Gossamer Bio, Inc. (GOSS) Earnings Call Transcript & Summary

Hello, and welcome to the Gossamer Bio TORREY Topline Results Call. [Operator Instructions] Please note, today's event is being recorded. I'd now like to turn the conference over to your host today, Faheem Hasnain. Please go ahead.

Hello, everyone, and thank you for joining us this morning. We're here today to discuss a critical milestone in Gossamer and seralutinib history, the positive topline results of the Phase II TORREY study in patients with PAH, Pulmonary Arterial Hypertension. Most of you have, no doubt, seen the press release we issued this morning announcing our topline results and we're looking forward to diving into the data a bit further with you on the call today. We'll be making forward-looking statements on this call, and I encourage you to read through the details on the slide. Now with us from the Gossamer team, we have Dr. Richard Aranda, our Chief Medical Officer; Dr. Rob Roscigno, the clinical lead of the seralutinib program, Dr. Larry Zisman, the inventor of seralutinib and a critical Gossamer team member, Dr. Ed Parsley, a pulmonologist who acted as our medical monitor for the TORREY study; Matt Cravets, our Head of Biometrics; Dr. Laura Carter, our Chief Scientific Officer; Caryn Peterson, who heads our regulatory group; and Bryan Giraudo, our Chief Operating Officer and Chief Financial Officer. We are also very grateful to have 2 leading clinical thought leaders in the field of pulmonary arterial hypertension with us today to help contextualize TORREY's topline results and to answer your questions. Dr. Ardi Ghofrani joins us from the University Hospital Giessen in Germany, where he heads the pulmonary hypertension division and has been at the forefront of the development of PDGFR inhibitors for the treatment of PAH. Dr. Ray Benza joins us from Ohio State University, where he is a Professor of Medicine in the cardiovascular division. He's also a former pulmonary hypertension associate physician of the year and developed the REVEAL 2.0 Risk Score tool that we used in the TORREY study, and we'll be talking about that more later. Both Dr. Ghofrani and Dr. Benza have played prominent roles in the development of every approved drug for the treatment of PAH since the introduction of Tracleer in 2001. And we're happy to have both of them on our steering committee for the seralutinib program and here this morning to review our topline results with you. So let's get into the results. Well, we are very pleased to announce the Phase II TORREY study met its primary endpoint and achieved statistical significance in change in pulmonary vascular resistance as compared to placebo. This positive result occurred despite enrolling a less severe PAH patient population as compared to recent studies. Now not only did we meet our primary endpoint of improvement in PVR but we also had multiple indications that we were improving right heart function. We saw a statistically significant reduction in NT-proBNP, which is a key biomarker measure of right heart stress. The result is congruent with significant and clinically meaningful changes we observed in right heart parameters as measured by right heart catheterization and echocardiography. Now remember, though PAH is a disease that initially manifests in the pulmonary vasculature. Patients ultimately die of right heart failure. All of these end points, both individually and collectively indicate that seralutinib is having a clinically meaningful and positive effect on the right heart function. And that speaks to the real strength of our data, the striking consistency across all measures in this heavily treated patient population, with meaningful concordance pointing to the benefits of treatment with seralutinib. Now while we did not design this study to see a statistically significant improvement in exercise capacity, we did have a 6-minute walk result that favored seralutinib in our overall population. This result was more robust in patients with elevated risk score at baseline and reached statistical significance in functional Class III patients. We saw positive results in our overall study population despite a notable imbalance of functional Class II and III patients in the seralutinib arm as compared to the placebo arm. Class II patients outnumbered the more severe Class III patients by more than 2:1 ratio in the seralutinib arm compared to a 1:1 ratio in the placebo arm thus, an imbalance in the patient population. And just as importantly, we were able to generate these meaningful improvements in patients without seeing the side effect profile that was observed in the imatinib development program in PAH. Seralutinib with its molecular and PK characteristics alongside its inhaled route of delivery was specifically designed to avoid these side effects and we're very happy with the safety experience for the patients in TORREY. Seralutinib was well tolerated with a mild cough being the most highly reported adverse event. We didn't see the types of GI and fluid retention side effects that caused a high rate of early dropouts in the IMPRES study. We also saw zero subdural hematomas. Now before we dive in further, I wanted to quickly mention one thing. We are really excited about these data, and we believe that the more transparent we are with you about our results, the more excited you will be about the potential of seralutinib. So today in that spirit, we're going to provide a robust level of detail on the primary, secondary and exploratory endpoints as well as safety from the Phase II TORREY study. We'll take the first step into the details of the study as we turn to the next slide. So in our overall patient population, we saw a statistically significant 14% reduction in PVR and a 6.5-meter improvement in 6-minute walk distance that favored the seralutinib arm. Now let me tell you why I'm excited about the data we generated in TORREY. First is the consistency and the concordance observed throughout our entire data set and what that tells us about seralutinib's positive clinical effect. Consistent benefits of seralutinib treatment were seen across all of our pre-specified subgroups and all of the key hemodynamic and right heart endpoints we evaluated in the study. It's unusual to see improvements in right heart function in a 24-week Phase II study. The fact that we observed statistically significant improvements in these measures, some of which have never been reported in a Phase II trial, speak to the clinical meaningfulness of the impact that seralutinib is having on our overall patient population. And second, we observed enhanced effects on hemodynamics and exercise capacity in patients with more severe disease at baseline, similar to prior development programs in PAH. In functional Class III patients, we saw a statistically significant improvement in PVR of 21%, and a statistically significant improvement in 6-minute walk of 37 meters, which was similar to the walk improvement seen in functional Class III patients in the PULSAR study. In patients with elevated risk scores at baseline, we saw a significant 23% improvement in PVR and a clinically meaningful 22-meter improvement in the 6-minute walk distance. The patients who needed the help the most, those who were maxed out on available therapies and who were at higher risk of mortality at baseline saw a consistent positive effect from seralutinib treatment. These 2 factors, the concordance of our data and the enhanced effects we observed in sicker patient groups, give us confidence as we move forward into Phase III and in the potential impact that seralutinib can offer as a new mechanism of action for patients with PAH. Before we go deeper, I do want to set the stage a bit. PAH. PAH is a rare progressive disease that is characterized by vascular remodeling and it has no known cure. Many treatments have emerged over the past couple of decades, but far too many patients continue to suffer with progressive disease and ultimately succumb from right heart failure and other PAH disease complications. All current therapies target 3 vasodilatory pathways, and there's a high unmet need for therapies that address the underlying pathological mechanisms of disease. Seralutinib's anti-proliferative, anti-inflammatory and anti-fibrotic mechanism of action has the potential to do just that by directly addressing the biological pathways driving PAH as a potential reverse remodeling agent. Patients with PAH are classified by their physicians into functional classes based on how their disease affects their physical activity. But we know from registry data that functional class correlates strongly with risk status and predict survival which you can see in the bar graphs in the lower right-hand section of this page. In TORREY, we enrolled patients with either functional Class II or Class III disease, and as you can see, patients with functional Class III only have a 57% 5-year survival rate despite currently available treatments. Now beyond functional class, other parameters are being incorporated to better categorize risk for morbidity and mortality events. A number of composite risk measures have been developed and the goal of PAH treatment has evolved to helping patients achieve low-risk status, which has been correlated with reduced mortality risk. As mentioned earlier, we used the REVEAL 2.0 Risk Score that was developed by Dr. Benza in our study. So we're especially glad to have him on the call today to help interpret our results. So what treatments are currently available to help patients achieve those low-risk scores. And as I previously mentioned, there are 4 classes of therapies that target 3 vasodilatory pathways. Now what's needed are therapies that address the underlying proliferative, inflammatory and fibrotic causes of the disease. I think we, like the rest of the PAH community, are excited for the first of these therapies, sotatercept to potentially reach the market. We believe sotatercept is a promising drug that's poised to have a positive impact on the disease course for many PAH patients. But I want to relay to you one of the key messages that came from our steering committee as we reviewed this data with them. PAH is a progressive and heterogeneous disease. And unfortunately, there are no cures. The majority of patients are not currently meeting their treatment goals. And even with the introduction of sotatercept we expect that will continue to be the case in the future. The results we generated in the TORREY study suggest that seralutinib, with its novel and complementary mechanism of action, could potentially be an important part of the future treatment algorithm. And with that backdrop, let's go into the TORREY results. Now as you recall, the TORREY study was designed to enroll 80 patients, 40 per arm between seralutinib and placebo. Patients were required to have a PVR of greater than 400 dynes and a 6-minute walk distance of between 150 and 550 meters, among other entry criteria to enroll into the study. Subjects started on a 60-milligram dose twice daily and after 2 weeks escalated to 90 milligrams twice daily. The vast majority of patients on both the placebo and seralutinib arms were able to reach and stay on the 90-milligram BID dose throughout the study. Randomization was stratified by our primary endpoint of PVR at a cut point of 800 dynes. We assessed our primary endpoint of PVR during screening and week 24. ECHO assessments were performed at baseline, week 12 and week 24. Both PVR and ECHO assessments were evaluated in the study by a blinded central reader. We assessed 6-minute walk during screening, at baseline, week 12 and week 24. We collected NT-proBNP during most of the in-clinic visits designated by the green triangles on this slide at baseline, at weeks 4, 8, 12 and 24. After week 24, patients had the option to roll into an open-label extension study. So we ended up enrolling 86 patients, 42 in the placebo arm and 44 in the seralutinib. We had one patient discontinued treatment in the placebo arm and 7 in the seralutinib arm. We'll go into further detail for the reasons for discontinuation during a safety discussion, but there were no patterns in the adverse events that led to treatment discontinuation. And we think the imbalance we saw here was more likely a consequence of enrolling a small study. The patients that discontinued treatment on placebo stayed on study and completed all required assessments, so we ended up with 100% of patients that completed the study in the placebo arm, which is pretty remarkable for any 6-month clinical study. We were very pleased with patients -- the patient's adherence to their treatment regimen in the study, and the survey results collected suggested that patients were pleased with the convenience and the overall presentation of our dosage delivery via the dry powder inhaling. Approximately 90% of patients entered into the open-label extension after finishing the 24-week study. Now getting into the baseline demographics of the enrolled patients, we didn't have any surprises here. We enrolled a mainly female study. Just under 70% of patients came from sites in the U.S. and all demographic factors were relatively well balanced between the treatment arms. We were also very well balanced in terms of background therapy between the 2 arms, approximately 57% of patients in the study Were on background triple therapy in both arms, which mirrors the background treatment split from the PULSAR study. 29 subjects on each arm were on prostacyclin or prostacyclin receptor agonist which broke down evenly to 19 on parenteral prostacyclins and 10 on orals in each arm. That means that roughly 44% of our study population was on background parenteral prostacyclins, a high percentage when speaks to how heavily treated our patients were at baseline. Now turning to the next slide. You'll see that TORREY enrolled a very prevalent patient population, roughly 8.5 years since diagnosis. This was, as I mentioned earlier, a relatively less severe population baseline means as compared to other recent studies. Now if you just direct your eyes to the lower right portion of the table, you'll see that our overall baseline PVR for the study was 669 dynes or about 110 dynes lower than the population enrolled in the PULSAR study. The mean 6-minute walk distance was 408 meters or about 10 meters higher than the PULSAR study. To our knowledge, these metrics make TORREY an unprecedented study. The PAH clinical trial was the lowest baseline PVR and highest baseline walk distance to achieve a statistically significant result on its primary endpoint. The 2 arms were fairly well balanced in terms of PAH disease classification. The only substantial imbalance between the 2 arms is in the functional class section from the slide in the black box. Placebo arm was split nearly 50-50 between Class II and Class III patients. While the ratio was greater than 2:1 in favor of less severe Class II in the seralutinib arm. On the next slide, we'll go into a bit further detail on how this imbalance may have impacted overall results. Now this slide is a breakdown of baseline demographics by WHO Functional Class with the Class II patients summarized to the left and the Class III patients on the right. What you'll see is that surprisingly across multiple metrics, the functional Class III patients had more severe disease at baseline. If you directorize to the highlighted values, you'll see that roughly 60% of the Class III patients had elevated risk compared with 30% of the Class II patients. Moving to other metrics. There's a 70-meter delta between the mean baseline 6-minute walks and roughly 250-nanogram per liter gap for NT-proBNP levels between the 2 classes. And as we've seen in prior development programs, these differences are important. Patients with more severe disease typically have more room to improve than patients who are healthier and are protein sealing or floor effects on measures of disease. As happens sometimes in small studies, we had fewer of those sicker, easier to improve patients on the seralutinib arm. But thankfully, as you already know, this imbalance wasn't enough to prevent us from achieving a positive study. I'm now going to turn the call over to Dr. Roscigno. Aranda, Ghofrani and Benza, who will walk you through the top line results and their interpretations of those results in further detail. Rob?

