aTyr Pharma, Inc. (ATYR) Earnings Call Transcript & Summary

Good morning, ladies and gentlemen, and welcome to the aTyr Pharma conference call to review results for the Phase Ib/IIa clinical trial of ATYR1923 in pulmonary sarcoidosis patients. [Operator Instructions] As a reminder: This conference is being recorded for replay purposes. It is now my pleasure to hand the conference call over to Ashlee Dunston, aTyr's Director of Investor Relations and Corporate Communications. Ms. Dunston, you may begin.

Thank you, operator. And good morning, everyone. Thank you for joining us today to discuss the results for our Phase Ib/IIa study of ATYR1923 in pulmonary sarcoidosis patients. We are joined today by Dr. Sanjay Shukla, our President and CEO. On the call, Sanjay will provide an overview of the results, including data that supports proof of concept, comments on next steps for the program. We will then open up the call to questions. Before we begin, I would like to remind everyone that, except for statements of historical facts, the statements made by management and responses to questions on this conference call are forward-looking statements under the safe harbor provision of the Private Securities Litigation Reform Act of 1995. These statements involve risks and uncertainties that can cause actual results to differ materially from those in such forward-looking statements. Please see the forward-looking statement disclaimer in the company's press release issued this morning as well as the risk factors in the company's SEC filings and included in our most recent annual report on Form 10-K and quarterly reports on Form 10-Q. Undue reliance should not be placed on forward-looking statements, which speak only as of the date they are made, as facts and circumstances underlying these forward-looking statements may change. Except as required by law, aTyr Pharma disclaims any obligations to update these forward-looking statements to reflect future information, events or circumstances. I will now turn the call over to Sanjay.

