United Therapeutics Corporation (UTHR) Earnings Call Transcript & Summary

Good afternoon, and welcome to the United Therapeutics Corporation Phase 3 TETON-2 Results Conference Call. My name is Drew, and I will be your conference operator today. [Operator Instructions] Please note that this call is being recorded. I would now like to turn the webcast over to Harry Silvers, Investor Relations Manager at United Therapeutics.

Thank you, Drew. Good day, everyone. It is my pleasure to welcome you to the United Therapeutics Corporation Phase 3 TETON-2 results conference call. Remarks today will include forward-looking statements representing our expectations or beliefs regarding future events. These statements involve risks and uncertainties that may cause actual results to differ materially. Our latest SEC filings, including Forms 10-K and 10-Q, contain additional information on these risks and uncertainties. We assume no obligation to update forward-looking statements. Today's remarks may discuss the progress and results of clinical trials or other developments with respect to our products. These remarks are intended solely to educate investors and are not intended to serve as the basis for medical decision-making or to suggest that any products are safe and effective for any unapproved or investigational uses. Full prescribing information for the products is available on our website. Accompanying me on today's call are Dr. Steve Nathan, Schar Chair of the Advanced Lung Disease and Lung Transplant Program at Inova Fairfax Hospital and Chair of the TETON Steering Committee; Dr. Leigh Peterson, Executive Vice President of Product Development and lead for the global TETON program; and Dr. C.Q. Deng, our Senior Vice President of Biostatistics, statistical programming and data management. Now I'll turn the call over to Leigh to begin our overview. Leigh?

Thank you, Harry, and good afternoon, everyone. We have slides available, and I encourage you to review those when you have a chance. Earlier this month, we were incredibly pleased to announce positive results from the TETON-2 pivotal study evaluating the use of nebulized inhaled treprostinil for the treatment of idiopathic pulmonary fibrosis, or IPF. The TETON-2 study met its primary efficacy endpoint of demonstrating improvement in absolute forced vital capacity or FVC relative to placebo. Statistically significant improvements relative to placebo were also observed in most secondary end points. Earlier today, we presented additional analyses from the trial at the European Respiratory Society Conference here in Amsterdam, which we will review during this webcast. These results mark a historic landmark in the landscape of IPF treatment. We believe these results not only demonstrate the efficacy of nebulized Tyvaso but also signal a turning point for patients who have had limited options and uncertain outcomes. These findings are more than an unprecedented statistical success in IPF, these positive results represent hope and progress validating years of collaborative clinical research and patient commitment. At the heart of these positive results is the power of treprostinil which we have long believed to be more than just a vasodilator. Through rigorous study design and thorough analysis, TETON-2 provides robust evidence that Treprostinil's antifibrotic effects can improve key measures such as lung function as measured by FVC and quality of life. Importantly, the trial also reinforces the underlying mechanism of action or MOA, demonstrating how targeting multiple pathways can bring tangible benefits to patients. This scientific validation of Treprostinil's MOA highlights its promise and transformative potential to advance pulmonary medicine. We believe that the benefits truly extend beyond the data, offering patients the promise of better days and greater stability in their journey with IPF. Dr. Nathan will provide additional details a detailed analysis of the TETON-2 pivotal study. But first, I'll turn the call over to Dr. Peter Smith, who is responsible for running our global TETON program. He will begin with a brief review of the TETON-2 pivotal study trial design. Peter?

Great. Thank you, Leigh. Okay. So we'll take a look at the TETON-2 study design. TETON-2, the pivotal study was a 597 patient multicenter, randomized, double-blind, placebo-controlled Phase III study evaluating nebulized Tyvaso in IPF patients over 52 weeks and 16 countries outside the U.S. and Canada. Full enrollment was reached in July 2024. Participants in the study were randomized to nebulized Tyvaso or placebo starting at 3 breaths 4 times daily or QID. And then titrated as tolerated up to 12 breaths QID. Participants who completed week 52 on study drug were eligible to then enter an open-label extension study. The primary endpoint was absolute change in Force Cloud capacity at week 52. Secondary endpoints included time to first clinical worsening event, first acute IPF exacerbation, overall survival percent predicted FVC, quality of life as measured by the King's brief interstitial lung disease questionnaire and change in DLCO. Safety was assessed via adverse events, labs, vital signs and ECGs. Next slide, please. So let's take a look at the eligibility criteria. In terms of key inclusion criteria of note, we included subjects who were over 40 years of age, had a predicted FVC of 45% or more subjects should have been on a stable dose of nintedanib or pirfenidone if they were using one of those medications. And diagnosed with IPF as confirmed by a high-resolution CT within the previous 12 months. Key exclusion criteria of note included evidence of obstructive disease, high supplemental oxygen use of drugs commonly used for PAH, recent IPF exacerbations or pulmonary infections. Next slide, please. On Slide 9, you can see the baseline characteristics for the TETON-2 pivotal study by treatment group. The study arms were well balanced across all criteria and no significant differences were observed between the treatment groups at Phase. Both treatment arms were representative of a typical IPF patient today. Now I'll turn the call over to Dr. Nathan to walk us through the data in more detail. Steve?

