DiaMedica Therapeutics Inc. (DMAC) Earnings Call Transcript & Summary

Thank you, Operator. Everyone, good morning, and I appreciate you all being on the call for our third-quarter update. Dr. Lorianne Masuoka, who serves as our Chief Medical Officer, is here with me, along with Scott Kellen, our Chief Financial Officer. Progress across both the stroke and pre-eclampsia initiatives has been substantial, and we see 2025 as a truly transformative year ahead — for the company and for those who hold our shares. Getting our leading U.S. locations up and running in the AIS trial required more time than we had anticipated, yet I'm pleased to report that among our top 15 sites, 13 have either completed activation or already put clinical trial agreements in place. When it comes to getting a site up and running, the contract negotiation phase is usually what eats up the most time. For these sites — especially the five biggest ones — we're expecting them to bring in a share of patients that's far larger than their numbers would suggest. Houston Methodist was the first among our top five to get activated, going live back in September. Two others we've flagged as top-five priorities are Chattanooga, Tennessee, and Oregon Health. Both have contracts in place, and we've already scheduled their site initiation visits. Our goal is for every one of those top five U.S. sites to be activated and enrolling patients before the year is out. On top of that, Canada will bring on five to seven additional sites, and we'd expect their enrollment pace to be on par with what we see in the U.S. We're starting with the priority locations, and the earliest ones should be up and running prior to the end of the year. Lorianne will walk you through this in greater depth, but alongside launching these tier-one sites, we're rolling out an update to the protocol that's meant to speed up enrollment further and focus on high-quality patients who have serious unmet medical needs; we factored in what these sites told us and introduced a number of modest but meaningful adjustments. The biggest one: we've switched to a freshly manufactured batch of drug that can be kept refrigerated for our investigational product. In the past, it had to be frozen, which meant an investigational pharmacist had to handle dispensing it, and those pharmacists typically only work Monday through Friday, 9:00 A.M. to 5:00 P.M. When the drug is kept in refrigerated vials, our thinking is that emergency department storage will let sites sign up patients during weekends and after normal hours. Great progress is also being made with our pre-eclampsia program; we're happy that approval from the South African Health Products Regulatory Authority came through so quickly, and only yesterday the first pregnant participant who has pre-eclampsia was enrolled; in our view DM199 could potentially modify the disease in pre-eclampsia, and we're hoping to show that during part one A of the study, with an announcement coming in the first half of 2025. For pre-eclampsia, we see DM199 as having the potential to be both first and best-in-class, and sadly, no therapies are approved in the U.S. at present. At this point, I'd like to hand things off to Lorianne.

