Fractyl Health, Inc. (GUTS) Earnings Call Transcript & Summary

Greetings, and welcome to Fractyl Health's 1-year REMAIN-1 Midpoint Cohort Data Call. [Operator Instructions] As a reminder, this call is being recorded, and a replay will be made available on the Fractyl Health website following the conclusion of the event. I'd now like to turn the call over to Brian Luque, Head of Investor Relations and Corporate Development at Fractyl Health. Please go ahead, Brian.

Thank you. Welcome, everybody. This morning, we issued a press release that outlines the topics we plan to discuss today. The press release is available at www.fractyl.com under the Investors tab. Presenting today will be Dr. Harith Rajagopalan, Co-Founder and CEO of Fractyl Health, with Dr. Adarsh Thaker from UCLA joining to share his perspectives after the formal presentation. During this call, we make forward-looking statements, which involve risks and uncertainties that may cause our actual results to differ materially from those expressed or implied by forward-looking statements. A discussion of these risks and uncertainties is included in our filings with the SEC from time to time, including the section titled Risk Factors in our annual form -- on our annual report on Form 10-K for the year ended December 31, 2025, and our quarterly report on Form 10-Q filed on May 12, 2026, which I encourage you to review. Any forward-looking statements on the call are subject to substantial risks and uncertainties, speak only as of the call's original date, and we undertake no obligation to update or revise any of these statements even if subsequent events cause our views to change. It is now my pleasure to pass the call over to Harith.