Thank you, Peter. And moving on to the next slide. There you see the results of the primary which, as Peter mentioned, is change -- placebo-corrected change in FVC 52 weeks. What's not notable on the silo of a couple of things. First of all, patients are up-titrating their drug between 0 to 8 weeks, approximately in excess of 82% to 85% of patients who are on at least 10 to 12 reps by 8 weeks. So they weren't on really an effective dose for the first 4 weeks when you see that up. But then at 8 weeks start to come apart, and we hit statistical significance as early as 16 weeks and the curves look like they continue to come apart. Now towards the end, what we have here is the Hodges-Lehmann Estimate of the difference between the two groups, which came out at 95.6 mL, which was highly significant. What you see for all the top points along the way in circles is the observed mean data, mean FVC. And you'll note a square plus the circles at the end at 52 weeks. The squares represent include imputed data, and it's reported as the median. So we have the median difference at 52 weeks, including imputed data of 95.6 mL. This is a forest plot, which includes imputed data at each time point of change in FVC and once again see that we reached statistical significance at week 16. And the point after that's continued to come apart or move further from the line of unity up until 52 weeks. So it appears that there is ongoing improvement when we use imputed data in all these analysis at all these various time points. Moving on to the next slide. Yes, we have the change in FVC by background therapy. We'll start off on the left with no background therapy in the treatment arm, we have a change over 52 weeks of 54 cc compared to the placebo arm, where it comes in at 107 cc. The placebo-corrected difference using the Hodges-Lehmann estimate once again, it's just under 44 cc. And this was not statistical significant -- but it's not statistically significant. You can see the p-value there at 0.33. Now there are probably two reasons for this. numerically, there is a quite a difference in the two numbers, but the numbers are relatively small. And the placebo group didn't deteriorate as much as expected as we'll see in other groups as we move on to background anticipated therapy. If we move to the middle panel, here, we see patients who are on background than tender. And this did reach statistical significance with the Hodges-Lehmann estimate of the difference being 97.6 mL. The treatment arm went down by 43 cc over the 52 weeks versus a placebo arm at 115 cc. And then moving to the last panel to the right, we have changed based on background pirfenidone. And for the treatment arm, 50 month, there was a decrement of 52 cc versus the treatment arm of just under 188 cc. Now what's notable, if you look at all the treatment arms, the dark blue, the treatment effect was very consistent across these three groups, 54, 43, 52. So almost replicate numbers be they on antifibrotic therapy or not. Moving on to the next slide, where we look at probably our most important secondary endpoint, which is time to clinical worsening. This also met statistical significance with a 29% relative reduction in the risk of clinical worsening. This -- I don't think we mentioned it, but clinical listening was a composite, and I'll show you the components of the composite in the next slide. If you look at the absolute numbers, there were 27.2% of the patients on active inhaled Treprostinil who had clinical worsening versus 39% in the placebo arm hazard ratio of 0.71. And once again, this met statistical significance of the P value as shown of 0.019. And here are the components of clinical worsening. It was the time to the first of these three events, death, respiratory hospitalization or 10% relative decline in the predicted FVC. Most of these events were driven by change in FVC, which by itself favored inhaled treprostinil. Then we have respiratory hospitalization once again, numerically favoring inhaled treprostinil. And death at far us almost equivalent numerically more patients in ultraprecise point as a first event, four to three. But as you'll see in a subsequent slide, there were actually more deaths during the course of the study on placebo versus inhaled