Thanks, Rick. To begin today, I want to walk through the latest changes made to the ReMEDy2 protocol along with the statistical analysis plan. For those of you listening online, I'll be pointing to slides that are up on your screen right now. Those same slides were submitted to the SEC in a Form 8-K last night as well. I came on board at DiaMedica in late January of this year, and my enthusiasm for this opportunity is enormous; it became obvious very quickly that a transformation of the medical organization was necessary and that we had to move clinical site-facing work back in-house to DiaMedica, since the results coming from our contract research organization simply were not acceptable. Our progress so far makes me very happy, and I'll go into more specifics on that in just a moment. While that transition is ongoing, the next thing we turned our attention to was the design of the trial and the statistical analysis plan. If you turn to slide two in the deck, you'll see that we have put in place a set of updates and additional clarifications to the ReMEDy2 protocol, and in our view these are quite favorable for the trial. We sent these updates to the FDA at the end of August, and since no comments had come back as of yesterday, we are now putting the new protocol modifications into effect. There are two main updates worth covering: first, including patients who failed to respond to thrombolytic therapy — meaning treatment with thrombolytics, which are Activase or Tenecteplase — and second, raising the interim analysis sample size from 144 up to 200 participants, which greatly strengthens the power of the Bayesian simulation that forecasts the total sample size required, and as a result sharpens the precision of, and confidence in, the final determination of sample size for the trial. I'll go through each one in detail, but understand that these changes will assist us in reaching the outcomes below. First, they accelerate enrollment rates at each site; when we previewed these changes with the clinical sites, their response was that by including thrombolytic non-responders, they could potentially boost their enrollment considerably. A number of sites mentioned figures as high as 50%-100%. Second, adding thrombolytic non-responders should improve our overall response rate relative to placebo. In our ReMEDy1 phase II study, this patient group showed the highest response rate of any subgroup. I'll say more about this, but when we reanalyzed our data, we wanted to bring in these potentially excellent patients in order to raise the probability of success for the ReMEDy2 trial. Third, for trials that use an adaptive design structure, the interim analysis must rest on a participant count large enough to optimize how precisely the estimated final sample size is determined, so that we avoid enrolling an unnecessarily excessive number of participants. Taken together, we expect the protocol updates to be very favorable for our study sites, they should raise the probability of success for the ReMEDy2 trial, and they should speed up our timeline to completion. Reaching the interim enrollment could require another 4 to 6 months because of this modification. Even so, what matters is that a larger sample may allow the eventual sample to be smaller, possibly shortening the whole study duration and delivering meaningful savings in cost. Let me state without any ambiguity: had anyone believed or worried that these adjustments would bring about even the tiniest decline in the trial's quality or efficacy, we would never have implemented them. I stress this point because over my 30 years of developing drugs, I have seen many trials collapse after inclusion criteria were widened in an effort to accelerate enrollment. Moving to slide 3, I'll provide further detail. Looking at what our ReMEDy1 phase II trial data showed, we determined that most subjects who had previously received tPA were given DM199 toward the end of the 24-hour treatment window. What's striking is that while enrollment for such patients could have happened merely 1 hour following tPA administration, they were enrolled, on average, 13 and a half hours afterward. That matters because it is widely accepted that thrombolytics possess a very brief half-life — roughly 24 minutes — so the body clears them quickly; such agents complete 7 half-lives within under 3 hours and are essentially gone from a patient's system 3 hours after being dosed. Consequently, when someone shows up with stroke severity that remains moderate to severe a number of hours after tPA has been cleared from their system, that signals the thrombolytics failed to meaningfully fix the neurological deficit. In everyday terms, we call such a patient a tPA or thrombolytic non-responder. Capturing this kind of patient is our aim, since we expect they will exhibit a low placebo response rate, and compared with DM199 therapy there is potential for a sizable treatment effect. Over on the slide's right side, the data from our ReMEDy1 study involving participants who got tPA before enrollment is displayed. In the placebo group, as you can observe, the response rate stood at 0%, which in our view confirms these participants were tPA non-responders. By contrast, the response rate in the DM199 arm reached 25%. Among every subgroup examined in the ReMEDy1 trial, this 25% improvement over placebo was in fact the largest performance improvement. I'd also like to point out that the main reason tPA non-responders weren't part of the ReMEDy2 study at the outset was the considerable outcome disparity depending on how quickly a patient receives tPA after a stroke; patients given tPA at 1 hour post-stroke have been shown to fare far better than those given tPA at 4 hours post-stroke. Back then, DiaMedica was unwilling to bring that potential variability into the study, nor the clinical risk of a TPA patient imbalance favoring the placebo arm. Yet after weighing the ReMEDy1 results, I think we can remove this variability risk provided we define a thrombolytic non-responder correctly. Now, moving on to slide 4, I'll walk you through what the inclusion criteria specify for patients who don't respond to TPA. Our scientific advisory board, along with top vascular neurologists who serve as consultants to DiaMedica, have put in a tremendous number of hours to make sure these criteria — the ones you see on this slide — are defined appropriately. Following our conversations with the scientific advisory board and the stroke neurologists involved in our trial, our view is that the optimal moment to determine non-responder status comes 6 hours after the thrombolytic has been given. Should you step in sooner — say, at 4 hours — the chance rises that the patient is simply a delayed responder. On the other hand, waiting until the 12-hour point creates a danger that additional brain tissue will die because blood flow is compromised. Our judgment is that 6 hours offers the proper balance. Keep in mind that during our earlier ReMEDy1 trial, DM199 was given, on average, 13 and a half hours following TPA. If we're able to reach patients faster than that, there's an opportunity to achieve better treatment effects than those we observed previously in the ReMEDy1 study. The second thing I'd like to stress is that once that 6-hour threshold arrives — or at any point afterward — the patient has to undergo reevaluation. The neurological deficit needs to be ongoing and must sit in the same stroke severity band as that of non-TPA patients, which means an NIHSS score falling between 5 and 15. By ongoing neurological deficit, what we mean is that the patient's NIHSS score cannot have shown an improvement of 4 points or greater after receiving the thrombolytic.