Good morning, everyone. Thank you for joining us today. We are thrilled to present strong 1-year data from the randomized REMAIN-1 Midpoint Cohort, demonstrating meaningful and durable weight maintenance with Revita, 1 full year after the discontinuation of GLP-1 medicines. The key result is that patients treated with Revita maintained up to 84% of their GLP-1-induced weight loss at 1 year compared to only 46% in the sham arm among patients who received complete duodenal ablations. This addresses the most significant unmet need in obesity today, durable weight maintenance after stopping GLP-1 therapy. The data shows something many people believe was not possible that durable drug-free weight maintenance can be achieved with a single outpatient endoscopic procedure. We are also pleased to have with us today Dr. Adarsh Thaker, the co-lead of the UCLA Bariatric Endoscopy program and the principal investigator of the REMAIN-1 study, who will share his perspectives on the data and his expectations for the upcoming pivotal trial results. We believe these positive results significantly strengthen the probability of success of our ongoing pivotal trial. We look forward to reporting top-line 6-month data from our landmark REMAIN-1 study early in fourth quarter of 2026, followed by potential FDA De Novo submission later the same quarter. With positive 1-year randomized data in hand and major catalysts approaching, we believe now is a particularly compelling time for investors to engage with our story. We have very strong conviction in Revita based on 4 key pillars shown here. First, the clinical signal is real. Second, our pivotal study is built to win. Third, there is a clear path to commercial value. And fourth, we are funded through definitive data later this year. For those new to our story, Revita is an endoscopic procedure that targets the duodenal mucosa or the first 25 centimeters of the small intestine just beyond the stomach. This short segment plays a critical role in metabolic regulation. Chronic exposure to high-fat, high-sugar diets cause long-term changes to the duodenal mucosa that impair normal nutrient sensing and contribute to metabolic dysfunction. Revita is designed to ablate this damaged tissue and allow the regeneration of healthier duodenal mucosa, offering a potential new way to restore normal metabolic regulation. Revita is like LASIK for obesity. It targets the underlying cause of disease with a single outpatient procedure rather than relying on chronic medication. This approach is particularly valuable in the post-GLP-1 setting where millions of patients are looking to maintain their weight loss without staying on therapy long term. The REMAIN-1 Midpoint Cohort is our randomized sham-controlled, proof-of-concept study in post-GLP-1 weight maintenance. The study was not formally powered for statistical significance. Our goal was to generate descriptive data to better understand the treatment effect exactly as a Phase II informs a Phase III and help inform the design of our pivotal trial. We enrolled patients with obesity with a BMI between 30 and 45 who were GLP-1 naive and did not have Type 2 diabetes. Patients were started on tirzepatide to achieve at least 15% total body weight loss. GLP-1 therapy was then discontinued in these participants, and they were then randomized 2:1 to either Revita or sham. The primary efficacy endpoints were the percent total body weight change from post-GLP-1 nadir between Revita and sham at 3 and 6 months. Today, we are presenting the 12-month data with patients and investigators remaining blinded to their treatment allocation throughout the full year. 41 of 45 randomized patients completed the full 52 weeks of follow-up. All discontinuations were due to patient withdrawal of consent. And importantly, there was no reinitiation of GLP-1 therapy after randomization. Today's safety and efficacy analysis include all available data. When we presented the 6-month results from this midpoint cohort earlier this year, we highlighted 2 key observations that directly informed our prespecified pivotal statistical analysis plan. First, we observed a clear dose-response. Patients who received more complete duodenal ablation maintained significantly more weight at 6 months. Based on this finding, we defined the per protocol population in our pivotal study as those who received a complete ablation of more than 14 centimeters. Second, we saw that patients who lost more weight on GLP-1 therapy were at higher risk of more regain after discontinuation. In the midpoint cohort, the separation between Revita and sham was, therefore, greater in patients with higher GLP-1-induced weight loss. Based on this, our pivotal statistical analysis plan includes run-in weight loss as a key covariate in the primary analysis. And we applied this exact same methodology to today's 12-month analysis to give investors as much read-through as possible to the potential results of the pivotal study. We are presenting data today across 4 analysis populations. The full mITT population includes all randomized and treated patients, consistent with the primary analysis in our pivotal study. We are also showing data in a high-run-in population, a complete ablation population and an optimized population that combines both high run-in weight loss and complete ablation. We believe this optimized population best represents Revita's ultimate clinical profile. Turning to the efficacy results at 1 year. Revita clearly reduced weight regain