG repeat that drives the pathology of DM1. We have demonstrated in patient cells that EDODM1 can reduce toxic RNA foci, liberate MBNL1 protein and improve RNA mis-splicing that is the hallmark of this disease. Last year, we presented the results of the FREEDOM Phase I single ascending dose study of PGN-EDODM1, where we demonstrated a dose-dependent increase in correction of mis-splicing. We are now pleased to present to you the first results from the FREEDOM2 multiple ascending dose study with PGN-EDODM1. I'll now hand over the presentation to my colleague, Dr. Paul Streck, EVP and Head of R&D, to take you through our clinical results.

Thank you, James. Here, you can see the schematic of our FREEDOM2 Phase II multiple ascending dose study. Today, we will take you through the results from the 5 mg/kg starting dose cohort and provide you an update with our progress in the 10-mg cohort, which we are currently enrolling. As you can see, the FREEDOM2 clinical study is designed as a 3-cohort 4-dose study where patients received 4 monthly doses. Each group has 8 patients randomized 3:1 active to placebo. As we look at the demographics of the 5 mg/kg MAD cohort, a couple of things to note. First, the splicing index for the group was in the moderate range for myotonic dystrophy with a mean of 69.2. Second, the baseline hand myotonia or mean vHOT at baseline for the group based on the time to extend the middle finger was 10.3 seconds, lower than we had in our Phase I FREEDOM placebo and 5-mg cohorts. As we look at the blinded safety data for the cohort, we are pleased to see that PGN-EDODM1 has a favorable emerging safety profile following 4 doses at 5 mg/kg. Importantly, we did not observe any cumulative increase in safety signals with multiple doses. Further, all treatment-emergent adverse events were mild or moderate in nature. Nausea was the most common AE and did not get worse with multiple doses. There were no serious adverse events or notable adverse events and no dose-limiting toxicities. In addition, there were no signs of hypersensitivity following drug administration. And importantly, there were no kidney-related treatment-emergent adverse events. eGFR and creatinine measurements remained within the normal range and transient albuminuria was observed in 2 subjects, however, with no cumulative signal after 4 doses. In summary, this safety profile supports our continued advancement of EDODM1 at 10 mg/kg. Turning to Slide 13. To assess splicing correction, tibialis anterior muscle needle biopsy samples were taken approximately 7 days following the fourth dose. Splicing levels were compared in these samples to TA muscle biopsies that were taken prior to dosing. In the 5 mg/kg cohort, mean splicing correction in the treatment group improved 7.3% compared to mean splicing correction of 6.8% in the placebo group. One of the treatment cohort patients was a notable outlier, demonstrating a 70.8% worsening of splicing. Note that this level of response was not observed in our 28-day Phase I study and is a notable outlier based on conversations we have had with third-party experts. When this splicing outlier is removed and we analyze the 5 remaining DM1 patient treatment samples, the mean splicing correction increased to 22.9%, 7 days following last dose compared to mean splicing correction of 6.8% in the placebo group. Along with this, we observed high mean TA muscle concentrations of EDODM1 of 158 ng/g in the day 7 biopsies from the treatment group. Now let me turn to functional outcome measures. Myotonia measured by middle finger vHOT or video hand opening time, demonstrated improvements in the EDODM1 treatment group compared to a worsening in the placebo group. This occurred from week 4 through week 13. vHOT improved up to 4.1 seconds through week 13 in the analysis of the 6 EDODM1 treatment group patients before returning to baseline at week 16. Moreover, in the vHOT analysis, excluding the splicing outlier I highlighted earlier, vHOT improved up to [ 3.4 seconds ] at all time points through week 16. With that, let me hand the call back over to James.