Patients who remain inside the NIHSS range yet are on a path toward recovery are not the ones we're seeking. And last, the patient needs a repeat scan to exclude deterioration stemming from any hemorrhagic transformation possibly brought on by the TPA, plus the patient has to satisfy every remaining requirement just as non-TPA patients do — for instance, receiving treatment inside that same 24-hour window from the onset of stroke symptoms. With this added criterion, our view is that we've achieved the proper balance for enrolling a patient population facing a substantial unmet need, one we expect will react well to DM199. Moving to slide five, the second major update concerns how large the interim analysis will be. Across the last 20 years, I've collaborated closely with a number of statistical analysis leaders on my studies, and I had them examine our statistical analysis plan. Given the input received and the possible shrinking of the trial's total sample size, we chose to conduct the interim analysis once 200 patients have been enrolled — up from the 144 participants previously planned. Slide 6 offers a strong illustration of why being underpowered at the interim analysis is something we want to avoid. What you're seeing is a genuine simulation from the model, comparing final sample size projections when 144 versus 200 patients are enrolled at interim. To protect trial integrity, we cannot reveal the actual treatment response and placebo response, yet in the example shown, those values match exactly, with the interim sample size being the sole variable. In this scenario, as you can observe, with 144 participants the reestimated sample size comes to 485 participants, while at 200 it comes to 339 participants. Let me stress again that this is a genuine simulation within the model that drives our statistical analysis plan. Looking at it from a cost perspective, a gap like this might mean savings for DiaMedica of $10 million or more; our belief is that this change could speed up completion of the entire study, despite the fact that it will postpone when we receive our interim analysis results. Hopefully, the impact on the timing of our interim analysis will be partly offset by widening our inclusion criteria and adding thrombolytic non-responder patients. Before, we had expected our interim analysis to be ready in the summer of 2025; at present we anticipate it in quarter four 2025. We believe our financing is sufficient to carry us through the interim analysis. Moving to slide six, let me restate that we strongly believe these protocol updates will boost enrollment rates by adding a subgroup of stroke patients likely to be high responders to DM199 therapy, while also enhancing the potential commercial value of DM199. Moreover, with the revised statistical analysis plan, we have the greatest potential for clinical success using the smallest study feasible. I'll now hand the call back to Rick.

Appreciate that, Lorianne. When it comes to ReMEDy2, our attention stays fixed on keeping the drive going with top-tier research organizations. To bring you up to speed, the bulk of the 15 priority study sites we identified have now been switched on as of today. At this point, I'll pass things along to Scott Kellen so he can walk through the financial results for this quarter.

Appreciate that, Rick, and hello everybody, good morning — glad you could be with us on the call today. Let's begin with our financial position: on September 30, 2024, the total of our cash, cash equivalents, and investments stood at $50.2 million, a decline from the $52.9 million figure recorded on December 31, 2023. Operating activities consumed $15.6 million in net cash across the nine months that ended September 30, 2024, compared with $14.9 million in the equivalent stretch of the prior year. The main drivers behind the higher cash consumed by operating activities were a larger net loss along with prepayments made to vendors for the ReMEDy2 clinical trial in the current year period. Those increases were partly offset by shifts in operating assets and liabilities during the same period, and we believe our existing cash and investments give us enough runway to reach Q3 of 2026. R&D expenses came to $5 million for the three months ended September 30, 2024, compared with $3.3 million a year earlier. For the nine-month stretch that closed on September 30, 2024, research and development spending came in at $12.6 million, up from $9.4 million during the same period ending September 30, 2023. The main drivers of that higher figure were rising costs tied to keeping our ReMEDy2 clinical trial going, growing our clinical team, and stepping up manufacturing development work. Working in the other direction, we saw savings from clinical trial efforts that wrapped up in 2023, such as our Phase 1C and REDUX trials, along with the 2023 finish of the in-use study work that had been undertaken to resolve the earlier clinical hold placed on the ReMEDy2 trial. We anticipate a moderate rise in R&D costs compared with recent earlier periods, driven by expanding ReMEDy2 trial site activations worldwide and ongoing participant enrollment. For both the three months concluded September 30, 2024, and the same period in 2023, our general and administrative expenses totaled $1.9 million. Over the nine months concluded September 30, 2024, G&A expenses came to $5.7 million, a decline from $6 million during the nine months concluded September 30, 2023. That nine-month decrease stemmed from combining lower premiums for directors and officers' liability insurance with reduced legal fees tied to our litigation against PRA Netherlands. Higher personnel expenses tied to growing our workforce, along with greater non-cash share-based compensation costs, helped counterbalance part of those reductions. Looking ahead, we anticipate that G&A expenses will hold at levels consistent with what we've seen in recent prior periods. For the three months concluded September 30, 2024, other income net came to $616,000, while the nine-month figure for the same period was $1.7 million. That compares against $693,000 and $1.2 million, respectively, for the three and nine-month periods that ended September 30, 2023. For the comparison covering nine months, the gain came from more interest income being recorded this year, which stemmed from larger balances in marketable securities than we had during the equivalent stretch a year earlier. Having said that, I'll turn it over to the operator to open the lines for questions.