with the benefit becoming more pronounced when an adequate dose of ablation was delivered. In the full mITT population, Revita reduced weight regain by approximately 40% versus sham, and this increased to over a 60% reduction in regain in patients who received complete ablations. The sham arm regained approximately 13% from nadir or 54% of their GLP-1 weight loss at 1 year, which is consistent with but slightly lower than the roughly 60% regain that is reported in SURMOUNT-4 and other post-GLP-1 meta-analyses. This similarity provides an important external validation for this study's findings. Looking at the complete ablation population over the full year, analyzing the MMRM that we will be using in the pivotal statistical analysis plan, patients first underwent the GLP-1 titration phase, losing a substantial amount of weight. After discontinuing GLP-1, they were randomized. At 1 year, Revita patients retained 81% of their weight loss, while sham patients retained less than half. This large treatment effect appears durable throughout the 1-year follow-up period. Importantly, the median ablation length in our ongoing pivotal trial is 16 centimeters, meaning that a majority of patients in the pivotal are expected to have received complete ablations. Turning now to the optimized population, the right dose in the right patient. In this group, Revita patients regained only 16% of their GLP-1 weight loss at 1 year compared to 54% in the sham arm. This represents over a 70% reduction in weight regain versus sham. The pivotal cohort had a mean GLP-1-induced weight loss of over 18% during the run-in period, which suggests that a majority of pivotal participants will be well represented from this high run-in weight loss population. Let's put this in practical terms. A patient starts at 220 pounds, loses 44 pounds on tirzepatide, weighs 176 pounds at the time of randomization. They would regain 23 pounds on sham over a year, but only 8 pounds with Revita, meaning they would maintain 36 pounds of their GLP-1 weight loss 1 year later. Turning to safety. Revita continued to demonstrate a favorable safety profile through 1 year. There were no new device-related treatment-emergent adverse events between 6 and 12 months, no device-related serious adverse events in the entire study. We saw no excess frequency of any treatment-emergent adverse events with Revita at all. There were only 4 incidents of TEAEs in total. Intriguingly, there were no new diagnoses of Type 2 diabetes in the Revita arm, but there was 1 diagnosis of Type 2 diabetes in the sham arm. These 4 mild treatment-emergent adverse events occurred in 2 patients, let's zoom in on them, abdominal discomfort, sore throat, nausea, dry mouth, all resolved within 2 days with no further device-related adverse events at any point in the study. This is an incredibly mild periprocedural profile, especially when compared to the GLP-1 medicines that represent the only treatment alternative for these patients. We believe it supports a potentially very broad outpatient utilization for a large population. Okay. With 1-year data presented, let's now turn to our ongoing REMAIN-1 pivotal study. This trial has the same design, patient population and enrollment criteria to the midpoint cohort you just saw. It is being conducted by the same investigators using the same protocol. So what this means is that the pivotal study is essentially a larger, well-powered version of the exact same experiment you just saw: same patients, same doctors, same protocol. This significantly derisks the pivotal results for us. Because of the strong similarity between the 2 studies, the midpoint cohort provides valuable read-through to the ongoing pivotal. Based on the data we've just shown you, both co-primary endpoints in the pivotal study are powered at over 95%. Let's take the first co-primary. It is the percent of total body weight regain in Revita versus sham at 6 months. The table shows illustrative sham 6-month regain in the mITT population versus what Revita regain needs to be in order to be able to win statistically. Hold that up against what we observed in the midpoint cohort in the mITT population, and you will see the results clear comfortably. Let's take the second co-primary endpoint, the responder rate. The FDA-mandated prespecified performance goal is a single-arm result of at least 50% of patients maintaining at least 5% of their total body weight loss at 12 months. In the midpoint cohort, this figure was 73% in the mITT population and over 90% in the complete ablation cohort. So in summary, we saw substantial weight loss maintenance with up to 84% of on-drug weight loss retained 1 year off GLP-1 when an adequate dose of ablation is used versus only 46% in the sham arm. Given the absence of any dose-limiting toxicity observed, we see room for even further profile optimization in the future as we continue to develop our technology. The periprocedural AE profile supports what we believe to be a very broad potential outpatient endoscopic use. And all of this translates to a well-powered pivotal with top line data expected in early Q4 2026 and a potential De Novo regulatory filing later next quarter. So with that, I'd like to turn the microphone over to Dr. Adarsh Thaker. 12 days ago, he became Associate Professor of Medicine at the David Geffen School of Medicine at UCLA, and he's also the co-lead of their Bariatric Endoscopy Program. He's also a clinical adviser to us and a REMAIN-1 investigator. Dr. Thaker?