treprostinil. Moving on to the next slide. One of the other secondary endpoints was time to first get exacerbation of IPF in the source adjudicated by a panel. And there was a 46% relative reduction in time to first acute exacerbation, but there weren't that many events so this didn't hit statistical significance. The number of events was 9% versus 17 or 3% versus 5.8%. So certainly numerically trending in the right direction, but this does not hit statistical significance. Moving on to the next slide. Another secondary endpoint was overall survival at 52 weeks. Same story here. numerically, there were more deaths in the placebo arm versus inhaled treprostinil arm, 19 versus 24 or 6.4% for inhaled treprostinil versus 8.1% for placebo, 23% relative reduction, that because of the small number of events is does not get statistical significance, but numerically, the certainly favored inhaled treprostinil. Another one of our secondary endpoints was change in percent predicted FVC at 52 weeks. And this slide looks almost replicate of the slide of our primary, which was change in FVC and mL. And you can see the same kind of thing here with the initial decrement in both groups as they upside traded and in the curve start to separate at week 16 with the difference as shown of 2.7% at 52 weeks. Moving on to the next secondary endpoint, which was our patient reported outcome measure that cable, as Peter mentioned. And here, we hit statistical significance, again shown to the left for the overall cable. The numbers are there. a p-value of 0.01. And the various domains that constitute the cable are shown to the right. Most of the difference driven by the difference in breakfast and activities domain as well as psychological score. Both of these independently hit statistical significance. The next secondary endpoint, which was also positive, favoring in heliport was change in the percent predicted diffusing capacity and showing the difference here of 1.91% and a p-value of 0.04, favoring once again inhaled treprostinil. This is a busy slide. This is a forest plot of various subgroups. And the major point to take home from this is that all point estimates are to the right of the line of Unity. So there was no subgroup that didn't have a benefit some of these by themselves in statistical significance because of the numbers. But all of them, again, were to the right of the line immunity, indicating a consistent effect across all the different subgroups. I will draw your attention to the top right, last study drug dose less than we dichotomized this by less than 9 breaths. The patients are able to achieve only 7 or 8 breaths. The numbers are fairly small, 62 versus 30. But what you'll note underneath that are the patients who had 9 to 12 breaths per session and that was the majority of patients. And if you look at this higher dose group, the difference was actually 107.8 cc in terms of FVC. So those rates the question of whether there might be some kind of a dose response. Moving on to our adverse events. This was an IPS study over 52 weeks. These are sick patients, and it's no surprise that the majority of them, 919 had at least one adverse event. There were no adverse events that were a big surprise. A number of subjects with at least one study drug-related AE. There was a difference there of 64.8% versus 42.4%. But generally, the drug works pretty well tolerated. These are the major adverse events that we're seeing. Cough occurred most frequently at 48% versus 24% I think the thing to bear in mind is that coffee is a common symptom of these patients with IPF anyway. So how much of this cost was preexisting versus exacerbated by the medication is a little bit uncertain. The rest of the side effects in a area are typically typical customer side effects that we've seen in many of the clinical studies that inhaled treprostinil in that patients experience in the clinical trenches as well. So in conclusion, TETON2 met its primary endpoint of change from baseline in absolute FVC at mean 52. The study also made statistical significance. Maybe last divestor was well tolerated and the safety profile was consistent with previous inhaled treprostinil studies and non prostacyclin-related adverse events. So with that, I will turn it back over to Leigh.