Thank you, Ashlee. Good morning, everyone, and thank you for joining us for our conference call. I'm delighted to be here today to discuss the results of our Phase Ib/IIa clinical trial of our lead therapeutic candidate ATYR1923 or 1923 in patients with pulmonary sarcoidosis, a major form of interstitial lung disease or ILD. Earlier this morning, we issued a press release summarizing the positive results for this trial. I'm very pleased to report that the trial met its primary safety end point. And the trial demonstrated a consistent dose response for 1923 on key efficacy end points and improvements compared to placebo, including measures of steroid reduction, lung function, sarcoidosis symptom measures and inflammatory biomarkers. Some of the key findings include 1923 was safe and well tolerated, with no signal of immunogenicity; steroid reduction of 58% overall from baseline and 22% relative reduction, compared to placebo, in steroid usage post taper in the 5 milligram per kilogram treatment group; complete steroid taper to 0 milligrams achieved and maintained for 33% of the patients in the 5 milligram treatment group compared to no patients in any other group; absolute improvement in forced vital capacity or FVC, a measure of lung function, at week 24 of 3.3% in the 5 milligram treatment group compared to placebo; clinically meaningful improvement over placebo for shortness of breath, cough, fatigue, and the King's sarcoidosis scores for lung and general health, also in the 5 milligram per kilogram treatment group; and dose-dependent trends of improvement compared to placebo in key inflammatory biomarkers, compared to placebo, including IL-6, MCP-1, interferon gamma, IP-10 and TNF, as well as key markers of sarcoidosis, including ACE, IL-2 receptor and serum -- SAA, with tightest control in the 5 milligram per (sic) [ 5 milligram per kilogram ] treatment group. We're very excited with the results of this study, which provide the first clinical proof of concept for 1923 and a neuropilin-2 targeting compound as well as validation for our tRNA synthetase biology platform. The consistency in dose response and clinically meaningful benefit observed, along with 1923's favorable safety and tolerability profile, give us great confidence that 1923 could be a transformative therapy for pulmonary sarcoidosis patients. Based on the results of this study, we plan to meet with the U.S. Food and Drug Administration to present these data and our plans for subsequent clinical development and path to registration for 1923 for pulmonary sarcoidosis, and we expect to initiate a registrational trial next year. Before we get into some more details of the results, let's discuss a bit of background of 1923, which we believe is a potential first-in-class immunomodulator for pulmonary sarcoidosis and other forms of ILD. 1923 is a novel Fc fusion protein based on the naturally occurring splice variant of the lung-enriched tRNA synthetase HARS fragment that down-regulates aberrant immune responses in inflammatory disease states. 1923 has been shown preclinically to down-regulate inflammatory cytokine and chemokine signaling and reduce inflammation and fibrosis. Neuropilin-2 or NRP2 is up-regulated on key immune cells known to play a role in inflammation and is enriched in inflamed lung tissue. 1923 binds selectively to NRP2 and therefore has the potential to normalize the immune system, serving to resolve inflammation and prevent progressive fibrosis, thereby stabilizing lung function and alleviating morbidity and mortality. Our primary development focus for 1923 is to address the substantial unmet medical need in ILD, a group of rare immune-mediated lung disorders that can result in progressive fibrosis in the lung. All of these diseases have limited standard of care, with substantial morbidity and mortality. Our initial indication for 1923 is pulmonary sarcoidosis, a major form of ILD which is characterized by the formation of granulomas or clumps of immune cells in the lungs. The formation of these granulomas is driven by persistent aberrant inflammation. If left untreated, it can lead to irreversible scarring or fibrosis and diminished lung function, which may lead to respiratory failure or the need for a lung transplant. We estimate the patient population for pulmonary sarcoidosis to be approximately 200,000 patients in the U.S. About half of all patients will require some form of systemic therapy, and 30% of all patients will have chronic progressive disease despite available treatments. The current standard of care for pulmonary sarcoidosis typically includes treating the inflammation with corticosteroids and other immunosuppressive therapies. While these treatments can help manage inflammation and alleviate symptoms such as cough and shortness of breath, they have no demonstrated efficacy on disease progression; and can result in serious, long-term toxicity. Additionally, many patients do not respond to currently available treatments, so there is a substantial need for safer, more effective treatments that could reduce or replace the requirement for chronic corticosteroid or other immune-suppressive therapy and prevent disease progression. Now let's get back to discussing the results which we have presented to you today. As a reminder: The Phase Ib/IIa study was a randomized, double-blind, placebo-controlled, multiple-ascending-dose clinical trial consisting of 3 cohorts testing doses of 1, 3 and 5 milligrams per kilogram of 1923; or placebo, dosed intravenously every month for 6 months. The study enrolled 37 histologically confirmed pulmonary sarcoidosis patients. These patients had active and symptomatic disease at baseline and were stably managed on a dose of corticosteroid treatment of greater than or equal to 10 milligrams per day. The primary objective of the study was to evaluate the safety, tolerability, immunogenicity and pharmacokinetic profile of multiple doses of 1923 compared to placebo. Secondary objectives included assessment of the potential steroid-sparing effects of 1923; in addition to other exploratory assessments of efficacy, including lung function by evaluating FVC, serum biomarkers, health-related quality-of-life scales and PET imaging. Because it was a small study, the study was not powered for statistical significance. Regarding PET scans, due to operational issues related to the COVID-19 pandemic, this data was not evaluable, as we had approximately 50% of the patients missing post-baseline scan, rendering this analysis too limited to have any meaningful interpretation. In terms of the trial design, the study included a 4-steroid taper, where patients [ with daily ] steroid dose was tapered to 5 milligrams by week 8. 5 milligrams is considered a subtherapeutic dose, meaning that it's likely that patient symptoms such as cough and shortness of breath will not be able to be managed at this dose and may even get worse. The study aimed to keep patients at a dose of 5 milligrams of steroid per day until study completion at week 24. If a patient's symptoms got worse, the principal investigator or PI had the ability to increase the patient's dose of steroids to help improve their symptoms. If the patient's symptoms remained stable at week 16, the PI had the option to attempt to titrate the patient off steroids completely. At the end of the study, we anticipated being able to assess a potential reduction of steroid burden in the 1923 versus placebo arms. As we start to present the results, it's important to note that analysis from the study shows that key demographic and baseline characteristics were generally well balanced across all treatment groups, for demographics, including age, sex and race; and for disease characteristics, including FVC, duration of disease and baseline dyspnea index score. As you can see on the table with regard to baseline steroid use, treatment groups were mostly balanced. The mean daily steroid dose ranged from around 11 milligrams per day to just about 4.5 milligrams per day across the 1923 treatment groups and placebo. Some patients were also taking background immunomodulators which were permitted in the study. Okay, now let's get a bit more into the results. Let's start with the safety findings, which are a key part of 1923's value proposition considering current treatment options for pulmonary sarcoidosis typically have serious side effects and long-term toxicity. Overall, monthly dosing of 1923 was determined to be safe and well tolerated. There were no new or unexpected findings with repeat dosing and there was no dose-response relationship observed. Adverse events or AEs were mostly mild or moderate in severity and generally assessed as unrelated to the study drug. The most frequent AEs were consistent with the underlying disease; and include cough, fatigue and wheezing, all symptoms you might expect to see in pulmonary sarcoidosis patients. There were no drug-related serious adverse events, and there were no deaths observed in the study. Importantly, there was no signal immunogenicity observed, which is a common and sometimes expected adverse event with a biologic therapy. This study showed no significant induction of anti-drug antibodies with repeat dosing. These findings are consistent with previous studies of 1923 in patients and healthy volunteers and reinforce 1923's favorable safety profile. Now let's talk in some more detail about some of the trends we saw on key efficacy end points. We'll start with steroid reduction, one of the key assessments of efficacy included in the study. As a reminder: All patients enrolled in the study were stably managed on steroids. In terms of the forced steroid taper, most patients were able to taper to 5 milligrams. Nearly all of the patients in the 1923 treatment groups were able to taper to 5 milligrams, while only 75% of the patients on placebo were able to do so. Considering baseline steroid use, let's look at the dose-dependent reduction in steroid utilization that we saw with 1923 during the post-taper period. When