Thank you, Harith, for the introduction, and it's good to be here this morning to review this data. As you mentioned, I'm a gastroenterologist. I've been involved in the Type 2 diabetes trials and now I'm involved in the weight maintenance program for the Revita DMR procedure. And as you mentioned, I have a busy practice treating patients with obesity and metabolic syndrome with endoscopic therapies. And when I saw these data, I have to say that I'm highly enthusiastic about the results. I think particularly compared to the data before, we continue to see clinically significant separation between the 2 groups. The separation is meaningful and appears to be sustained. I think that one of the worries that I had and I think it was shared by others that perhaps there was no plateau in the earlier data and maybe performing the DMR procedure is not actually giving us true separation, but just delaying the inevitable with the weight gain and shifting it downstream. But I think as we start to see the curves get to their 1-year mark and beyond, we are starting to see hints of a plateau even with the small numbers in this pilot study. And so from my standpoint, I think with the numbers presented here, the benchmarks for my own clinical practice, for my partners to accept it in clinical practice and most importantly, for patient acceptance, I think those benchmarks have clearly been surpassed. There's a lot of different ways to present these numbers. And I think that I think about how I would describe it to a gastroenterologist or even a patient or considering incorporating or undergoing these procedures, respectively. And I think that slide where Harith showed how much weight regain in the 220-pound patient, the way I would say is that if you do this procedure, to keep an extra 15 pounds off compared to just stopping the GLP-1 agent. I think that's a meaningful and powerful message. That is not easy to do. And I think that framing it that way with simplicity, I think, again, we will gain a lot of acceptance in the field and by patients. I want to highlight that I think that the dose optimization information is quite important. As we went through the study, we had our own learning curve. We had our own kind of safety tolerance and as we saw the safety data is really, really good. And so the training wheels and kind of the kid gloves or gentle hands that we were using early on in our learning curve experience kind of came off, and we were able to be more aggressive. And for that reason, I'm hoping that these data represent a floor and not a ceiling in terms of how effective the treatment can be to maintaining the weight loss. And so I'm looking forward to the pivotal numbers. And the safety data is in line with my own observations of what patients have been experiencing after undergoing the procedure. Kind of looking forward, as we await the pivotal results from the larger cohort, I think it's important to reflect on that there's often a difference between what we as providers want to see in terms of data and what patients actually want, particularly in a world where upcoming GLP-1 agents are revealing ever-increasing weight loss numbers. Patients don't need a grand slam. They don't need perfect separation in terms of how much weight regain they might have between this and just doing nothing, stopping cold turkey. Patients just need a boost. So even some separation I think will be good enough to get widespread buy-in and certainly keep me immediately busy from even just my own patients, my own practice. Even a 5% to 10% weight loss compared to the original baseline is linked to very meaningful health benefits. And that's often what we counsel patients with things like fatty liver, diabetes. They don't have to get 30% weight loss, even 5% to 10% will help them. And so I think if we can -- if I can say to a patient that doing a Revita DMR procedure is more likely than not going to keep most of your weight off after you stop this GLP-1 medication, that will be enough to get buy-in. And the reason for that is that the safety data is so good and the stakes are so low. The procedure really is more of a time inconvenience. You have to spend half a day, take off a work and find the ride home after undergoing anesthesia. But why not do this? It's very well tolerated. It's a straightforward procedure. We're only going to get better and more efficient at it. And really for the -- again, the time commitment and having to go under anesthesia for a little while, the patients are being free from a pharmacologic agent. I'll finish with kind of just an anecdote where we were in the REMAIN-1 study, but for one reason or another, our start-up and activation was towards the back end. And we were told that we only had a couple of weeks to fill our entire study cohort if we wanted to participate. And my obesity medicine partner went to work and just from her own clinic, within 1 week, we found 25 patients and filled our entire quota of participants. And that was without even leaving or advertising or talking to other people, just from her own clinic. And that's how quickly we filled the study. Now you could say, okay, maybe it was the free tirzepatide that they're getting and that was the reason that patients wanted this. But I was in the REVITALIZE 1 trial for Type 2 diabetes. And back then, we had minimal to no data, very little safety data. And even then, patients, when they heard about the possibility of coming off of medications, the buy-in was very high. I've never had a single patient, even when I have very little data to back up my claims on what this could do, never had a single patient refuse to consent for the procedure. So I think the patient drive is strong. And I think that, again, we don't need to see grand slam numbers for myself and other gastroenterologists who do bariatric and metabolic endoscopy procedures to immediately be overwhelmed with patient demand after this procedure. So I'm hopeful, and I look forward to seeing the overall cohort data. And I'll turn it back to Harith.