Thanks a lot, Dr. Nathan. And for your trailblazing contribution contributions in the field. We're really grateful to have you with us today, and we appreciate the valuable perspectives you've shared. In summary, these data are a significant milestone for individuals living with IPF. And I extend my heartfelt gratitude to all of the trial participants and their families the dedicated teams at United Therapeutics and the investigators whose commitment made this achievement possible. The robust and carefully executed TETON2 trial has delivered persuasive evidence that treprostinil can meaningfully improve FVC for patients who previously faced limited treatment options. Most of which offer only modest benefits and often come with difficult side effects. Equally, we observed benefits on several key secondary end points all while demonstrating a well-tolerated safety profile that's consistent with previous Tyvaso studies and known prostacyclin-related adverse events. We consider these findings a decisive achievement and are committed to swiftly seeking regulatory approval for the IPF indication and advancing this important therapy to those patients who stand to benefit the most. We look forward to providing results from TETON-1 pivotal study, which evaluates participants in the U.S. and Canada in the first half of next year. So now I'd like to open the call to questions. Operator?

We will now begin the question-and-answer session. [Operator Instructions] the first question comes from Olivia Brayer and Cancer Fitzgerald.

Congratulations on a really great result here. Can you talk about the decision to use Hodges-Lehmann in the TETON studies. And really what's driving that difference between the observed versus the imputed mean. I'm under the impression that HL hasn't necessarily been a stat method of choice in prior IPF studies. So just any thoughts there? And then around the discontinuation rates, is that something you guys have disclosed yet, both for the overall study but also across the different treatment arms. If I'm looking at this correctly, it looks like only about 203 and 212 patients were evaluable at 52 weeks. So any color there would be really helpful. And then I've got one more question if we've got time at the end.

Thanks, Olivia. C.Q., we'll handle the stats question. And then we'll kick it back to Dr. Nathan for your question on discontinuations.

Yes, for the strategical model, initially, we propose from Metros using mixed model repeated measure. But in all of these analysis that if you use a pragmatic model, you have to check the assumption for using that model, see if that assumption still hold. And in our TETON-2 study, we -- after you fit the model, we check the assumption, essentially you check the normal distribution of the residual from that model. Then the monogenic the result indicate that assumption wide. So we then switch to the nonprime rate-based approach to analyze the data. But we did all these analysis from prime age and nonprime age in both the way he says. So results are consistent. So but from a digital standpoint, if your assumption for using primatic model, it's violated use supposed to switch it to the number of mice. Now I'll turn to the question to Steve.

Yes, sure. No, that's -- you did pick up that there were 203 and 212 patients at the end of 52 weeks to continuing on therapy. And there were premature discontinuation. If you look at the 298 who started out in the treatment arm or inhaled treprostinil, 100 prematerial discontinued treatment and 50 of these were due to adverse events another 32 due to patient decision. And then what we have coded is other for the other team. In the placebo arm, there were 73 who pre materially discontinued treatment, 37% for adverse events, 22% was a patient's decision and 14% for other reasons. And there patients who discontinued before week 52, but still continued in the study. And so we had 224 versus 239 patients who completed the study at 52 weeks. i will have to say that this was a big ask of patients. We were giving them a medication that's nebulized four times a day. We were asking them to go 52 weeks. And we're giving them a medication that we and they didn't know if it worked night and we and they didn't know if they were getting placebo or the real deal. And so I think the component of this is really just nebulized a fatigue over 52 weeks. Once the drug is approved for IPF, I think there's added motivation for the providers as well as the patients to continue on therapy because now we know it works, and now they will know that they're getting the real deal. So we might actually see -- I would speculate, you might actually see less discontinuation once it's in the clinical arena for IPF.

The next question comes from Andreas Argyrides and Oppenheimer.

All right. Congrats on the impressive results across the board. Can you just maybe -- it may be hard to do so, but can you just speak to what you think is driving the magnitude of benefit? We've talked about the preclinical evidence of the treprostinil antifibrotic effects. So maybe a little bit of color there? And then looking at these results, is there any one particular end point that you find important for clinicians. And then just one more addition to that is. Looking at the results on the background medications, any conclusions you can draw on why a more pronounced benefit, let's just say, the background pirfenidone versus nintedanib.