you look at the average daily dose, overall we saw a relative reduction and overall reduction of steroid use in all of the 1923 treatment groups compared to placebo, and we saw a dose response. When you look at the 5 milligram per kilogram treatment group, we see some impressive activity. Here, throughout the duration of the trial, there was a 22% relative reduction of steroid dose compared to placebo, demonstrating these patients were able to tolerate lower steroid doses by receiving 1923. And when you look at the change of baseline overall, patients were able to reduce their steroid dose by 58% from what they were on when they entered the study on average. To put that in context. Many PIs have stated that, for patients receiving baseline steroid of around 15 milligrams per day, an overall reduction of 50% or greater will be highly impactful to patients and especially if you also see concomitant symptom improvement. Finally, if you recall, the trial design did present the opportunity for PIs to try to taper patients completely off steroids; and we had 3 patients who indeed were able to do so, all in the 5 milligram treatment group. This is an outstanding finding and indicates 1923 could be a transformative steroid-replacement therapy. We're very pleased with the findings of steroid reduction, which on their own show a dose response and a measure of drug activity, but it is the steroid reduction combined with improvements in other efficacy end points like lung function and clinical symptoms which can have the potential to be particularly impactful to patients. Let's move to FVC, a well-validated measure of lung function and an important measure of disease in this patient population. Because of the granulomas that form in the lungs of patients with pulmonary sarcoidosis, patients' lungs can become fibrotic, meaning they become stiff and make it more difficult to breathe. Forced vital capacity or FVC is a measure of volume of air exhaled over one full breath and is a standard measure of lung function in patients with ILD. The greater the FVC, the better the lung function. There were some differences in baseline FVC across treatment groups, but we're looking to assess the mean percent change from baseline to week 24 for each treatment group. For context, remember steroids are used in these patients to help control inflammation, so by decreasing steroid use, it is expected that FVC may decline without this control. So let's take a look at what we saw. In the placebo and 1 milligram treatment groups, FVC remained stable or slightly declined, but in the higher-dose treatment groups, we saw FVC improve. Patients had an improvement of 2.8% in the 3 milligram treatment group and a 3.3% improvement in the 5 milligram treatment group compared to placebo at week 24. The data we see here was an outstanding result and would not typically be expected in such a short study. So simply stated, in the 3 milligram and 5 milligram treatment groups, patients have their steroid dose substantially decreased, received 1923. And their lung function greatly improved compared to placebo. What is also notable is the percentage improvement seen which are consistent with previous studies in ILD, including studies in sarcoidosis and idiopathic pulmonary fibrosis or IPF. FVC improvement of greater than 2.5% is often considered clinically meaningful, so the percent increases of 2.8% and 3.3% that we demonstrated with 1923 are remarkable findings to our experts. We've discussed steroid reduction and improvements in FVC, 2 objective measures of efficacy. Now let's take a look at the quality-of-life assessments used for sarcoidosis symptoms and see how patients felt at baseline compared to the end of study. There were validated symptom measures that we used in the study for dyspnea or shortness of breath, cough, fatigue, the King's sarcoidosis scores for lung and for general health. Patients completed these symptom assessments at baseline monthly during the study and at completion of the study at week 24. What we are looking to assess is compared to a defined minimal clinically important difference or MCID for each of these indices. So did the patients improve, remain stable or worsen? When we look at the results, again here we see dose-dependent clinically meaningful symptom improvements in the 1923 treatment groups compared to placebo. While patient symptom scores remain mostly stable in the 1 milligram treatment group compared to placebo, we see clinically meaningful improvements in 4 out of 5 symptom scores in the 3 milligram treatment group and, in the eyes of our experts, dramatic symptom improvement in all of the scores in the 5 milligram treatment group. These are very important findings, as steroids are known to have side effects that can greatly impact patients' quality of life as much or even more so than the disease itself. And quality of life compared to current treatment options is of high priority to patients with pulmonary sarcoidosis. What these findings demonstrate is that not only in the context of a steroid taper that 1923 reduced the amount of steroid burden and improved lung function, but this also translated to dramatic improvements in how patients felt. Finally, we'll quickly describe some of the biomarker findings. Again consistent with the other measures of efficacy we discussed, we saw dose-dependent trends of improvement, compared to placebo, on key inflammatory biomarkers, including IL-6, MCP-1, interferon gamma, IP-10 and TNF alpha; as well as key markers of sarcoidosis, including ACE, IL-2 receptor and SAA, with best anti-inflammatory control in the 5 milligram per kilogram treatment group. These are the important biomarkers related to inflammation and sarcoidosis. And many are the very same biomarkers that we saw positively impacted by 1923 in both our preclinical studies and our clinical study in COVID-19 pneumonia patients. A few key findings for the biomarker data include: Inflammatory biomarkers increased overall in the placebo population and largely remained elevated, which is what we expected to see. Inflammatory biomarkers were more tightly controlled in all 1923 treatment groups, demonstrating that 1923 had an anti-inflammatory effect with a dose response. As an example, over the course of the study, we see IL-6 and TNF levels improve in our 3 and 5 milligram populations compared to almost a twofold worsening in the placebo population. The important takeaway here is that, by removing steroids and adding 1923, inflammation was controlled in a dose-dependent manner; and this aligns to what we saw clinically. The results of this study support the advancement of 1923 to its next stage of clinical development for pulmonary sarcoidosis and expanding 1923's potential benefit to other forms of ILD. In terms of the next steps for pulmonary sarcoidosis, we plan to request an end-of-Phase II meeting with the U.S. Food and Drug Administration to discuss these data and subsequent clinical development and path to registration for 1923 for pulmonary sarcoidosis. We expect to initiate a registrational trial in pulmonary sarcoidosis next year. Also we are planning with our investigators to submit these data for publication in a major journal and for presentation at a medical conference in the future. While our initial focus for 1923 is on pulmonary sarcoidosis, the mechanisms of action, translational work and clinical data combined with the overlapping disease pathology across ILD strongly suggest that 1923 could have the potential in other ILD indications as well. ILD is an umbrella term that refers to a group of over 200 individual diseases that can cause fibrosis or scarring of the lung tissue primarily driven by aberrant immune response to an inhaled exposure or other insult. The main 4 types make up roughly 80% of the population. These include connected tissue disease-related ILD where lung manifestations are secondary to autoimmune diseases such as systemic sclerosis; chronic hypersensitivity pneumonitis, which results from an exaggerated immune response to environmental antigens; and idiopathic pulmonary fibrosis, the prototypical fibrotic lung disease. We estimate that there are over 500,000 patients currently living with ILD in the United States. And outcomes for these patients remain poor despite current treatment options, with median survival as low as 3 years in certain cases. Scarring of the lung can occur in upwards of 20% of patients across all forms of ILD. Standard of care outside of IPF is treatment with immunomodulatory therapies similar to those used in sarcoidosis. These treatments have limited clinical evidence of efficacy and are associated with significant long-term toxicity. We think we have a potentially transformative therapy in 1923 with an addressable multibillion-dollar market. Overall, we're very pleased with the results of this study and we're excited at the potential that 1923 holds. Based on the study results, we believe we have a safe drug that we expect to advance to a larger registrational trial in pulmonary sarcoidosis next year, which is one step closer to a new treatment with improved outcomes for this debilitating disease. We're very fortunate to be -- to have been guided and encouraged by some of the leading PIs in the field and can sense the excitement at their reaction to these findings. You can read some of their supportive statements in today's press release. It's a great day for aTyr. This is a culmination of many years of hard work. It has taken a lot for us to get to this readout. And I'd like to acknowledge, in fact, the entire aTyr team, along with our partner, the Foundation for Sarcoidosis Research; our clinical sites; our investigators; and most importantly, the patients for their hard work, dedication and support in helping us achieve this milestone for the company. We appreciate your interest and continued support and look forward to providing updates in the future. At this time, along with Jill Broadfoot, our CFO, we'll be happy to take your questions.