Morning, and thanks for fielding these. Lorianne, staying on ReMEDy2, what first triggered the decision to revisit the protocol along with the statistical analysis plan? Was higher enrollment the only reason, or did something else set off that whole conversation that produced these modifications?

Honestly, it came from two things at once — we wanted to drive enrollment, and we'd also held a lot of talks with our SAB plus the KOL experts we collaborate with; they were genuinely eager to find out how patients who got thrombolytic therapy yet didn't respond would fare. We consulted statistical specialists, and their view was that having no fewer than 50% of patients enrolled mattered before running a top-quality Bayesian sample size adjustment; that's why we applied the adjustment, drawing on substantial statistical guidance.

Understood. Then here's a question that's maybe a bit technical. When exactly do patients get randomized at this stage, especially the ones who turn out to be TPA failures? The reason I ask is that you'd want those patients spread evenly between the two arms, and you can't actually tell whether they're failures when randomization happens before they get TPA. So could you take us through how that works in practice, step by step?

Sure. When a patient gets randomized, that patient does not get TPA. That patient goes into our study through randomization. If TPA is given before the patient is randomized into our study, then a wait of six hours is required before we can determine whether that patient is a non-responder and therefore eligible to enter our study.

Understood. Then, if I could, one last item. According to the initial plan you laid out, those 144 patients were meant to serve as the interim cohort in a study of 364 patients. And I realize you just walked us through a simulation example, but I believe that one involved 350 patients with 200 as the interim analysis. So does 364 remain a workable figure, or what's the right way to interpret this? Since those two things appear to conflict. Yeah.

Well, with 200 patients at the interim analysis, and assuming the same drug effect as before at the 144, we'd expect a smaller overall patient count, meaning under 364.

All right. Understood. I appreciate your taking the question. Thank you.

Morning. Thanks for fielding the questions. So, to kick things off, just so I'm clear on this. Could you give me a bit more detail on the sorts of assumptions that lead to a faster final readout given these modifications — again, with a larger number of patients at the interim? I believe you may have already addressed part of this, but is it simply that with greater certainty about the effect you're observing, you're able to resample with fewer patients, or is it more about the higher enrollment rate you'd anticipate by permitting TPA non-responders? Thanks.

Yes. So essentially, what happens with Bayesian analysis is that the more patients you include at the interim, the greater the precision you achieve in evaluating the sample size. For instance, assuming a 14% excellent outcome rate, which is what we had initially determined back when we had 364 patients, we might be able to reduce that to 300 with an interim look at 200; and clearly, that would mean substantial savings in money and substantial savings in the time needed for enrollment. So essentially, moving to 200 allows us to potentially cut a significant amount of time from the trial and to operate far more efficiently regarding the dollars we expend on the trial.

Understood. It's possible you already addressed this — I've been bouncing between a few calls, so I apologize. Did anything stand out regarding enrollment rates or the general baseline characteristics of patient care that played a part in driving these adjustments? Or was it mostly tied to that second point you just raised? Thank you.

Yes. It isn't something observed within the study as it was actually running. The truth is, when we were putting the finishing touches on the report for the Bayesian analysis accompanying the SAP being submitted to the FDA, we arrived at these modifications after consulting our statistical experts. So expert opinion was truly the foundation.

And Chase, I can add; so, part of our analysis we conducted with these TPA patients that did not respond from our phase 2 trial, there was a very low placebo response rate, so a zero, whereas we had a 25% improvement with our drug; and so if we carry that into our current trial, that should greatly help the statistical analysis plan. So if we have a large number of patients that have a very low placebo but a strong drug effect, it should be very beneficial for our statistical analysis plan.

In the earlier trial, what number of patients were the TPA non-responders, and what was that N?

The figure comes to 20 in total. Beyond that, I would mention that a fairly thorough review of how human urinary form gets used in China at present formed a piece of what we examined. Roughly six clinical studies have demonstrated an added benefit from urinary KLK1 when given alongside TPA. That finding also lines up with what we've heard in conversations with several distinct parties across China — the very people who are currently treating patients in that country.