Sure. That was three questions. I need I'm not remembering and I'll start with the last because I remember that one. The three groups that we showed, treatment naive on preferred and nintedanib. For some reason, the pirfenidone group appear to deteriorate the most. The group that got placebo. And that's why we've got that big difference. But I think an important point is if you look at the 3 groups who got treatment, they were all very replicate in terms of their response around the 50cc loss of lung function over 52 weeks. To put it in further context, if you look at normal elderly individuals, normal elderly individuals over the course of 52 weeks, will lose about 30 cc of land faction. And so you put the 50 cc in that context, and I think it's a pretty impressive result. Now, the second question is, is there anything else in the data set that will resonate with clinicians. I think a big one for me is quality of life. And I showed you that. And if we can put patients on therapy that preserves or improves their quality of life, that's huge, that's going to resonate with patients. That's going to motivate them more to take the medication 4 times a day. And as you know, the agency, the FDA is all about how patients feel, function and survive. Well, this gets to how they feel directly. And that's the first study in IPF to my knowledge, that has shown an improvement in quality of life. So I think that's a pretty big secondary endpoint that we had in the study. I'll ask one of my colleague. if you don't mind. I going to ask one portfolio my colleague. The mechanism. Thank you. Yes. I think- as mentioned during this call that prostanoids and treprostinil in particular, have properties beyond their traditional vasodilatory effects. When we saw the signal from the INCREASE study, of the FEC change over 16 weeks. We went back into the literature, and we did a deep dive. And in actual fact, there's a large body of evidence attesting to the antifibrotic properties of treprostinil. It engages at least four different prostanoid receptors that have downstream antifibrosis properties. And so it appears based on biologic evidence that it's a per traffic molecule in terms of its antifibrotic properties. Now what I really like about it is we're giving it inhaled, a different route of administration. We're getting deposition of drug directly to where we want it within LBI. And if you think about the current paradigm of the pathogenesis of IPF, we talk about epithelial to mesenchymal transition, that's lining alveolar epithelial cells that metamorphosize and transition to fibroblasts. And now we're getting drug directly down there where that activity is taking place versus giving systemic drugs currently available for pirfenidone and nintedanib gets into the blood stream we're not entirely sure where in the lungs is going. So to attack this disease directly by in held right makes a lot of sense, and it's very gratifying to see that borne out in the results from TETON-2. The next question comes from Joseph Thome at TD Cowen.

I appreciate the presentation. Maybe for Dr. Nathan, can you discuss how you would incorporate Tyvaso into your treatment paradigm if it's approved, would this be a frontline combination therapy? And earlier in your ERS session, have that pirfenidone BBI. Whereas Tyvaso looks to have about a 50-milliliter decrease, no matter what, how would that impact your prescribing decision of boiler? And then anything different in the U.S. and Canada that we should think about will we extrapolate these results into the findings expected findings for two time line?

Sure. If or when Tyvaso is approved for IPF, I think it will be nice to have different treatment options. I think we are in the era of multi-modality therapy for IPF, it makes a lot of sense in all the different diseases that we've managed to chip away at, including pulmonary arterial hypertension. Now it's dual therapy triple therapy patrol therapy. I think we're probably going in that direction with IPF as well. And as I mentioned, giving an inhaled medication on top of systemic medication makes a lot of sense to me. And I did the presentation this morning, I gave the analogy of attacking the disease from the front and attacking the disease from the rear. So to me, it's a very attractive option to give multi-modality therapy with an oral medication as well as an inhaled medication. How I might do this on a case-by-case basis is really going to depend on that patient interaction. But I would lean more to doing multi-modality therapy in these patients. I will say that we started multi-modality therapy in PAH before we knew it actually worked in PAH. And so I think we might see the same kind of paradigm emerge in IPF. Yes, we actually, I don't think that there's going to be a difference, at least based on the baseline demographics of these two patient populations. There doesn't appear to be anything different in these populations. I know that folks had hypothesized that, well, maybe more patients with PH-ILD will get into TETON-2 because Tyvaso isn't available in many of these different countries. -- but we actually compared the baseline demographics of TETON-1 to TETON-2, and this was an abstract presentation at ATS this year, and there really wasn't any difference between the two groups. The FEC was right about the same the DLCO was right about the same. If anything, there were more patients on supplemental oxygen from TETON-1 than they were from TETON-2. But overall, the patient populations look very similar.

The next question comes from Ash Verma with UBS.

This is on behalf of Ash. Congrats on the data. So I have two2 questions. So the first one, just looking at the time course of FVC decline in the Tyvaso arm, what do you think drove this rapid decline after week or 40 versus prior to that, it was fairly stable. And then my second question is related to like competing net do you believe they would have to pursue head-to-head study versus have so to establish clinical differentiation?

I'll address the FEC and I'm going to ask CQ sitting here next to me to help me out with that. I think we see that in IPF studies. I think the imputation also helps drive that curve down because as you get further into the study, you're getting more dropouts and then you do the imputation which generally is not favorable to the patients whose data is being imputed. That we see this in many different IPF studies where you tend to get a little bit more of a drop off towards the end. But if you look -- if you think back to that graph that I showed of the primary and you look actually at the imputed data, the squares that I spoke about, it looks like they are coming more apart than the lines would suggest. CQ, do you have anything to add to that?

Yes. especially for patients who would die before week 52 we have to apply that pretty conservative imputation approach. So for patients would die before week 52. We impute that using 2.5 percentile. In the gain, we actually even -- he is a more strict rule to impute that then after we discuss with FDA, FDA suggests we used a 2.5 percentile. So all of these imputation actually the drag it down just the mean value or medium value for all of these submissions at the late time points.