Thank you, Paul. In conclusion, we're very pleased by these results today. At our starting dose of 5 mg/kg, PGN-EDODM1 has demonstrated promising safety splicing and vHOT data, which gives us confidence for the ongoing FREEDOM2 study in the 10 mg/kg cohort. Moreover, EDODM1 was well tolerated in the MAD study with all adverse events being mild to moderate with no serious adverse events and no evidence of cumulative toxicity. In addition, 12 patients have progressed into the open-label extension at 5 mg/kg, including 5 patients from the FREEDOM2 multiple ascending dose study. From a target engagement standpoint, we observed promising splicing correction with EDODM1 at 5 mg/kg, our starting dose cohort, particularly when the splicing outlier is excluded. We also observed vHOT improvements in the treatment group, suggesting that higher dosing may drive even further improvements. As of today, the company has now dosed 5 of 8 patients in the 10 mg/kg multiple ascending dose cohort of the FREEDOM2 clinical study with subjects having received up to 3 doses of EDODM1. As such, we look forward to reporting the results from the 10 mg/kg cohort in the second half of this year. Lastly, for our last guidance, our cash runway extends into the second half of 2027. In accordance with our current plans, we believe this is sufficient to see us through reporting the 10 mg/kg FREEDOM2 data later this year and results from the third cohort of 12.5 mg/kg in 2027. With that, we're happy to take questions. Operator?

[Operator Instructions] Our first question comes from the line of Konstantinos Biliouris of Oppenheimer.

Congrats on the update. A couple of questions from us. The first one is, what does the consideration, the drug consideration look like of the outlier here who had 70.8% relapse in the splicing?

Yes. So that particular patient did clearly have drug on board. In fact, it was relatively significant levels of drug on board. But we do not understand why that particular patient did not see an improvement in splicing.

And maybe a follow-up on the placebo 1 patient who also seems to have a very good response. Any thoughts on what may these have been attributed to?

In an analysis that has been done by experts in the field, you can certainly go and see movement in terms of 20% or so, both in terms of negative and positive improvement in splicing that occurs. This doesn't appear to be variability in the splicing assay. It appears to be simply movement in the patients over time. In our Phase I study, we did have in the placebo group, some patients who showed as much as a 30% worsening or 25% or so improvement. They do tend to be not as common as the sort of smaller movement that you see around the baseline measurement, however.

Our next question comes from the line of Joseph P. Schwartz of Leerink Partners.

I was wondering, first, is there any mechanistic or clinical rationale for why the outlier patient in the drug arm behaved differently in terms of splicing? And was this biomarker change associated with any clinical changes?

So unfortunately, at this point, we remain blinded to, for instance, some of the underlying baseline characteristics of this patient. We don't know, for instance, the size or the number of repeats the patient had or if there were other characteristics that might distinguish them from other patients, and we won't become unblinded until the end of the study. In terms of why they didn't respond, as I mentioned earlier, we did note that they had reasonable levels of drug in the tissue sample that we took. But clearly, we saw a significant worsening in the response in this one individual. Based on our prior experience in the Phase I data over the course of 28 days, we did not see this kind of change in the placebo group. And in our conversations with experts in the field, they've noted that the likelihood of seeing this kind of change is very rare, low single-digit percentage in terms of the samples that they've looked at. So unfortunately, at this point in time, we cannot ascribe a particular reason for why the patient responded in this particular way. It will be something we look at going forward. I will note that this is our starting dose, and we anticipate as we move forward that we'll be able to improve upon these splicing results in the 10 and 12.5 mg/kg cohorts.

Okay. And then how should we view the post-hoc metric of average splicing index improvement of 23% in the drug arm, excluding the outlier patient, considering one of the placebo patients had their splicing index improved by 27%.

Absolutely. So the placebo group has only 2 individuals in it right now. As we add additional placebo patients, we anticipate that the placebo group will move closer to baseline as was the case in our Phase I study. And then in terms of what we would expect going forward for the active group, we did note that 5 of the individual -- sorry, 4 of the 6 individuals in the active group saw higher levels of splicing than what we observed in the Phase I study at the 5 mg/kg dose level. So we expect to be able to build upon that as well.

Our next question comes from the line of Paul Matteis of Stifel.

I was wondering if you can comment on what you're seeing as it relates to the safety profile of the 10 mg/kg dose on a blinded basis and your level of confidence that you can mitigate side effects there and have a wide enough TI.

My colleague, Paul will take that one.

Yes, Paul, we feel quite good about what we saw in terms of movement. As I mentioned, we didn't have a significant amount of AEs, and we saw no evidence of any cumulative effect, i.e. any renal lab parameters getting increasingly worse with subsequent dosing. So -- and we didn't really see any other movement outside of the albuminuria, which is quite sensitive for anything going on. I think most importantly is that what we saw is we take labs are drawn at baseline 24 hours post dosing, then 7 days post dose. The albuminuria was present at 24 hours and then returned to baseline by day 7. And like I say, each dose appears to behave as its own discrete event as opposed to something that's building upon previous events.