Thank you, Adarsh. In closing, we see a clear near-term path to a potential launch. Later this quarter, we're going to have an Investor Day to walk through our view on the substantial commercial opportunity ahead of us. You'll remember earlier this quarter, we hired Mike Zumdahl as our SVP of Market Access and Commercial Strategy. So he's going to walk you through our strategy to tackle this exciting new market development opportunity in front of us. And in early Q4, we will present pivotal top line 6-month data. And then, in late Q4, we anticipate our potential FDA De Novo submission. With that, I'll pause. Dr. Thaker and I would both be happy to open the floor for questions.

[Operator Instructions] So our first question comes from Michael DiFiore at Evercore.

Congrats on the great data. So 2 questions for me. For the pivotal trial, how will dropouts be handled statistically? And separately, can you remind us of the dietary modifications in the pivotal trial? And how they will be monitored and forced throughout the study?

Sure. So Michael, dropouts are handled in the MMRM analysis that we've prespecified as missing at random and MMRM handles them through their -- through its own modeling. We will have some sensitivity analyses as well, but that's the primary analysis. Dietary modifications in the pivotal are exactly like what we have done here. You will note that there were some patients who are better about following it than others. None of the data was excluded from this analysis, and we are confident that we have done a good job of centralizing an appropriate moderate diet and lifestyle program through centralized blinded dietitians who are overseeing the work of individual site level dietitians at every clinical trial site. It's literally the same diet and exercise program that all of the GLP-1 studies have also run, 500,000 calorie deficit plus some exercise recommendation each week.

Our next question comes from Whitney Ijem at Canaccord.

Congrats on this update. A question for Dr. Thaker. I guess, can you -- what are you currently offering patients who come to you for endoscopy-based procedure, obesity patients who come to you for an endoscopy-based procedure? And can you maybe compare and contrast how these data compare to those options?

Sure, Whitney. Right now, the primary procedures we perform are endoscopic sleeve gastroplasty. So it would be a suturing-based procedure where we're reducing the size of the stomach through the mouth. And we use that as an alternative to patients who may undergo surgical sleeve gastrectomy but don't want to have the incisions or the aggressiveness or changes in surgery. And then followed by that is actually revisions of patients who had actually already had bariatric surgery and need tightening because their anatomy has become dilated. With the primary ESG, I'll talk mostly about that one, but typically quote patients is having between 14% to 17% total body weight loss at 1 year. That's the average. Real-world numbers vary significantly based on diet and exercise programs and settings, but that looks to be around the number. So I think it's a slightly different procedure because we are doing that primarily for weight loss. It's something that I'm certainly interested in looking at to see if that can be used as an off-ramp for GLP-1, but I don't think that the physiology alteration is as good as what we're doing here in the duodenum. In the duodenum, we're not changing the anatomy. We're trying to restore a normal healthy gut tissue physiology. And so I see no reason why you couldn't do this in combination with the ESG either at the same time or in tandem and really see if we could start to meet surgery level numbers by doing everything endoscopically.

Got it. Really helpful. And then just one quick follow-up for the company. I think in the previous randomized cohort data, you talked about the p-value and kind of the -- with an understanding that this is a small n and read-through to the pivotal. Unless I missed it, can you remind what did you see as a p-value here? Just kind of curious how that compares to the last update and read-throughs to the pivotal?

Well, we have not powered the study for formal statistical inference, you can look at the standard error bars, and you can see that there's clear separation. There is no question that what we are seeing is going to be highly, highly statistically significant if it continues in the pivotal trial.

Our next question comes from Chase Knickerbocker at Craig-Hallum.

Maybe just 2 for Dr. Thaker to start. We greatly appreciate your thoughts as far as kind of what the level of efficacy that needs to be seen in the pivotal for however you want to define it, but substantial adoption of Revita in your practice ultimately, if approved. So is that bar however you want to define it, whether that's retention of weight loss or kind of absolute weight loss from GLP-1 baseline? Just your thoughts there would, I think, be really helpful to contextualize this for us.

Yes. I think the second point that you made is the easier way, how am I going to describe this to my partners, to referring physicians, or to patients, which is from the original start before the GLP-1, how much weight can we keep off if we did a one-two punch of GLP-1 and then DMR to get the patient off of it? And I think that even 5% weight loss maintained compared to the original baseline would be a win. 10% would be great. And that would mean a lot of the other endoscopic procedures I do and have meaningful health benefits. So I believe that's one of the secondary study endpoints where if we can keep more than 5% total body weight loss from the original baseline after a DMR treatment, to me, that would be enough to offer this to patients. And if it's more than 10% from total body weight baseline, then I think that it will be a very popular and an easy sell.