Yes, thanks for the question on TPIP. We're happy to do some follow-up. But I think today, with respect to Documents time, we're going to focus on the results presented.

The next question will come from Lisa Walter with RBC.

And congrats on the data today. Can you just a couple for Dr. Nathan on the adverse events. Just Dr. Nathan, I wonder if you can comment on patient tolerability with using the nebulizer versus using an oral pill. And were there any drop out specifically due to costs in the study? And secondly, just wondering if you were surprised by the slightly lower rates of diarrhea in the treatment arm? Was this mainly just driven by the fact that there was almost 2% less use of OFEV in the Tyvaso arm? Or was there another factor at play here? Any color here would be helpful.

Yes. I think that nebulized medication has a different adverse event profile to the oral medications. And Certainly, cough is a big one we worry about when we reported carpet was 48% versus 24%. What I would say is -- we don't have good granularity on the cost in terms of how sustained it was some patients, not from our own clinical practice of cough for 5 minutes a treatment and then it will ameliorate. So we don't know how long the cost lasted for. A lot of this cough was in the uptitration period. And once you get them through the uptitration period and generally, it stabilizes out. So even though cough is reported as ported as an AE. What we don't report is when cough gets more tolerable and becomes less of an AE and they get potentially back to a tolerable baseline cough. Now what I will say is that there certainly were patients who discontinue from a sale forecast. I mentioned the number of patients who prematurely discontinued for an adverse event. It was 50 versus 37. And between the two groups, a lot of that was cough -- but what I would say is, of those patients who discontinued for cough, there were only two who reported severe cough as an AI. So there were some other patients with mild of moderate care might have discontinued and maybe it was accumulation of gosh, I've been on this medication for so long. I'm coughing. I've had enough and we're going to stop. So I think sometimes when patients discontinue it's combination of some AE, maybe nebulizer fatigue, maybe not feeling much of a difference. This is also very different. This is an IPS study. This isn't like increase where patients have polony hypertension and some of them felt better could do more with their 6-minute walk test. This is an IPF study where the goal was and what we showed was slowing the rate of progression and preserving the quality of life.

The next question comes from Jessica Fye with JPMorgan.

Congrats on the results. you've heard Dr. Nathan I'm curious which of these clinical endpoints you think most convincingly makes the case for a direct antifibrotic effect for is Tyvaso. And was total lung capacity also measured in this trial corroborating the FVC results? The second one is just various papers suggest that a meaningful proportion of IPF patients eventually end up with pulmonary hypertension over time. And I guess recognizing that's not in the presentation today, did the TETON trial capture and collect information around any IPF patients who might have received a PH diagnosis well on the trial? And then lastly, recognizing that right heart cath is the gold standard diagnostic for PH in your earlier comments about the incredibly balanced baseline characteristics between TETON-1 and TETON-2. Beyond the trials just being balanced, I'm curious if there are any inferences that you would make from the various baseline characteristics that we do have about the proportion of patients in the trial who might have also had PH?

Sure. Okay. A lot to unpack there, so I'll try and address as much as I can. The study by purpose was totally agnostic to the presence or absence of common hypertension because our goal was purely to pursue an antifibrotic effect. Getting to your first question, which a metric or outcome directly is most attributable to an antibiotic effect. I'll have to say FEC. I think that clearly, the 95.6% difference does attest to direct antifibrotic effects. As CQ mentioned, we did multiple sensitivity analyses doing various imputations. -- and they all came out at around 90 to 95 cc. So whichever way we looked at the data, whichever way we imputed it was pretty much the same result. I think that's a pretty impressive result, especially on top of current antifibrotic therapy, as I showed you for the pirfenidone and nintedanib arms. Now you're quite right. I mean, a lot of these patients or some of these patients probably had PH that were looking below the surface. If you look at some of the papers that are out there going back to an old study, ATMOS IPF, which was looking at another drug for IPF, all those patients got cath. The inclusion criteria were fairly similar, and about 15% of those patients had pulmonary hypertension, and that was under the guys of an older definition of pulmonary hypertension. So very likely that some of these patients didn't have forming hypertension. The best surrogate we and perhaps around the few surrogates we have for pulmonary hypertension is the -- and there wasn't a difference in the NT-proBNP in terms of their baseline values. So I think I answered your first question, I think most of your questions are around formerly hypertension. So hopefully, I address everything that you asked.