Thanks, Faheem. This first slide shows the results for the primary endpoint of the TORREY study, which was changed from baseline in pulmonary vascular resistance at week 24. We observed a mean 21-dyne increase in the placebo arm and a 75-dyne decrease in the seralutinib arm resulting in a least squares mean difference between the 2 arms of just under 100 dynes or an adjusted reduction of about 14%. Importantly, these results reached a statistically significant p-value of 0.031. And recall where these patients started. The mean baseline PVR for the study was about 670 dynes. Accordingly, patients on the study had less room to improve than in prior studies such as the imatinib studies where patients came in with a mean baseline PVR exceeding 1,100 dynes. And as Faheem mentioned, one of the strengths of our data is its consistency. This is a forest plot that shows the point estimate for change in PVR for all of our pre-specified subgroups. If the diamond is to the left of the zero line, the estimate favors seralutinib and if it's to the right, it favors placebo. As you can see, every point estimate favors seralutinib, which is really exciting to us indicating that seralutinib is having a positive impact across all these pre-specified patient groups. There are several that reach a statistically significant result for seralutinib, but I want to point out 2 towards the bottom of the slide related to baseline disease severity. Third from the bottom is functional class. In functional Class III patients, we saw a 137-dyne difference favoring seralutinib with a p-value of 0.04. The bottom row shows patients with an elevated REVEAL 2.0 Risk Score of greater than or equal to 6 at baseline. This subgroup includes both functional Class II and III patients. This was our best-performing subgroup with a 168-dyne difference favoring seralutinib with a p-value of 0.0134. This next slide provides more detail on those 2 pre-specified subgroups. On the left is change in PVR by WHO Functional Class and on the right is PVR change by REVEAL 2.0 Risk Score. You will notice that seralutinib has a positive effect in all of these pre-specified subgroups, decreasing PVR and functional Class II or III low or elevated risk patients. In Class III and elevated risk patients, seralutinib has an enhanced effect. As you can see, the placebo arm PVR increases in both of these subgroups. This is what you might expect as higher functional class and elevated risk score, both correlate with disease progression and mortality risk. Seralutinib treatment not only prevents that progression, it also drives improvements in PVR in both of these patient subgroups. Ultimately, this leads to more pronounced, statistically significant treatment effects. So what we see when we look into the components of PVR is that much of the benefit observed in TORREY was driven by a significant reduction in pulmonary artery pressure, which is shown here on the left. We observed a highly significant reduction in mean pulmonary artery pressure, generating a p-value of 0.0094. We do also see a directionally positive improvement in cardiac output on the right, though it doesn't reach statistical significance with a p-value of 0.326. Contributions from both drivers of PVR confirm that we are achieving a meaningful change in resistance as represented by the overall change in PVR. Now I think this is a good time to ask Dr. Ghofrani to provide his thoughts on the significance of our hemodynamic data. Ardi?