[Operator Instructions] Your first question will come from the line of Zegbeh Jallah with ROTH Capital Partners.

Congratulations on the data, very exciting. I think, like you said, Sanjay, it validates 1923 but also [ the journeys in today's ] platforms. So excited. I just have a couple of quick questions. I think the first is can you just provide a bit of details on the patient population, how representative this patient population is?

Sure. So when we think about this population: They're all stably managed on steroids. They've advanced in their disease, so they do have advanced form of sarcoidosis. Our view is we're looking at moderate to severe disease, has not progressed to where these patients are highly fibrotic, but this is where we think there is the largest addressable population within sarcoidosis. It's also the area where I think patients feel perhaps the most pain from the current therapy of steroids, so I'd consider this an advanced cohort of sarcoidosis patients. And I do think it represents a larger swath of the, say, 200,000 patients or so that have been estimated here in the U.S. living with this disease.

And then it's really nice to see the efficacy across the different end points that you kind of looked at, particularly FVC where we didn't really expect to see much because there's been other 24-week studies that haven't been able to show improvements in FVC. And so I was just curious. For a pivotal study, what do you think the end point might be? Is it FVC? Or do you think it will be steroid sparing?

So traditionally with this division and really with all the approved therapies in IPF, because there are none for these inflammatory forms of ILD, for example, FVC has been used as a gold standard. Well, however, there is quite a bit of momentum to also look at steroid reduction and look for steroid-sparing agents. The fact that we have outstanding results on both of these end points really bodes well for us, so I think it's something that we're going to discuss with the regulators on how to order these 2 end points. There have also been discussions around composite end points with many of the lead PIs in this area worldwide, but I think we have now strong options, whether you call one a primary and the other a secondary, that we feel like we can move forward. Drugs in this area have been approved, just hitting their secondary end point, so I think we've -- we're in a real enviable position right now because we see dramatic findings with both of these 2 important efficacy end points.

And the safety was also really good, especially like you said, when you're comparing to steroids. I was just kind of curious about [ the -- and for the ] safety analysis. I think you enrolled 12 patients per cohort and I think you had safety and efficacy [ in that view ], so I was just wondering if there were some discontinuations and what may have happened there. And then the last bit, a follow-up to this one, it's just about the doses tiers. So we saw the best response at 5 milligrams, and so I was just wondering what you thought about the COVID data that you generated a while back. Perhaps you could have used a higher dose or something like that. I was just curious [ to know if it somewhat ] moved past that COVID data.