The next question comes from Jason Gerberry with Bank of America.

Maybe for Dr. Nathan, just curious how you look at this group of patients who just cannot tolerate the existing NFI products in this study, the 25% of patients were not on background meds, we've heard anecdotally maybe 1/3 of IPF patients just can't take their OPEBs for tolerability reasons. So does the data here inform in your mind that Tyvaso is an option as a monotherapy? Just kind of curious how you're thinking about that and/or maybe what you need to see from the pooled TETON 1 & 2 data as this approach as monotherapy would be helpful to understand. And then when you talk about polypharmacy, right, and you've got Tyvaso, daridomolast, potentially Bristol's LPA, like is it the fields going to just be experimenting with this? Or is the catalyst for polypharmacy with triple or even quadruple therapy need to be really driven by registrational studies.

Yes. I think how we use the drug when it's approved will really be on a case-by-case basis. And you're right, those patients who are intolerant of the current antifibrotics might be first up for monotherapy. And so I think this is a very nice alternative for them. And -- the numbers you mentioned, if you read various registries and database reports, the number of patients who aren't antifibrotic at one year even much lower. Some reports around 10% or 20% of patients who started on antifibrotics continuing in a year later. So certainly, there's a large patient population who could or would be candidates for this. And hopefully, they will be much more tolerant of this versus the systemic side effects that we get with our current antifibrotics. I think how we use this drug in relation to what's approved will really be on a case-by-case basis. I think as clinicians in the trenches, we have been adopting dual therapy, triple therapy, quadruple therapy in the PAH field before that got program. And I think it's based on clinical need. When you have patients who are deteriorating and you have limited options. Sometimes you have to use your best judgment before the clinical trials come out to prove that triple therapy or quadruple therapy works better than dual or monotherapy. So it's really going to be driven by how the patients do and each clinician's interaction with individual patients how we use this drug.

The next question comes from Mazi Ali Mohamad with Leerink.

This is Maxi on for Roanna. Just kind of one really from us, which is, what do you think drove the lower decline in FVC in the placebo group versus that on background standard of care. I'm recognizing that the treatment-naive group had a smaller sample size, but just kind of seeing that there was a much larger difference there. What do you think was the primary driver there?

I can't be certain. And very interestingly, we've seen the same phenomenon in other IPS studies, including the neurodermatitfor some reason, the treatment-naive patients deteriorate less. I think folks have hypothesized that, well, maybe these patients were stable and that's why they weren't started on antifibrotic therapy, and that's why we get less of a decline in these patients because they were predestined to continue to stay stable versus patients who are deteriorating and more likely to go on therapy. One thing that's made me think about, and this is entirely speculative. I have no data to support this is do our current antifibrotics have a limited period of time where they might be effective. The studies that got the pirfenidone approved were 52 to 72 weeks. And maybe the escape mechanisms that kick in 3, 4, 5 years later. We don't know. As I say, this is entirely speculative on my part. I don't know but it's either that or patients who are predestined to stay stable who got included in these studies. But an interesting phenomenon that we've seen in other half year studies as well.

Operator, we have time for one more question.

And that last questioner will be Roger Song with Jefferies.

This is Leon Chen for Roger. So kind of following on the placebo arm in a population without background therapy. So just wondering with that data seen here but still want -- do you still see potential use of Tyvaso in the first line without background service?

Yes, I believe that it can and will be used as first-line therapy because what I would say to you and to other providers is look at the treatment effect and the treatment effect was consistent across all three groups at about minus 50 cc in terms of loss of lung function. And so in the clinical tranches, you might be -- or we might, as providers be dealing with a different patient population versus those who came in treatment naive into the clinical trial. One other point I need to make is that if you look at the baseline demographics of the treatment-naive patients compared to the rest, their baseline FVC was higher. So these patients might have also come in having milder disease, and that's why they haven't as yet been started on antifibrotic therapy. And if you take patients with more moderate or severe disease, that becomes an enrichment strategy for further deterioration. So that certainly could have also played a role.

Thank you for participating in today's webcast. A rebroadcast of this call will be available for replay for 1 year by visiting the Events and Presentations section of the United Therapeutics Investor Relations website at ir. unithgr.com. You may now disconnect.