Yes. Thanks, Rob. I hope you can hear me well. Well, I think the first and most important finding is that the study met its primary endpoint and we all realize how important this primary endpoint for the patient is by means of reduction in the pulmonary vascular resistance, which is actually one of the main drivers of disease progression. So it's a desirable and achievable goal to reduce pulmonary vascular resistance with this novel therapy. The response that we've seen is on top of an optimized background standard therapy. And having said that, it is a therapy that has maxed out for a substantial proportion of these patients because it includes triple combination therapy with parenteral prostacyclin. This is very reflective of the prevalent nature of the disease in the currently study patient population and also shows that they have been treated with the best standard of care that is currently available. And nevertheless, had a disease, which is still chronic and has the potential to progress. What we also appreciate was that in the higher-risk patients, it is very important to achieve improvements because they need these improvements in pulmonary hemodynamics and other indicators of disease severity more acutely. And therefore, it's a very encouraging finding to see that the magnitude of response in the more severely affected patients is bigger not meaning that those with less advanced disease had no effect. But the margins and the magnitude of response was clearly higher in the more severe patients. In the field of pulmonary hypertension, it's known for a very long time that there's a concordance of the PVR with right heart function. And the results that we saw here are to be discussed as very important findings. We've seen -- unlike with sotatercept in the previously mentioned PULSAR study, in the TORREY trial, directional improvement of the cardiac output. However, it is good to say or it's fair to say that the cardiac output in the actively treated group was more or less preserved while it deteriorated even in as short as 6 months in the placebo group. So it's a desirable directional change of the cardiac output with seralutinib. Regarding the historical perspective, the safety profile that was demonstrated with seralutinib shows that it is safe, effective and it will be a novel mechanism of action by means of tackling the disease with an inhaled tyrosine kinase inhibitor and introducing these novel complementary pathways into the treatment armamentarium of pulmonary hypertension is very exciting.

Thank you, Dr. Ghofrani. Now let's turn to our secondary endpoint of change in 6-minute walk distance. As you recall, the TORREY study was designed as a hemodynamic study first and foremost and was not designed as an exercise capacity improvement study. We were not powered to see statistical significance but we wanted to see trends that would help us design our Phase III program. We had a high entry criteria cut-off for the study at 550 meters. We went that high because of the COVID-19 pandemic and patient recruitment competition driven by the sotatercept Phase III STELLAR study. Despite the challenges that conducting a 6-minute walk test during the pandemic presented, we did see some interesting results, particularly in patients with more severe baseline disease consistent with prior studies in PAH. On the next slide, you'll see our results in the overall study population. As is often seen in clinical studies, we saw a placebo effect represented by the gray line early at week 12, which wanes by week 24. Seralutinib group, shown in green, has an early improvement that remains relatively constant at about the 15-meter mark resulting in a difference that favors the seralutinib arm by 6.5 meters at week 24. With the overlapping error bars at week 24, this isn't a statistically significant result. But again, we weren't powered on this endpoint for statistical significance. And overall, patients entered the study at a very high mean baseline walk distance of 408 meters. On the next slide, we've isolated 6-minute walk data for the functional Class III patients, you'll recall, these patients came in with a 6-minute baseline walk distance of about 370 meters and in general, are much more similar to the overall patient population that used to be enrolled in PAH clinical studies. And what you see is a result that is very similar to what we used to see in successful vasodilator studies on 6-minute walk distance, a placebo arm that holds steady and eventually declines at week 24 and an active arm that improves over time, approaching a 25-meter improvement at week 24 resulting in a clinically meaningful and statistically significant 37-meter difference from the placebo arm. As Faheem mentioned earlier, this effect has dampened into overall population by the imbalance in Class III subjects in the treatment groups, which you can see in the ends at the bottom of the graph. As with PVR, we broke out the differences in 6-minute walk at week 24 for both functional class and risk score. Starting with the left most group on the slide, the functional Class II patients. You'll see that seralutinib patients do have an average increase in walk of 14 meters at week 24. But what is more notable is what we see in the placebo group. This group of patients began the study with a mean walk distance of 455 meters, an increase of 30 meters means that they ended the study at over 485 meters which is a very high mean result for a relatively large PAH subgroup in a clinical trial. This played a role in muting our 6-minute walk distance treatment effect size in the overall patient population. The next set of bars to the right depicts results for the functional Class III patients we covered in further detail on the prior slide. Again, to summarize those results, we saw a 37-meter statistically significant difference in this subgroup. Looking at the right side of the page, we do also see a nice improvement in elevated REVEAL risk patients on seralutinib with an end-of-study improvement of 27 meters. This results in a non-statistically significant but clinically meaningful placebo-adjusted improvement of about 22 meters. We are confident with these results in hand, we can properly power a Phase III study on the 6-minute walk endpoint. Now we're going to turn to some very encouraging and exciting data Faheem mentioned earlier. These data confirm to us that not only was seralutinib having an impact on the disease course for patients in TORREY, but that it was a clinically meaningful impact. This first slide shows mean changes in NT-proBNP over weeks 4, 8, 12 and 24. As a reminder, NT-proBNP is a peptide that is released by cardiomyocytes when they are subjected to stress and it is an important biomarker of right heart failure. And what you can see is that there is an early separation of the curves at week 4, which becomes significant by week 12 and is highly significant at week 24. It's fantastic to see such a sustained decrease in NT-proBNP levels throughout the blinded study period even as the placebo arm increases over time. So we've been calling this our hemodynamic and ECHO scorecard. And I think it gives you a clear picture of the concordance of the results we've been referencing. Across all of these hemodynamic and right heart functional and structural echocardiographic endpoints, we see either statistically significant changes or results that are directionally favoring seralutinib. For many of these end points, we saw improvements by week 12, which aligns with our early improvement in NT-proBNP. I will first focus your attention on right atrium area. This measurement is in the current European Society of Cardiology European Respiratory Society guidelines and is a known prognostic indicator of disease worsening in right heart failure, which again is the leading cause of disease-related mortality for PAH patients. As the patient's disease worsens, the right atrium enlarges, so seeing a statistically significant decrease in right atrium area relative to placebo is a critical finding, indicating a positive change in disease course for patients on the seralutinib arm. Likewise, right ventricle free wall strain is another metric that has been associated with mortality and PAH. Strain is a measure of right ventricle deferation and when we see a significant decrease in strain relative to placebo at week 24, it means we're seeing an important movement -- improvement in right ventricular function. We also saw a significant improvement in pulmonary artery compliance, which gives us information about the coupling of the right ventricle to the pulmonary vasculature, and improved pulmonary artery compliance indicates a more normal physiological interaction between the right ventricle and pulmonary circuit, indicating a potential remodeling effect. So we've heard from many of you what you believe are relevant data in a PAH population, including hemodynamics, 6-minute walk distance and NT-proBNP. Looking at this slide, we have an analysis looking at change in REVEAL 2.0 Risk Score among patients on seralutinib and placebo. On the left-hand side, you'll see a bar chart. The 2 comps on the left of this chart represent the proportion of patients that improved by at least 1 point on the REVEAL Risk Score. The 2 bars in the center of the chart represents a proportion of patients whose risk score did not change during the study. The paired bars on the right-hand side of the bar chart represent patients that worsened by 1 point or more. Risk score changes are critical in their impact to lives of PAH patients. REVEAL is a broadly used tool that predicts survival. A single point improvement on the REVEAL 2.0 Risk Score confirms a substantial mortality benefit to a PAH patient. As you can understand, ultimately, what patients and clinicians are the most focused on above all else is long-term outcomes, including survival. I want to point your attention to the left-hand side of the bar chart. As you can see, the majority of patients on the seralutinib arm experienced an improvement of at least 1 point in risk score over the course of the study. This result narrowly met statistical significance as compared to placebo with a p-value of 0.059. You can also see that a much smaller portion of patients on seralutinib demonstrated worsening of risk score on the right-hand portion of the bar chart. As compared to the 40% of placebo patients who saw at least a 1-point increase in risk score. These data further corroborate the changes we're seeing in the right heart. As you can see in our heavily treated study population, seralutinib made a difference on both disease improvement and disease worsening. Seralutinib patients have almost 2.5x the odds of achieving a risk score improvement compared to placebo patients. When you look at the totality of the efficacy data we presented today, these are the most clinically impactful data. Our team is thrilled to be sharing it with you. I'm now going to invite Dr. Benza, as a leading cardiologist in the pulmonary hypertension field to provide his perspective on the meaningfulness of the data we've presented thus far on right heart parameters and the REVEAL Risk Score reduction. Ray?