Sure, sure. So with regards to -- the safety data set looked at really anyone that got any kind of exposure to our therapy, so that's the 37 patients in our trial. So with regards to discontinuations, we did have a few discontinuations primarily due to the impact of COVID. And this impacted mostly our 3 milligram cohort, where we had -- some patients had to discontinue due to really COVID. Not much we can do there with sites were starting to shut down, but I think the evaluable population here. We just chose to look at all patients that have received therapy, and I think the findings are really consistent. Whether we do a sensitivity analysis on those dropouts or not, it's pretty much in-line and the same. So I think the impacts were -- we're fortunate they were minimal. We had to battle through COVID. There are PIs who are really proud of the fact that we were able to do that, but I think more importantly the patients, many of them, really were quite dedicated to getting their therapy and battling through COVID themselves; many of them traveling across the country sometimes to fly in to the clinical sites, those sites that permitted these patients to come in. So discontinuations were minimal and largely due to COVID. When you mentioned COVID. I think that was a study where we looked at 1 and 3 milligrams. At that time, we really couldn't go to 5 because we were actively in this trial, but we had conducted several drug safety monitoring boards looking at 1 and 3. So the FDA at that time thought we've got good safety with 1 and 3 tracking in this trial, so those doses were viewed as something we could look at. If you remember, in that trial we saw really, really nice activity at 3 milligrams. That gave us our first clue that 3 milligrams was going to be our first real active dose. This trial backs that up. However, we are able to test 5 milligrams in this trial, which is even better. And one could potentially think that 5 milligrams in that population -- I wouldn't have been surprised if we would have even seen better impacts. It's just that at that time we didn't have a safety data set that we could really test that. I wish we could have because I think those COVID patients could have even benefited more so with even one dose of 5 milligrams.

Yes, same, yes. Those are my thoughts exactly but really excited about the data. Congrats again to you and the team.

Thank you.

Your next question will come from one of Joe Pantginis with H.C. Wainwright.

Congratulations on the data. First, let me just go back to the Phase III. I was just curious if you have a idea yet about potential size.

Yes, great question, Joe. So if you look at the kind of deltas we produce, if you then want to power for that kind of difference, which as we've highlighted in nearly -- in really all of the end points we've looked at, they're all clinically meaningful. We passed the threshold here. That impresses the experts. If we want to keep to that delta and model and power a new trial, we're right in the ballpark of about 200 patients, which is what we thought going into these readouts. So we thought a trial looking at about 200 patients, yes. A global registrational trial is what we're targeting. So this is holding up really, really well with regards to these reductions. And going back to the previous question: Whether you look at steroid reduction or FVC improvement, if we power a study, we'll be able to [ maybe ] enroll about 200 patients who will hit a significant p-value. And I think it's fairly straightforward approval then.

That's great. And then 2 questions on the data, but I guess it's from a follow-up standpoint. So I think obviously the steroid tapering and getting some patients off as well, those 3 patients, I think, is pretty exciting and compelling, so I was just curious. First, with regard to follow-up of those patients past week 24, do you have any or do you plan to give any duration data? And then secondly, with regard to obviously COVID impacted the scans, the FDG-PET scans, curious if there were, maybe at minimum, any anecdotes you can provide from the patients that did have follow-up scans.

Sure. So with regard to following up or any extension, there wasn't an extension component to this study. I think some of those patients, especially those that tapered 0, which there was -- unfortunately, we had to obviously make a cut here and really analyze this data. I think anecdotally we've seen that, patients who were able to receive 1923 and taper off, they don't want to go back on steroids. They were pretty clear, in the patient notes, that they're just feeling great. We've seen that. We now understand that it matches what we saw with our data, but there wasn't an extension component to this trial, so unfortunately they had to likely go back to standard of care, which is mostly steroids or sometimes bringing a mix of steroids and immunomodulators. So that's really the first piece there. With regards to PET, Joe, it's really tough for me. I tend to be pretty conservative looking at trends and things of that nature. There's just not enough data in the 3 and 5 milligram. I can tell you that there's no overt signals of worsening when we compare, say, placebo to 1. Anecdotally I can always find 1 or 2 patients where I look at the PET scan in 3 and 5 and say, "Wow, that's really trending the right way," but this is -- to me doesn't make it a trend, if you will. I -- just I wish I had more data there, but I do think that, if we had a full data set there, we might have seen some real interesting findings in the 3 and 5 milligram population. But too small of a data set for me to make a call there. I'd like to be a little bit more sure. We will be looking at some of that data, of course, with the PET experts, but unfortunately we -- I just can't make a call there.

Of course, completely fair. And I've always appreciated the conservatism on your part. And then my last, I guess, relatively quick question is just logistical. Do you have any visibility as to when we might get an update out of Kyorin?

Well, yes. I think -- Kyorin is our partner in Japan. And they have moved through their Phase I trial, I think, a Phase I trial in healthy volunteers. Typically no news is good news because that's a safety trial. What I can tell you is they are on track to join us in a worldwide registrational trial. So they're very much on track. They've demonstrated, in my mind, in my opinion, a good safety check there with healthy volunteers. And certainly this is a kind of data that's going to really electrify worldwide PIs in the ILD space, including those in Japan. So I can just tell you early sort of touch points with Kyorin is -- our partner, is extremely happy. And I expect that the experts in Japan will also start to want to really tap into 1923 quickly and get a trial started there.

Your next question will come from the line of Hartaj Singh with Oppenheimer.