And Paul, I was asking specifically about the 10 mg/kg dose coming up. Are you talking about 10? Or are you speaking about 5?

Speaking about 5 at this point, Paul. Like I say, it'd be premature for me to comment on blinded data at 10. Suffice to say, if there was anything material that was happening, we would certainly share that.

And I think one of the sort of things to underline, Paul, is that as we expected, we have not seen cumulative toxicity with any of the signals in the 5 mg/kg, which is what we had seen in nonhuman primate preclinical toxicology studies as well. So as we project forward, we would anticipate seeing something similar.

And just to clarify, James, when you guys talk about safety and your comfort, what is the line in your mind? Like I guess, said another way, is the target product profile here, no hypomag at all, only mild? How do you think about drawing the line for an acceptable risk benefit?

Well, I'll let Paul speak to what would be acceptable. But I do want to underline, we have seen no hypomagnesemia in the Phase I study nor have we seen this in this study as well. Paul?

Yes. I think it's well said, James. I would suggest that anything that is transient and reversible, whether it's reversible with magnesium supplementation, I think most importantly, I want to remind folks, bad disease, lots of progression of clinical symptoms. So anything that we're seeing that is mild in nature and able to be managed without any evidence of long-term effects, I think, would be a very adequate long-term safety profile.

Yes. And sorry, just slipping this in. Any time lines you can provide on trying to lift the FDA hold?

So we continue to engage with the FDA, and we believe we'll be able to address the questions that they raised. But given the way things are right now with the FDA, we really can't provide a time line, I'm afraid.

Our next question comes from the line of Laura Chico of Wedbush.

I guess just one on efficacy. I know we have vHOT data included here, but what are the expectations for releasing additional functional endpoints such as 10-minute walk/run test. I know you mentioned, James, folks are entering the open-label portion of the study now. And then just with respect to safety, I apologize. I just wanted to clarify again. Could you actually quantify the albuminuria increase? I guess I'm trying to understand if there were transient elevations occurring in that 10 mg/kg cohort, I guess I need to understand what would actually warrant a disclosure ahead of the data release?

So in terms of a disclosure, if we had to materially change the study stop dosing, we had a hold from a regulatory agency regarding the data, something along those lines would be something we would disclose. But as Paul highlighted, what we have seen so far has been very mild in nature, not requiring any intervention, has been transient, self-contained, not getting worse over the progression of the 4 doses. And I'll note that of the sites that have the open-label extension open and available to them, patients are embracing that and moving forward and moving into the open-label extension, both patients from the FREEDOM study as well as from the FREEDOM2 study. And then to get to your activity or functional question, so we did not see meaningful movement in terms of 10-meter walk/run test at 5 mg/kg. Keep in mind, this is our starting dose. So if you look back at the data of other folks, you'll see that the level of vHOT improvement that we're seeing here is, I think, quite noteworthy. And we're pleased to go and see that indeed, it persisted over the 3-plus months that we're looking before at the last time point, both placebo and active across the baseline. So we're pleased with the vHOT signal. We're not surprised we're not getting movement yet in other outcome measures, given this is our starting dose at 5 mg/kg. We are looking at a range of other functional endpoints as well. But again, at this point, it doesn't make sense, I think, for us to report that data until, a, we have more placebo patients in those groups; and b, we've explored higher dose ranges as well. So we're excited by what we're seeing at the starting dose level and looking forward to reporting at the 10 mg/kg data later this year.

Our next question comes from the line of Debjit Chattopadhyay of Guggenheim.

I have a question on the vHOT observations, especially between week 13 and week 16. Why did the placebo drop to baseline? And why does the treatment arm recede back to baseline that quickly towards the end of the study?

Yes. Thanks for the question, Debjit. So we don't know if this is just variability of the assay itself. But like I said earlier, we're just pleased to note that for over 3-plus months, we see nice separation of both the placebo versus the active group. It was interesting to note that one individual who was a splicing outlier also had the greatest variability, I will say, in the vHOT assay as well. So when you separate that individual out, you can see the bars continue to be separated, although to your point, they both move towards baseline. I think with higher dose levels and in particular, adding to the placebo group, this group contains only 2 patients, we may begin to start seeing more defined changes over time between these 2 groups.