Thanks very much, Rob. I appreciate the opportunity to express my opinion about what I feel is very exciting results from this new class of agent that could be used in our patients who have remained ill despite maximally treated therapy. From most of you that do not know how the REVEAL score work. It is a multiparameter score that comprises not only demographic features about the patients, but also really gives a great description of the totality of the risk by including things as vital signs, comorbidities, walk in BNP as well as functional class as well and in addition to hemodynamic parameters like the PVR, right atrial pressure, echocardiographic features of high-risk and pulmonary function features of high risk, including low DLCO so that these scores really give you a composite at how these patients will fare that is statistically much more relevant than looking at individual parameters like 6-minute walk test or BNP or functional class. And this is the way the community is now segregating our patients and allowing us to drive patients to what we feel is the ultimate goal of keeping low risk, which is defined by a REVEAL risk score of less than 6 or less than 5% mortality at 1 year. The fact that we can show this type of reduction in score in a heavily treated patient and a less ill patient is quite remarkable and attest to the -- I think, the strength of this new class of drug on top of the armamentarium that we're already treating these patients with. I think when you look at the differences in the 6-minute walk test, it's important to note that the less ill patients primarily have a more prominent placebo effect than more ill patients, and that's distinctly seen when you carve out the Class III patients who really have a remarkable change and a meaningful change in 6-minute walk test, as many of you remember, a change of anywhere between 25 to 35 meters in a 6-minute walk correlates to improved quality of life and also improvement in long-term survival. So that's a very important finding. The fact that we can see improvement in right ventricular remodeling as measured by a very sensitive test, which is the right ventricular strain rate is also very important. Again, we were treating a majority of low-risk patients in this trial and to see a reduction and improvement in right ventricular strain means we're seeing very early signs of right ventricular remodeling in the positive direction. By coupling, we mean that the right ventricle and the pulmonary vascular system are in sync with one another. The right ventricle has enough power to eject blood into a high-resistant circuit is a very important finding, it's showing improvement in compliance, again, with a heavily treated population is quite significant. I think these results are all very consistent and objective and suggest that in the properly used patient, which is those at higher risk that this is going to be a very important part of our armamentarium. And I think it's important to point out that the majority of the patients that we see in clinic are people who had elevated risk. I mean the bulk of our patients are the intermediate risk zone, which is a REVEAL score of 7 to 8 and the fact that we included patients with REVEAL score of 6 in this patient showed an improvement in them suggest that even these people at low risk for mortality, that we're going to have a low risk for morbid events in these patients, which is reflected by achieving the risk score less than 6. So again, very excited about the results of this trial and encouraged in these early results.

Thank you, Dr. Benza. I'm now going to turn things over to our Chief Medical Officer, Richard Aranda, to take you through the safety and tolerability profile that was observed in the study. Richard?

Thank you, Rob. Moving into our discussion of safety, it's helpful to remember that seralutinib's design and route of administration was meant to avoid the systemic adverse events that were observed in the IMPRES study of imatinib that was conducted over 12 years ago. The high rate of gastrointestinal and fluid retention events in that study led to a high dropout rate. They also observed a high frequency of severe adverse events, such as subdural hematomas in their study. We are pleased to see that in the TORREY study, seralutinib had an improved safety profile, including no subdural hematomas. On this slide, we present the events observed at a significantly higher frequency in the imatinib arms of IMPRES provide the event rates from the IMPRES study in the center and then to the right, report the number of these events observed in the TORREY study. As you can see, there were far fewer events in TORREY. The events that did occur were much more well balanced between the seralutinib and placebo arms than in IMPRES. We believe seralutinib's distinct safety profile can be attributed to its kinase inhibition profile, low oral bioavailability, route of inhalation, limited systemic PK and fast systemic clearance providing significant differentiation from the imatinib tolerability profile. So what events did we see? On this page are all events reported in 5% or more of patients on seralutinib. The top event reported is cough, which is at about 40% and is consistent with the results seen with dry powder inhaled drugs in other therapeutic areas. Of note, the vast majority of these cases were reported as mild. In the treatment arm, 17 where the 19 case -- of the 19 cases were mild, with 2 cases being reported as moderate. Beyond cough, most of the other AEs on this table are the typical events and events rates that one would see in any randomized clinical trial or our events related to the patient's underlying pulmonary arterial hypertension. Of note, all of the events on this slide were reported as mild to moderate in severity. Let me add that 8 severe adverse events were reported during the study, mainly related to the underlying pulmonary arterial hypertension and only one was also reported as a serious adverse event, and that was reported as possibly or probably related to study drug by the investigator. This was a case of hemoptysis in a patient in the seralutinib arm. This patient had underlying interstitial lung disease in lupus and presented to their doctor after finding a small -- very small amount of blood in their sputum after coffee. The patient was briefly hospitalized for observation, had no further events and discharged the next state. Hemoptysis is known to occur at low background rates in PAH. Here is a list of the adverse events leading to discontinuation that Faheem mentioned earlier. As you can see, there's no pattern in these events. The first patient with lower abdominal pain had issues with their prostacyclin pump and ultimately decided to come off the study. Two patients who discontinued in the seralutinib arm did so more for personal reasons. One patient had dry mouth due to underlying Sjogren's and this was exacerbated during the study. The other patient had a history of lupus and experienced cough. We had 2 patients discontinued treatment due to liver enzyme elevations in the seralutinib arm and one in the placebo arm. In general, with regards to our safety and tolerability profile, I don't think we could ask for much more. We knew that cough would be frequent with the dry powder inhaler but the overwhelming majority of the cases were reported as mild and did not lead to discontinuation except for the case mentioned previously. Further, we avoided high rates of gastrointestinal and fluid retention adverse events, which we believe will be a differentiating factor for seralutinib as we move into Phase III. With that, I will turn things back over to Faheem, to summarize the results and provide next steps. Faheem?