Great. Sanjay and team, really, really nice data. Sanjay, a couple of questions from my end. So your side effect profile stayed pretty consistent. There's really no increase in side effects as you increase the dose, but you're seeing increases in efficacy through the 3 doses. My first question is -- and also it looks like your confidence intervals are getting better and better through the 3 doses, which is I think an even better sign. My question to you is, if you have followed this study and just looking at the data that you have -- and I know you had to stop the study at 24 weeks. If you were to extend the study, if it kept on going further, would you have seen sort of a -- that temporal effect continue, meaning the separation of those curves get greater and greater? So that's my first question. I've got a couple of follow-ups.

I think that's the way -- I mean you asked a question that already some of our experts are thinking about. They see the trend moving upwards, and it's steady. So I do think that it's a combination of 2 things there. What you start to see is hints of durability with our therapy, that it gets higher and higher. And as you can imagine, as you start to see more exacerbation or relapse in placebo, those rates will start to taper down. So I think we are really in an advantageous situation here that we could see the trajectory of our therapy, whether it speaks around steroid reduction or FVC improvement or clinical improvement, continue to get better. And I think the other component of that is you could see placebo in a longer study start to worsen even more, and that's going to create even more -- larger deltas for us. So I think your question is fair. And I think, when we look at the trends, yes, we'd love to do a longer study because we think we can actually really show even larger deltas compared to placebo, especially when you think about what we're seeing in the 5 milligram population which is really outstanding and dramatic.

Yes. That's great, Sanjay. And then my other question is you had 3 patients in the 5 milligram per kg on -- go to 0 on steroids. And as -- you had done your [ KOL calls and a lot of our checks ]. I mean that's pretty amazing for patients. Were there other patients that were close to getting to 0 other than those 3? I mean, did you see any other patients where, if they had a few more weeks, they could have gone to 0 milligrams on steroids at the 5 mg per kg arm?

Well, what I'll tell you is, when you look at the steroid reduction number, 58% in that treatment arm and more or less the average dose of just over 5 -- which is essentially 5, right, if you think about it, because you're also taking into account post-steroid taper there. What you really -- what I'm seeing is a number of patients that I might have maybe pushed the gas a little bit more, but I'm not a pulmonologist. We've seen, we see patients basically knocked down from, say, 20 or 25 milligrams; and then they get to 5 quickly and then they just held that. So I think what's going to happen here is, after this data, more of those PIs might look back and say, "Well, I maybe could have pumped the gas a little bit more," but to get 33% of these patients in the 5 milligram group completely off -- you also have to consider in that treatment arm one of these patients, because of COVID impact, only had one dose. So there's 8 completers there, so it's 3 out of 8 who completed -- completely weaned off steroids. So I think certainly there was a few patients where I might have said, "Okay, patients doing really, really well with their symptoms. It looks like their FVC is getting better." Maybe in the next study that PI has a little bit more confidence because now they've seen these effects. That's typically sometimes what happens here, but we're going to have an investigator meeting. And that very question, we're going to go through every patient profile of all 37. And I do not doubt that there is going to be a few of the PIs who might say, "Wow, I could have maybe tried on this one to go to 0," but nonetheless, I think to see 30%, 40% of the patients here completely get off steroids in such a short time really demonstrates the profound attributable benefit of 1923 for these patients.

Yes. No, it's -- I mean, the directionality and the consistency of the directionality in the dose responses and the temporal consistencies, I mean, it's really nice. Last question is just you've got data on key inflammatory biomarkers compared to placebo. We look forward to whenever you publish this at a medical conference, but do they give you any line of sight into which other indication that you mentioned in your slides aside from pulmonary sarcoidosis you could sort of -- you'd want to test this drug in? And when could we see some movement there from a clinical perspective?

Sure. So when you think about these sort of top line inflammatory markers that I mentioned of the interferons, TNF, IL-6, these are -- there's commonality in other inflammatory ILDs where those are also inflamed. These are involved in the same kind of aberrant immune response you see in scleroderma patients who have ILD or hypersensitivity pneumonitis patients. In fact, when you look at some of the animal work we've done, we've put out translational efficacy in models of scleroderma -- for example, we presented a poster at the Scleroderma Foundation several years ago that showed 1923, frankly, outperform the existing therapy, nintedanib, in a very aggressive mouse model of scleroderma ILD. In that same model, we see declines in these same inflammatory biomarkers, so there's consistency in the fact that we've always -- and you can even go back to all of our presentations. These are the same biomarkers that we saw impacted in animals. These are the same biomarkers that we saw impacted in COVID pneumonia patients. And these are the important biomarkers involved in really all of the spectrum of ILD, so we believe that we've got a therapy well positioned to expand into, for example, scleroderma ILD or chronic hypersensitivity pneumonitis. This data is going to lead progress and really get those indications and experts who treat those patients interested in pursuing a Phase II trial, so this is something that we will be looking into rather rapidly. I think we now become the leader in the ILD space. We're the furthest along. And we've demonstrated, I think, some outstanding signals here, so undoubtedly I think there is going to be significant interest to expand and jump into Phase II trials looking at some of those other indications to expand the sort of breadth of 1923's applicability.

Yes, great, Sanjay. And really kudos to your team for such a consistent database in a very tough pandemic-driven time.