So as a follow-up, for the 10 mg/kg cohort, when you do have the data, how many placebo patients do you expect to have in that cohort?

So all of the cohorts are 3:1 randomization. And so there will be 2 placebo patients there, which we'll be able to add into the 2 placebo patients here. So at that time, we will have 4 placebo patients to go and compare to the 6 5 mg/kg and presumably 6 10 mg/kg patients, both in terms of the splicing analyses as well as all of the functional outcome measures.

Got it. And one last follow-up here. Just with regards to your splicing assay, could the splicing assay be off for the whole study? I mean the question is, do you believe the outlier patient splicing is a real result? Or is it a mistake?

Yes, it's a great question. And obviously, when we first saw the result, that's where our minds went. I will say the team here did a terrific job in terms of both making certain that there was no confusion in terms of mixing of tubes or mislabeling of samples, confirming the fidelity of the RNA extraction and the running of the assay and making certain that the calculations where they go and compare splicing at the end of study to the beginning study were done correctly. We don't believe there were any errors made in that analysis. So that leaves us with this 70% worsening in terms of the response. There's not a tremendous amount of data out there, unfortunately, in the public domain. We have from private communications with pioneers in the field heard that this type of event can occur if you look at 100-plus patients in a couple of patients. So it's a low single-digit percentage type event over the length of this kind of study. So it can happen in reality. And what we don't know right now is, is there something in particular that defines this patient as being different than other patients who either were stabilized or where we saw a response. And that's something which when we unblind at the end of the study, we'll be able to dig into more fulsomely.

If I could follow up with one more. From a therapeutic index perspective, how confident are you you can get to 10 mgs and if needed, go towards 12.5?

Yes. So the recruitment and dosing for the 10 mg/kg, as we highlighted today, is proceeding well. We have 5 of 8 patients dosed with up to 3 doses, and we continue to drive that forward with confidence. So based on our expectations, we will be dosing the 10 mg/kg cohort and moving to a third cohort of 12.5 mg/kg, and we'll be able to do that and build upon the results we have here.

Our next question comes from the line of Ananda Ghosh of H.C. Wainwright & Company.

So one question on -- as you were saying that you will be adding the placebos from the earlier studies as the trial goes on, what kind of statistical analysis are you doing in order to compare those cumulative placebo additions with respect to the active group? Is it -- that would be helpful to understand how you are comparing the 2 arms.

And I'm sorry, Ananda, you cut out for a second there. If you could repeat your question for me?

So as you mentioned that you are -- the placebos will be added as the trial goes on, what kind of statistical analysis you are performing in order to compare the placebo with the active group, let's say, when you are at the 10 mg/kg cohort?

Thank you. So this study is not powered to go and generate statistically significant data. It's really both a safety and dose-finding study as we go and explore doses up to 12.5 mg/kg. As we and others have done, we will go and take these patients and basically group them together in the placebo group as we move forward. We're doing our best to make certain that, number one, of course, they're randomized and we have the same inclusion/exclusion criteria, both across the placebos as well as across the actives as we move through Cohort 1, 2 and 3. So that's why we believe that we'll be able to pull these patients together as we did in our Phase I clinical study, FREEDOM.

Got it. And maybe a follow-up question was, what was the dose concentration across the patients -- number of patients you had for the 5 mg/kg? And based on your modeling, where do you expect the concentration might be at the 10 mg/kg?

It's a great question. Unfortunately, right now, we just don't have enough information to project forward. We do expect to be able to build upon these tissue levels. also to be able to build upon the splicing we've seen. And we're certainly hoping that we do not see another outlier like we saw in this particular group, which definitely skewed the group analysis, which we've provided.

I would now like to turn the conference back over to James McArthur for closing remarks. Sir?

Thank you for joining our call today. We'd like to thank, most importantly, all the patients and investigators who are involved with the ongoing studies and very much look forward to informing and updating you all on our future progress. Thank you very much, everyone.

This concludes today's conference call. Thank you for participating. You may now disconnect.