Thanks, Richard. Look, I know we've thrown a lot of data at you this morning. We felt it was important to share more than our peers have in the past because PAH is a multifactorial disease where there are many components that can lead to both improvement for patients and conversely deterioration. When we review these data with our key opinion leader steering committee, I was struck by the concept a number of KOLs emphasized. They don't think of the benefits of PAH treatments in terms of relative reductions but rather by where those treatments can get their patients to with respect to their overall disease state and their risk of progression. And in that vein, I'm excited that seralutinib was able to get patients on drug to a disease state that's comparable to the end of the study, disease state that sotatercept achieved in the blinded portion of the PULSAR study, in terms of PVR and other measures of the disease. So what does this all mean for seralutinib going forward? In TORREY, we had a positive Phase II study with striking, concordant, statistically significant improvements across multiple hemodynamic biomarker and right heart structural and functional endpoints. We saw all of these effects in the absence of the safety and tolerability issues seen with [indiscernible], which we believe will be a differentiating factor going forward. We saw improvements in 6-minute walk distance, especially in sicker patient groups, which we believe provides a clear path forward for a Phase III program in PAH. And lastly, and this is something we'll spend more time on educating the market in the coming year. We believe our results in TORREY and the mechanistic rationale open up a great opportunity for us in Group 3 pulmonary hypertension. Group 3 patients have tremendous unmet medical need with inhaled TYVASO as the only approved drug. So before briefly going over next steps to the program, I just want to stop and thank every patient and their families and all of the TORREY study investigators that participated in the TORREY study. We're so pleased to have a positive study, and we quite literally could not have gotten there without you. We also want to thank Dr. Ghofrani and Dr. Benza for joining us today and providing their invaluable perspective on the data we've generated in TORREY. I also want to thank everyone at Gossamer, who worked on the TORREY study and in any way touched the seralutinib program over the years. I personally continue to be inspired by your dedication and your commitment to patients and know this will carry us forward as we advance seralutinib into Phase III development. And briefly on next steps. Our team is working diligently to get in front of global health authorities in the first half of next year. And after we conclude those meetings, will go into further detail on our plans for PAH Phase III, which we expect to begin in the second half of next year. And lastly, as I alluded to earlier, we plan on beginning parallel development of seralutinib in Group 3 PAH later next year and into 2024. We'll have further details for you on what those studies would look like next year. With that, we're looking forward to answering your questions, and I'll turn the call over to the operator to begin the Q&A session.

Yes. Thank you. At this time, we will begin the question-and-answer session. [Operator Instructions] And the first question comes from Ellie Merle with UBS.

Okay. Maybe just first for the physicians on the call. If you could just elaborate a little bit on your thoughts on the results of where you might use seralutinib relative to sotatercept from this data?

Yes. Let's start with Dr. [indiscernible] maybe you could take that first one.

Yes. I think that we are all excited that there are novel therapeutics entering this space. And most importantly, because they are coming from a complementary signaling pathway background to those treatments that we currently have available. Among such, seralutinib obviously, is the most welcome addition to our therapeutic armamentarium, and it will be used as soon as it is approved according to the label, obviously, but it will be used in patients with PAH. And we've seen from the study data that there are patients who react very well and have a very significant clinical response to the treatment and tolerate the therapy on longer term. But there are also patients who have not such kind of impressive results as we've seen in the clinical studies in the average responses and there will also be patients most likely who may develop tolerability issues or safety issues over time. And of course, these patients will qualify for treatment with seralutinib. But one other thing that I'm very excited about is that both the signaling pathways for seralutinib and sotatercept are complementary to each other, which gives space to even combining both treatments, provided they both enter the market soon. And there is good preclinical evidence that a combination of this kind of tyrosine kinase inhibitor with an activin ligand trap such as sotatercept may even cause some synergistic effect on the pulmonary vasculature and the right ventricular performance. So long story short, there is room for both therapies in the field of pulmonary arterial hypertension therapies.

Dr. Benza, anything further to add to that?

And I agree with Dr. Ghofrani. I think there's plenty of room for both of these medicines, particularly with the proposed synergistic effect of both of them. I think we'll be using these in the majority of our patients who are above the very low-risk patients, which are really the minority of the patients we see in the clinical environment. So I think it's going to fit in well with our other armamentarium and be complementary to some of the newer therapeutics that are coming out.

Great. And just a quick follow-up question. I know you mentioned that next year, we'll learn more about potential Phase III design. But just in light of this data, can you elaborate a little bit maybe on how Phase III design could look maybe relative to the design of the Phase III that we saw from sotatercept? And I guess, any specific inclusion criteria or anything that like you think high level might differ relative to the sotatercept Phase III? And in particular, you mentioned excitement around potential to look at a broader population, even sort of like Class III, maybe like your thoughts there and how you would approve potential studies in the Class III patients or say, like interstitial lung disease things.

Yes. So first off, as it relates to Phase III for PAH, I think the results suggest kind of some obvious the design elements, 6-minute walk will clearly be the primary endpoint. We'll be looking to increase the baseline criteria for patients coming in on both 6-minute walk and PVR. I think the results certainly suggest that. I would say that the overall trial design probably would look reasonably similar to sotatercept, but Richard, our Chief Medical Officer, if you want to...

I think the results from the Phase II clearly have provided us some good insights and directionality on how we should approach our inclusion criteria for our next study everything that Faheem has outlined. In terms of ILD, we also think the results, particularly our NT-proBNP results are indicative that we could have an effect in Group 3. And in addition, we know the underlying mechanism of seralutinib may be applicable in the long part of ILD or the fibrotic part as well as treating the pulmonary arterial hypertension. But more to come on what we're exactly thinking about our ILD design.

And the next question comes from Andreas Argyrides from Wedbush.

Lots of data to [indiscernible] here. Could you just start with giving us a little bit more color on how many patients made it to the 90 mg BID dose. What was the distribution of doses? Did any of the patients have to down-titrate to 45 mg [indiscernible]?

Yes. So all but a handful [indiscernible] 90 mg BID dose about 5 or 6 stayed at 60. And -- I don't think we had anyone down-titrate to 45 to stay there.