Your next question comes from the line of Prakhar Agrawal with JonesTrading.

Congratulations on the data. My first question is on the patients who saw the best steroid reduction. Anything that stood out on their baseline characteristics, whether it's disease duration, background therapy, baseline efficacy, et cetera? Just wondering if there are ways to enrich your Phase III trials based on these data. And I have a few follow-ups.

Prakhar, that's a great point because I think looking at -- are there any sort of baseline signatures that may indicate that 1923 is even more sensitive? As you can understand, it's a small data set, so being able to pull out those sort of sub cohorts and looking at sensitivity that way -- we're going to be looking at that. Because the baseline characteristics were more or less balanced, I would say right now there isn't an overt signature, but as I said, we have 37 patient profiles to now go through with all of the experts and PIs in our trial. I think one covariant we're really looking at is race. African-Americans typically have more severe disease; and we did enroll, frankly, a higher proportion of those patients in the 5 milligram group. Conversely that's also the treatment arm that we performed the best, so that's a good sign. I think the fact that maybe the performance in a tougher cohort of patients that shows some of the durability that I'm talking about, but when we look at any of the other baseline characteristics right now, I don't see a covariant per se. But we are going to be looking at this post talk because this can best inform us on how to basically create a really tight, potentially efficacious population in our next trial through inclusion/exclusion criteria, but I'll be looking closely at that.

Got it. And secondly, on FVC improvement, could you frame the improvement of 3.3% in the context of any data that has been shown by some of the other drugs in the setting, even off-label anti-TNFs that are often used? And for patients who saw the -- who saw FVC improvement, did they also have the most steroid reduction? Just wondering if there is any correlation there.

Yes. So the first question is this is in-line and, honestly, better than what you were seeing with some of those approved or, I would say, efficacious therapies that might have toxicity. When you talk about infliximab, that was about a 2.5% improvement in a study that [ Dr. Baughman ] led about 10, 12 years ago, but infliximab comes with its own toxicity. And I think that's why ultimately it's a difficult therapy for some of these patients to take, but we're well over that threshold. And when you look at some of the other approved therapies in the space: Tocilizumab recently, for example, for scleroderma ILD, we are showing FVC changes that are above and beyond those changes. And let's remember another thing too. Many of those therapies for IPF, they're looking at a delta of less worsening. We're showing something, frankly, a lot better. We're showing improvement, actual improvement, here. So a regeneration, if you will, of lung function. So I think that should really be heightened here, that these are typically diseases where therapies are approved because they show 2.5% less worsening, if you will. So I think we feel really great about that data. And that is, I think, the data that has the experts really, really pointing out how remarkable it is to see something like that. So that's kind of my FVC point. Your other question -- what was the second half of the question? Sorry.

Just curious if, patients who the -- saw the most FVC improvement, they have the most steroid reduction, if there was -- yes.

Yes. I mean I think what I would basically say is generally it's lining up. I wouldn't say that there's maybe a pure correlation, but we are looking post talk here an exposure-response relationship. And it appears at least preliminarily that, the more drug you receive, the better response regardless of whether you're looking for a response in symptoms or response in FVC improvement. There seems to be a correlation occurring also with the biomarkers as well. The biomarkers are also correlating. So specifically looking at those that might taper off or get to really low doses, yes, there is a strong suspicion already that we've got a tight correlation there.

Your next question will come from the line of Yale Jen with Laidlaw & Company.

Also accept my congratulations to the outstanding outcomes. My first question is actually related more to the placebo. Just curious, any patients in 5 milligrams that eventually need some rescue? If so, how fast those rescues have happened.

Sure, sure. So when you talk about rescue, and that data has been presented, I obviously have to hold something back because this is going to go into a major publication, but I -- what we're basically seeing is the placebo rate of relapse, if you will, is somewhere around, say, 45% to 50% in the Kaplan-Meier curve, which is expected in a trial of this duration, according to the experts. If you take it a little bit longer, we expect those to trend up even upwards 60%, 70%, maybe 80% if you get to a year trial. What we're seeing in 5 milligrams and 3 milligrams is we're staying, frankly, below -- close to about 10%, very, very low numbers. I think, if you look at 3 and 5, that might be what I would consider maybe just one patient who might have really needed to be what I would classically call a rescue. And then there are some questions of whether or not that rescue has to do with sarcoidosis or not because some of these patients are coming in with other issues, other autoimmune disease or autoimmune-like symptoms. So we're teasing out some of that, but I will tell you that, what I can say overall, relapse rates appear to be, say, fourfold higher right now in this trial, in placebo, but that's data that will be published and spoken about at a major conference: very, very low relapse rates right now being observed in our [ high ] 3 and 5 milligram 23 arms.

Okay, great. That's very helpful and very supportive as well. Maybe just a question and follow-up on what Joe has asked: You mentioned that the next trial could potentially enroll about 200 patients and based on the statistical powering. Is that based on the FVC? Or that based on the steroid sparing. And what might be the -- sorry. What might be duration for the study if you think about for the next one?