Got it. And then maybe just second, as you've had greater experience with Revita, done more procedures, maybe just talk to us about how your kind of comfortability going longer with the ablation length has progressed and then also kind of the time to complete the procedure, how that's developed? And kind of how the level of complexity here compares to some of the other things that you do in your practice?

Yes, certainly. This is really the first time anyone's bothered to go after the duodenum like this, even though we've known that it's a key regulator for the metabolism. And one of the reasons that duodenum is it's thin-walled and it has a rich blood supply. So we've always been afraid of complications. The other major complication we were afraid of is pancreatitis. So early on, particularly in the diabetes studies and for providers who had picked up the device for the first time for this study, we were being very careful and of trying to not overdo it. And as a result, we ended up, I think, in many ways, underdoing it. And that's why you're seeing not every patient achieving this dose optimization, optimized dose length, which roughly to me, translates into 7 ablations. Can you do more than 7? As I got more comfortable and the way that I would teach it to now to someone who is picking it up for the first time, I would say that the duodenum can take it. It turns out the data and again, my own experience has shown that we don't need to be so careful. So while before we were obsessing over avoiding overlapping treatment areas, avoiding going too far, avoiding starting too close to the beginning of the duodenum, now we can be more aggressive without sacrificing safety. And so we can start pushing the bar. And I think that's a learning curve that I had to go through that just with simple training and coaching, other providers won't need to. So I think that we can get that into their hands. much more quickly to make sure we meet that benchmark of 7 ablations or more.

And how complicated is this?

At the moment, it is -- there was a little bit of a procedure and a technical learning curve of driving the catheter, et cetera. There's a lot of work being done to democratize this procedure really to make it as easy and as quick as a screening colonoscopy. And when you think about how many people -- I think the number needed to treat for screening colonoscopy is like 450. So the reality is that most patients that were doing screening colonoscopy are not being helped by that procedure, whereas I think with this, we're going to see a lot more meaningful benefit on an individual level. So I envision a world where you could do this in line with or even at the same time as a screening colonoscopy one day. And I think that's the goal to make this under 30-minute procedure that can be done by any gastroenterologist who wants to. And so we're working hard to make that happen.

Our next question comes from Chi Fong at Bank of America.

This is Chi on for Jason. The first one that I have here is I want to revisit Slide #15 from the corporate presentation where you provide some nice sensitivity analysis on what Revita need to hit in the pivotal cohort data for it to be statistically significant. And curious your thoughts about the base case assumption of 11.6% weight regain from the sham arm in 6 months. If I recall correctly, the 6-month data from the midpoint cohort the sham arm regained 7.5%. So if you can talk about the differential assumption there, that would be great. And in the event that the sham arm did perform more similar to the midpoint 6-month data, 7.5% weight regain, what would you need -- what would Revita need to hit to still hit that? Say with the 5.1%, they'll be sufficient or not? That would be my first question.

Yes. So great question. Bottom line is that the 5.1% and the 11.6% that we're showing here are measured in exactly the same way as the pivotal statistical analysis will be conducted. And that explains the difference between the number that we reported back in January and the number that we are reporting now because we're taking run in weight loss as a treatment interaction term as we laid out on in one of the earlier slides. So I think you can see that the Revita regain needed to win is -- the delta versus the illustrative regain in the mITT population is roughly 2.5 percentage points. That is the number that we would need in order to feel comfortable regardless of the methodology used.

Got it. And my second question is that under the De Novo pathway, do you think that the 1-year midpoint data that you have today, coupled with the 6-month pivotal data in 4Q could be just sufficient for approval consideration? Or do you think that even in the De Novo pathway, would the FDA still require the 1-year data from the pivotal cohort for registrational purposes?