Okay. Great. And then for Dr. Benza, with other KOLs that we've spoken to, an 18% reduction in PVR is typically considered clinically meaningful. How are you thinking about the reduction in PVR here? And yes, so we'll start that on a follow-up.

Maybe I can start, sure. I think that the magnitude of response that we saw here is something that needs to take into the consideration the baseline PVR at which the patient started off. So the baseline PVR was rather moderately elevated as compared to other clinical trials. It's clearly pathological. It's 6x above normal. But compared to other trial populations, it's a rather moderate elevated PVR. So the room for improvement, obviously, is less. But as Ray Benza pointed out, it's the absolute value that is reached with a novel therapy that is encouraging. So getting into the proximity of the 500 times with treatment already reduces -- is one of the key components of a reduced risk score for these patients. And then also, given the magnitude of response in the overall patient population, we should still also acknowledge that the percentage-wise reduction of PVR in the more sicker patient population in the Class 3's and those with higher risk scores was even more pronounced and was approximating 25% and some percent reduction, which then is also getting into proximity of results we've seen with other PAH-specific therapies and with seralutinib and with sotatercept in particular. So for me, it's an encouraging result. And the last piece of my comment is I was impressed by the subgroups that were shown in the forest plot. And every prespecified subgroup fulfilled the criteria of changing towards the right direction. And in the study sized like this and looking at forest plots, I'm not so much intrigued by the p values rather than by the direction of the change and it was all in the right direction. So I'm quite pleased with the results here.

Dr. Benza?

Yes. I think this is similar to what I already mentioned. I think thinking about the reductions in PVR relative to the patient population is extremely important in this trial. In a relatively low-risk patients, we really have a significant and impressive drop in PVR. I remember, when you drop the PVR below 400 times or 5 Wood units, you are prognostically important improvement in survival. So that's really important not to forget. And the fact that we've dropped the PVR by 25% in the more ill patients, that is getting to the point where we see a correlation between PVR and right ventricular modeling, so I think that's a really important point to remember that when we get a drop of around 25%, we're starting to see that correlation with improvement in right heart function. And I think that's what we're going to see in the correct population, which is the more intermediate and the high-risk patients that we're seeing more in our clinical environment.

Okay, great. Thanks. I'll jump back in the queue. Yes. Thanks.

And the next question comes from Joseph Schwartz with SVB Securities.

Congrats on the progress. I was wondering if you could help us understand why PVR was not imbalanced at baseline, but 6-minute walk, functional class [indiscernible] NT-proBNP were imbalanced? What explains this apparent disconnect? And what kind of things did you stratify for in TORREY? And will you be doing it any differently in Phase III?

So the TORREY study was stratified by PVR, so that really accounts for the difference. It was not stratified for the other factors you spoke about functional class or 6-minute walk distance.

Yes. And I'm sorry. Would it make sense to do additional stratifications in Phase III, given that the primary endpoint is expected to be 6-minute walk?

Yes, absolutely. Phase III will be a larger trial. So they'll be room to as 3 or even 4 potential stratification variables as opposed to smaller studies [indiscernible].

Right. Okay. That's helpful. And then -- some investors have noticed that 6 patients were missing from the seralutinib arm in the pulmonary arterial pressure, which is used in the PVR calculation. And I was wondering if you could explain the imputations you're making for PVR on Slide 19. And which imputation was prespecified and used? And is that similar to the convention for other PAH trials? And did you do any sensitivity analyses to evaluate other methods?

Yes. So for the primary endpoint, we have prespecified a multiple imputation method. There were 50 imputations performed, and then there's a roll-out procedure to get to our estimates. So for those 6, we multiply imputed our primary endpoint. This is a very common procedure done across many different therapeutic areas, including PAH. Multiple sensitivity analyses were performed. I can think of off the top of my head, at least 12 that I did and every single one of them gave the same answer with a significant p value and an estimate between a difference of 95 to 102 or 103x, including observed cases analyses, analyses with different [indiscernible] model and things like that. So at the end of the day, the message here is that the missing data, the 6 missing values for PVR did not affect our results in any way. What you saw on the [indiscernible] we didn't impute for all those components, so we're showing you the actual values and especially since the missing data didn't affect our primary endpoint, it's reasonable to show just the data we collected after that point.

It's very helpful. Thanks for the color.

And the next question comes from [indiscernible].

So just in terms of your Phase III study, so since the trial is not expected until the second half of next year, which is getting pretty close to potential sotatercept approval. Are you concerned the FDA may ask you to include sotatercept in the study? And also, how do you think sotatercept's potential adoption may impact your Phase III enrollment?

Yes. So as it relates to the Phase III study, we would expect to see sotatercept on the market in the U.S. by end of '23, beginning in '24. Clearly, we will be and are actively working now on getting ourselves prepared to be enrolling not only in the U.S., but also quite extensively in Europe where sotatercept won't be available for some period of time. So in the context of patient enrollment, we feel pretty comfortable given the results that we have here that we'll be able to enroll that Phase III study in a similar time frame at a similar clip to what you saw with sotatercept. As it relates to what the regulatory authorities will be saying about enrolling sotatercept patients, obviously, we've got to wait to our end of Phase II meeting to kind of get the full understanding there. So I won't jump ahead of kind of what their expectations are. But having said that, it's reasonable to assume that some portion of our Phase III will include sotatercept patients. And as Dr. Ghofrani and Dr. Benza said, we think these mechanisms are quite complementary. And this is a disease that is treated through combination approaches. And so that could be quite exciting in terms of potentially having that type of combination breakthrough what I call kind of the glass ceiling of effect that we see in these studies. It's really interesting to see where we ended up on PVR across the entire patient population and our subgroups and where we ended up on 6-minute walk ended up being in the same neighborhood, the same ZIP code as where the sotatercept patients ended up being. So unfortunately, we keep hitting this in this disease. We keep hitting this ceiling as it relates to 6-minute walk and floor as it relates to PVR, hopefully, through rational combinations, we can actually produce better results.

Okay. And just one more question. In terms of the background therapy in TORREY, can you just talk about to what extent patients were able to vary the background therapy? I think one of the doctors mentioned the background therapy was already optimal at baseline. I'm just trying to understand did placebo patients increased dosing? And if they did, what proportion of that did?

So background PAH-specific therapies that you're speaking about, were to remain constant throughout the trial, to not add a compounder. If a patient had a clinical worsening event, that is required hospitalization for worsening PAH or a change in one of those medications, then that event was recorded and was taken into consideration when analyzing for the primary endpoint. So there were no changes that occurred in either arm outside of a well-documented clinical worsening event.

Okay. And sorry, just one final question. It's about the single patient with ALT and AST increase. I mean what grade of ALT/AST increase was that? And what was the bilirubin for that patient?

Yes. The -- you're talking about in the active arm because we had the single patient in the placebo?

Yes. So there is discontinuation in the active arm.

Yes. So they're above 3x and bilirubin was normal in that particular patient.

And the next question comes from Paul Choi with Goldman Sachs.

We appreciate all the details, but I was wondering if you could maybe comment on either the numerical or magnitude of changes with what you saw in terms of the WHO functional class assessment at week 24, any numerical trends or commentary you can make on the magnitude of potential changes there? And then I have a follow-up question.