We're looking at -- yes. [ Go ahead. That's ] -- yes. So we're modeling it to be a 9- to 12-month study and we've looked at it both ways. We've looked at, okay, if we held this kind of delta with FVC improvement versus also steroid reduction -- and it's coming out to be about the same ballpark. So this is why I think we have real options here when we discuss this with the FDA in how we want to order this. We have the potential to really preserve alpha for either of these 2 end points. And as I've mentioned, drugs have been approved here and whether or not you hit the primary or the secondary end point, so we've got 2 real outstanding shots here. And when we sort of extrapolate into the power calculations, we're coming out to about a 200-patient trial. Whether you want to actually put steroid reduction first or FVC first, it's a similar kind of plan.

And maybe the last question here is that I think, earlier discussions, you guys may -- talked about maybe [ at least modeling ] as part of the infliximab study. At this point, is there any changes or different thoughts? Or there is total new different thinkings maybe after you spoke with the FDA.

No. I mean I think we had modeled some things based on that trial. That was an FVC improvement, but as I said, they have a toxicity element. We don't have that. We also see some florid symptom improvement. And remember that trial didn't have steroid reduction, so we've got a lot more in our arsenal to really, I think, have an even more informed discussion with the FDA. So I would say we would have an enhanced ability to have more options and more flexibility to win here for patients even compared to that infliximab trial.

Okay, great. And again, congrats on a very outstanding outcome.

Thank you, Yale.

Your last question will come from the line of Kennen MacKay with RBZ (sic) [ RBC ] Capital Markets.

Let me offer my big congrats on this readout as well but -- on the data and proof of concept but also just the operational execution here. It's no small feat to run a trial like this in fragile patients with compromised lung function during a global pandemic, so again big, big congrats there. Sanjay, I'd like to ask on the mechanism here. And for context, I'm asking because I'm just thinking about how this proof-of-concept data in pulmonary sarcoidosis can inform the potential in other indications, especially some of the ILDs, out there. And as I'm looking at it, it seems like maybe there are 2 major variables. The first is maybe the involvement of neuropilin-2, NRP2, in the etiology of the disease versus the involvement in pulmonary sarcoidosis and especially the inflammatory component; and then second, maybe the involvement of inflammation versus fibrosis in the etiology of the disease as it relates to pulmonary sarcoidosis. So again I'd just be very curious on mechanism and how that can be extrapolated to other diseases.

Yes. That's a great question because we have also validated the target here as a potential target in ILD sarcoidosis. We'll have to importantly look at that target. Neuropilin is a type of target that we see expressed in a number of inflammatory types of responses. In the literature you see it, for example, in the joints of RA patients. So we would expect that we will find this target also in other ILDs once we start to assay that tissue, but that's going to be an important component because, last year, we showed that neuropilin was enriched in granulomas with sarcoidosis patients. So we knew the target was there, waiting for a drug like 1923 to sort of shut things off. So this will be important, as we start to think about those other conditions, to really consider neuropilin biology as a target in those ILDs that -- these other ILDs. When you talk about that balance between inflammation and fibrosis, this goes hand-in-hand with neuropilin, which is expressed primarily during the inflammatory response. So what -- [ we want to partition ], and we've always thought this, that our drug is better when there is more of the inflammatory component of the disease. So maybe avoiding those patients that might be more end stage and fibrotic because that's typically where you see the inflammation, frankly, die down because now the lungs of these patients have become, unfortunately, fibrotic. So we think there's an opportunity here to get involved early when there is an inflammatory response. If we can actually tap into those pockets when neuropilin is driving this response, we think we can actually turn things off and really prevent the progression of fibrosis. So I think that's really the outcome that [ now one would really look at ] when you think of the connected-tissue disease ILDs and also some of the pneumonitis conditions out there. So you're thinking about it exactly the same way we are. And thanks for the shout-out about getting this done in a pandemic. I think that was no small feat, according to some of our PIs. And I think that speaks to the dedication of the patients but also the fact that now we can see that the majority of them on our drug were feeling really good. So that might have been a real motivation for them to power through a real difficult time here.

And maybe just a quick follow-up, if I may. I know you have been considering other ILDs for a trial here, CTD-ILD or CHP. I'm just wondering sort of where we are in terms of time lines there. And congrats again.

Yes. I mean I think, where we are, we're just digesting this data. I think there is going to be significant interest, probably starting today, to start to move into those conditions because patients need better options than steroids in those conditions as well. Demonstrating this dose response and also the safety is also going to really accelerate our thinking in using 5 milligrams in those populations. So while I don't have a time line right now, I think first things first. We really are focused on moving sarcoidosis forward, but I do think that there is going to be options here to potentially quickly move into those other conditions because of the type of data that we produced.

At this time, there are no further questions. Mr. Shukla, do you have any closing remarks?

Sure. So I'd just like to thank everyone's interest. I'd like to thank everyone who's supported us as we've obviously worked through what we think is a real transformative trial for us with some results that we hope to, as I said, share in a major journal and a larger medical conference in the short future here. So thank you, everyone, for your support, your interest. And we will be in touch in the future. Thanks so much.

Ladies and gentlemen, thank you for participating in today's aTyr Pharma conference call to review results for the Phase Ib/IIa clinical trial of ATYR1923 in pulmonary sarcoidosis patients. You may disconnect at this time.