Just to clarify your question, the magnitude of the changes for functional Class 3?

Any percentage of patients who were able to change functional classes? Or any assessment along those lines potentially?

Yes. So functional class, obviously, is a very applied and kind of -- there's only forward levels. So it is a challenge to change them, but we did have 7 patients to drop and improve their functional class in the [indiscernible] and I believe the same number in the placebo. But more importantly, we had 6 patients in the placebos actually had their functional class worsen, including some that moved to functional class IV. And we had none of that happened in the treatment group. So 14% of placebos got worse and here's a 0% in the [indiscernible]. So basically is a maintenance or improvement in functional class in the treatment group versus sort of more variability in the placebo in terms of improvement maintenance or getting worse.

Okay. That's helpful clarification. And then I wanted to maybe ask in terms of the study design and implications for Phase III, you did have patients titrate up from the 60 mg dose to 90 mg. Any thoughts given the safety profile that's been demonstrated here in the Phase II with just going with the 90 at the start that which you could have a potentially beneficial effect in terms of the treatment effect size?

Yes, sure. Just to provide some context, part of the reason we did the titration was for training of patients to use the DPI appropriately. So at this time, we're still thinking about incorporating the rapid titration that we use in our Phase II into our Phase III program. And we're still waiting for our pharmacokinetic data and which will inform us further on -- around our dosing and we need to do with respect to further adjustments in that. But as we presented, we're really quite pleased with the benefit-risk profile of the 90 milligram.

And the next question comes from Carter Gould with Barclays.

I guess first for the doctors on the call. We just came away from the [indiscernible] in which there was a lot of discussion around the challenges in conducting PAH studies going forward? And I guess, can you maybe just speak to the practicality of enrolling the potentially functional class III population with narrow or 6-minute walk distance criteria going forward? And then maybe for the company, just how you're thinking about managing expenses here now until you have clarity on that Phase III design, if there is anything you can do on that front?

Yes. So I'll turn it over to Dr. Ghofrani, but before I do, I don't think it would be our intention to just enroll a functional class 3 Phase III study. We'd be looking at patients certainly a functional class 3, but also functional class 2 with a higher risk score and probably higher walk -- lower walk baselines and higher PVR baseline. So it would cross both functional class 2 and 3. But Dr. Ghofrani, would you like to comment?

Yes. I think that you mentioned something very important, Faheem, which is that the risk and disease severity is multifactorial thing in PAH patients. It's not only defined by functional class or by the hemodynamics or by risk score, but it is really a multicomponent assessment. And I believe that with adjusting some of the entry criteria in the Phase III program, nevertheless, being open to include patients in various functional classes that the recruitment will not be so much of an issue, particularly as the trial is going to be performed as an international multicenter trial, and there are many regions in the world where patients are less fortunate to receive many, many background therapies. There's a high motivation to be included into clinical trials for good reasons, but also because the care in clinical trials, even if you should be in a placebo group is much better than the real world out in the field. So the patients get the privilege to be treated in expert centers and so forth. So there are multiple motivations to be included into a trial. And even more so, if you have a Phase II study, which is already indicative of a good efficacy complementary to the given drugs available. So I don't think it's going to be a problem for the upcoming Phase III studies to recruit.

And Carter, on the balance sheet side of the question and expenses. We have -- we ended the third quarter with 300 million of cash on the balance sheet. We continue to deploy that capital to get ready for that Phase III as already just said, our ability to go to places, especially for example, in Europe, where we know the sotatercept approval time lines will be longer, give us an opportunity to access those patients quickly. But clearly, we're very focused on making sure that we can deploy the capital appropriately, be in a position to get the study up and running. And certainly, as we think about the totality of expenses at Gossamer, we're very, very focused on continuing to support seralutinib program.

And I also think that there's clearly the potential for some interesting partnership options as well.

And the next question comes from Olivia Brayer with Cantor Fitzgerald.

I want to ask a follow-up on the dosing strategy for Phase III. Could there be an opportunity to actually increase the final dose level or even look at the 2 dose arms in a registrational program? I know you guys have talked about the 90 mg, but given the clean safety and the data today, just wondering how you guys are thinking about moving forward there?

Yes. I just want to reiterate one thing. I think, first of all, as Faheem mentioned, we need to assemble our data and then speak with the regulators because, obviously, they'll have an opinion on this, and that's the plan. And as I mentioned, we are waiting for our pharmacokinetic data that will inform us on our population PK, if you will, and what we're seeing. And that will guide us, obviously, if we should go beyond the 90-milligram dose that we currently use. And so I think you're absolutely right. We have a good benefit-risk. And once we get that data, we'll be able to more precisely assess that question.

And the next question comes from Yasmeen Rahimi with Piper Sandler.

Quite a bit for you. So given the encouraging data, you see the stock is trading at cash, I think fundamentally, it comes down to a few questions. One is, what is the possibility -- what is the probability of a successful Phase III, within the commentary you made that you will be enrolling Class II, Class III patients, 6-minute walk test, modifying some of the baseline characteristics in favor of a higher PVR and lower 6-minute walk test? I would like the doctors to kind of comment on probability of success in translation into a successful Phase III. And then the second question is how competitive is this product now, I think in terms of a bit in the current marketplace? And then the third aspect is can we really drill down a little bit more when we say higher PVR, like what would be an ideal PVR inclusion criteria, what would be a good 6-minute walk test, lower number that you would want to stratify patients on? So if you could provide some commentary on that, that would be really helpful.

Maybe Dr. Benza, you can start off.

Yes. I think by using a risk score instead of functional class, we get a much more objective baseline characteristic of the patient as opposed to just using functional class. So using REVEAL risk score of 6 or greater gives us an appropriately risk patient for our Phase III, particularly since these patients who more likely have a lower walk and a higher PVR. Remember, a walk of greater than 440 is associated with a good prognosis. And so trying to adjust for that using the risk scores within the range of 6 or greater usually will give us the population where we would expect to see a significant improvement in the walk-in just because they're objectively scored at a higher risk. I think using the scores as our -- as one of our enrichment characteristics will allow us to really have the type of patients we want that will manifest the important positive results and magnify the ones that we've seen here in the Phase II.

[indiscernible] Go ahead, please.

I was going to say maybe Dr. Ghofrani wants to add, because I think your other question, Yas, was about the profile, how competitive is the profile?

Yes. I think that Ray mentioned some very important points. And in addition, maybe regarding the competitiveness of a clinical trial in the Phase III program with sotatercept, I believe that the company is well advised to mainly approach high-volume expert centers around the world in which the current priorities are for original trials. For trials that investigate on novel signaling pathways rather than in times when very few really exciting complementary therapies or classes of drugs were investigated, many of the sites, particularly in the Western World, were entertained by useless, what I call useless Phase IV studies of drugs, which were already well characterized, approved and very well known. Nowadays, I think the high-volume expert sites, they prioritize original novel and innovative trials and amongst such will be the sotatercept study. So I have little doubt that the seralutinib study, I'm sorry, I have little doubt that this [indiscernible] once the Phase III program is initiated. And regarding the segment and pathway, it is complementary to what we have. It is complementary to sotatercept. So that's one of the main motivations, knowing that during the [indiscernible] development, the proof of principle for the utility of TKIs has already been delivered on the efficacy side. But with this novel TKI, which is now also applied on an inhaled route, I think we have advanced profile and advanced safety profile as well, which qualifies for further development.

And then Faheem, one last question. Is there an opportunity to dose higher in Phase III [indiscernible]?

I'd say, Yas, that we're still evaluating all of the PK data. More to come on that. We definitely think there's a may be a possibility in that front, but I think we've got to understand the totality of the PK data that comes in.

And the next question comes from Brian Cheng with JPMorgan.

We have a question on your analysis on [indiscernible] and also cardiac output on Slide 22. The -- so is there a little arm to be looking at only 38 patients in the analysis compared to 44 patients in total for the treatment arm. So we're wondering if you can provide some color on the reason behind a discrepancy. And secondly, can you talk about the potential contribution of background therapies in your clinical outcomes. We're wondering if the shorter duration of stable background therapies compared to other trials may have contributed to the outcomes that we saw today, because I think that [ PULSAR was about 3 months and TORREY 4 weeks for the duration for a stable background therapy.

Yes. First -- well, I'll turn both questions over to Matt. I think we already answered the first question.

So we were missing 6 right heart caths at the end of the study. So the primary analysis on PVR, use an imputation procedure that was multiple imputation to account for that. So that's why the end is 44 in the primary analysis. However, as I mentioned earlier, all of our [indiscernible] analysis, including an observed case analysis showed the same results. So we did not impute the components or any other right heart catheter, which is why [indiscernible].

And then the background therapy?

Yes, I think can...

So the stable background therapy, the dosing was to remain stable for a shorter period of time. But the agents and the agents being used were to be -- were similar, so that with what impulse are. So yes, while there could have been a dosing change, the regimen itself had to remain the same. So we've looked at this very closely for changes in dosing effect in that prior period and really haven't seen that, that's what occurred in the study.

And the next question comes from David Hoang with SMBC.

Thanks for the call, really informative. I just had a couple of questions. I think most have already been addressed. But just in terms of some of the Phase III criteria as you mentioned that may be implemented. Do you believe that would flow -- any of those would flow over to the label, the drug label itself and potentially have some limitations there just in terms of labeling language? And then as we think about the ideal population for a Phase III trial, does that change your assumptions at all on the overall market opportunity for seralutinib?

Well, again, as we mentioned, the expectation in our Phase III trial will include both functional class II and functional class III patients. But as Dr. Benza mentioned, probably be qualified through the REVEAL risk score and bringing in patients who have quite frankly, more room to see improvement. So I would expect to see a label that includes both functional class II and functional class III. I'll kind of answer the next part of the question, but I'll also invite Dr. Ghofrani or Dr. Benza to comment as well. That in context of the profile of a drug like this, where you have such a profound effect on -- a potential profound effect on right heart and potentially impacting kind of how these patients die, which is right heart failure, there would be an expectation that we would assume that physicians would use this drug earlier in functional class II patients to prevent progression and certainly use it in functional class III patients to prevent mortality. But please, Dr. Ghofrani and Dr. Benza, please comment from your perspectives.

Yes, I can see that the Phase III study will be quite comparable to other Phase III programs regarding inclusion and exclusion criteria. And most likely, the label will not very much differ from recently approved drugs with the comparable inclusion criteria. Traditionally and as far as I have noted, raising lower limits for the 6-minute walking distance and/or capping the upper limits for it as well as having certain inclusion/exclusion margins for the PVR, usually doesn't influence the label. It's rather the functional class that has been allowed to enter, and it's the entities of disease subcategories that were included and the absence or presence of background therapies in general. But it won't be as specific to say if you have a lower PVR value of 450 [indiscernible], then it will be carried forward into the label. So I'm not concerned about that.

And the next question comes from Gavin Clark-Gartner with Evercore ISI.

First, did you see a correlation between baseline PVR and PVR reduction or 6-minute walk improvement in this trial?

I mean typically, in PAH trials, the correlation with change in PVR with change in 6-minute walk has been very weak. Across all of the studies that we've done, we first reported this with sildenafil. And so that correlation, yes, we've looked at it, but it's no different than what we've seen with other trials.

Got it. And was there a correlation between baseline PVR and PVR reduction?

We have Matt, our stats guy digging into the database on that one. [indiscernible] have it correlated with the components of that, so we'll have to get back to you.

We'll come back to you.

Okay. Sounds good. And maybe a second question. Did you see a benefit on time to clinical worsening, hospitalization, death or any of the other outcomes measured?

Yes. We saw -- there weren't a lot of events on time to come worsening. They were 62.

[indiscernible] seralutinib.

Yes, we had a -- there were 6 events in the placebo group and 2 in the -- I'm sorry, yes, 5 to 2, the [indiscernible] hazard ratio, a reduction of about 60%. It's very small numbers. So not significant, however, certainly numerically favored the treatment group in terms of 2 [indiscernible].

Okay. Got it. Maybe a last question on for me. How did the PVR or 6-minute walk benefit looks for patients on doublet versus triplet therapy?

I mean most of our patients are on triple therapy. And I don't think we've done that analysis on one background but that's a great question. It's a proof [indiscernible] integrated in less than 3 and it's in the generally pretty close. Greater there less than 3. Yes, very close, the greater than 3 with the higher value than less than 3. To the forest plot.

And the next question is a follow-up from Andreas Argyrides from Wedbush.

I know there are a lot of questions here. But just for clarity sake here. So the baseline PVRs are historically low in this trial when compared to other PAH trials. What can we expect in PVR for patients on background therapies just generally now? Are they much lower? And then with plans to go into functional class III patients in the Phase II, what does that look like from a baseline PVR and 6-minute walk distance?

I mean I think for Phase III on 6-minute walk distance, we would kind of shorten or lower the criteria that was used in TORREY. As we pointed out earlier, the TORREY criteria was a 6-minute walk distance up to 550 meters, and we would kind of go to more traditional 6-minute walk distance that's been included in other Phase III trials of the usual 50 to 450. So to really to pull down those higher walkers that are approaching the ceiling for 6-minute walk in the Phase III moving forward.

And I'd go one step further, again, recall, we conducted the TORREY study not only with the backdrop of COVID, which made things obviously difficult but secondarily, against the competition of the sotatercept STELLAR study, where we know that physicians were putting less sick patients into TORREY. Again, part of the reason why we raised the capital, we did to put the infrastructure in place to be able to get to some of those more normal patients that we see in typical Phase III studies to be able to do that before sotatercept is broadly approved throughout the globe, we think will allow us to get a more normal patient population for Phase III than what we saw in Phase II.

Great. Thank you.

And this concludes question-and-answer session. I would like to hand the floor to Faheem for any closing comments.

All right. Well, thank you very much, and thanks to all of you for participating on this call, all the great questions. And special thanks to Dr. Ghofrani and Dr. Benza for your time today. I'll just wrap this up by saying that we're ready to make a difference for PAH patients, patients that -- where the chances of survival are not great at 5-year context. We've got a drug that not only is active, but demonstrating a meaningful effect across both functional class II and III patients. We look forward to going into Phase III, and we look forward to staying in touch to talk about our progress as we move forward. Thank you, everybody.

Thank you. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect your lines.