Theravance Biopharma, Inc. (TBPH) Earnings Call Transcript & Summary

Ladies and gentlemen, good afternoon. I'd like to welcome everyone to the Theravance Biopharma Conference Call. [Operator Instructions] Also, today's conference call is being recorded. And now I would like to turn the call over to Gail Cohen, Vice President, Corporate Communications. Please go ahead.

Good morning, and thank you for joining. As always, I remind you that this call will contain forward-looking statements that involve risks and uncertainties, including statements about our development pipeline, expected benefits of our products, anticipated timing of clinical trials, regulatory filings and expected financial results. Information concerning factors that could cause results to differ materially from our forward-looking statements is described further in our filings with the SEC. Joining us are Rick Winningham, Chief Executive Officer; and Rick Graham, Senior Vice President, Development. Following our prepared remarks, we will open the call for questions, and Andrew Hindman, Chief Financial Officer, will join. Now I will hand the call to Rick Winningham.

Thanks, Gail. Today, we announced top line results from our Phase II study of nezulcitinib in hospitalized patients requiring oxygen support with acute lung injury to severe COVID-19. Given our experience in developing respiratory therapies, with the onset of the pandemic, we chose to accelerate nezulcitinib, also known as TD-0903, our investigational inhaled lung-selective pan-JAK inhibitor into clinical development. We committed to studying nezulcitinib in a seriously ill population with respiratory distress driven by viral-induced pulmonary inflammation. The data announcement today represents almost 15 months of expedited drug development. In that time, our teams completed Phase I trial in healthy volunteers, an initial dose selection portion of the Phase II, culminating with this randomized, double-blind, placebo-controlled, multicenter trial in over 200 patients with severe COVID. While this Phase II study did not meet its primary endpoint, we're encouraged by the trend in the prespecified endpoint of 28-day all-cause mortality in the intent to treat population and additional key analysis that we'll be sharing with you today. Rick Graham, our Head of Development, will walk you through the specific details for this study. Rick?

Thanks, Rick. Beginning on Slide 3. Nezulcitinib is an inhaled, lung-selective pan-JAK inhibitor that has the potential to broadly inhibit the pulmonary inflammatory cascade caused by viral infection. We highlight several important observations from our preclinical work with nezulcitinib that suggests it may provide a novel therapeutic benefit through 3 distinct activities. Starting on the left panel. Nezulcitinib is a potent pan-JAK inhibitor. In COVID-19, we recognize the importance of targeting key inflammatory cytokine-signaling pathways that arise from viral infections in the airway, particularly interferon as well as others such as IL-6 and IP-10. The middle panel shows that nezulcitinib may protect against virus-induced cell death, augmenting or supplanting a partial response to dexamethasone. And the far right panel demonstrates the potential to limit mRNA expression of cell entry components harnessed by SARS-CoV-2 in the airways. Moving to Slide 4. This was a randomized, double-blind, placebo-controlled, multicenter study in hospitalized patients with COVID-19 requiring oxygen support with the majority of patients enrolled in February and March of this year. Key inclusion criteria include hospitalized patients aged 18 to 80 years old requiring supplemental oxygen to maintain greater than 90% saturation with a positive SARS-CoV-2 test less than 72 hours prior to randomization and symptom onset greater than 2 to 14 days prior to hospitalization. For clarity, patients enrolled at baseline were at low-flow oxygen, high-flow nasal cannula or invasive -- or noninvasive positive pressure ventilation. The study was conducted in South America, Europe, the United Kingdom and the United States. Randomization was 1:1 with 106 patients receiving 3-milligram nezulcitinib plus standard of care with a loading dose of 6-milligram on day 1 and 104 patients were randomized to receive placebo plus standard of care. Patients who were administered medication via the Aerogen Solo nebulization system dosed once daily for up to 7 days. End points were assessed through 28 days. Slide 5 describes the top line results. Regarding the primary endpoint, a respiratory failure-free day is defined as the clinical status of 4 or below on the NIAID 8-point ordinal scale severity. There is no difference between nezulcitinib and placebo for respiratory failure-free days from randomization to day 28. There's no difference on the secondary endpoints of S/F ratio, the proportion of patients in each category on the 8-point ordinal scale or the proportion of patients alive and respiratory failure-free at day 28. Notably, there was a trend in favor of nezulcitinib when compared to placebo for 28-day all-cause mortality and time to recovery. We will report a post-hoc analysis using the key inflammatory blood biomarker, C-reactive protein, or CRP, to further characterize these important end points in our trial. Nezulcitinib was well tolerated when administered once daily for up to 7 days. And finally, consistent with expectations for a lung-selective medicine, plasma exposure of nezulcitinib was low. Slide 6 shows a summary of patient disposition for the randomized population. The intent-to-treat population consisted of 210 patients. Of those 210 patients, 205 were randomized and treated with study drug, 5 patients were randomized but not dosed, 88% of randomized patients completed the study and a total of 24 patients discontinued early for an adverse event or loss to follow-up or as a result of patient decision. Slide 7 outlines the baseline demographic and clinical characteristics of the patients. The mean age was 58 years, approximately 60% of patients were male and main body mass index was 30. Approximately 45% of patients had 2 or more comorbidities. The most common comorbidities included hypertension, coronary artery disease and diabetes. Approximately 80% and 20% of patients were of clinical status score 5 and 6, respectively. Overall, these characteristics, including baseline CRP concentrations, were well balanced between treatment groups and consistent with demographics from a hospitalized severe COVID-19 population at large. Moving to Slide 8. The primary endpoint of respiratory failure-free days from randomization through day 28 was not met in the study. There was no statistically significant difference between nezulcitinib and placebo on the [indiscernible]. I'd now like to take you through encouraging results from the prespecified analyses in the ITT population, followed by results in the post-hoc analysis in which there was a meaningful reduction in all-cause 28-day mortality rate and time to recovery. Moving to Slide 9. On this and several subsequent slides, the nezulcitinib arm is represented by the orange line and placebo is represented by the gray line. The X axis on this slide is a cumulative proportion recovered and the X axis is time in days. As a reminder, time to recovery is defined as a clinical status of 3, 2 or 1 on the 8-point ordinal scale. There was a trend of improvement in median time to recovery of 10 days with nezulcitinib compared to 11 days with placebo. This rate of improvement and time to recovery is apparent from day 10 through day 28. Slide 10 illustrates 28-day all-cause mortality and time to mortality for the 2 study arms. Treatment with nezulcitinib resulted in a trend of improvement in the 28-day all-cause mortality rate and time to mortality. Of note, there were 19 total deaths in the study, 6 deaths or 5.7% in the nezulcitinib arm and 13 deaths or 12.5% in the placebo arm. The p-value for time to mortality was 0.08. The hazard ratio of 0.42 suggests that there is a 58% reduction in the likelihood of death in favor of nezulcitinib. Moving to Slide 11. Understanding the treatment of COVID-19 first requires understanding the immune response induced by this virus on the patient's respiratory system. CRP is an acute phase biomarker of inflammation and has been found to broadly correlate with disease severity. Normal CRP is generally less than 10-milligram per liter. However, CRP levels begin to increase in patients with SARS-CoV-2 infection. In those hospitalized with COVID-19, the epidemiologic studies have shown approximately 85% of these severe patients have significantly elevated CRP levels. Published literature shows in those that die or need high levels of respiratory support, the median CRP is approximately 100 milligram per liter and can be elevated to 150 milligram per liter or more in critically ill patients. As we showed in our demographic table, the baseline mean CRP level was between 70 to 75 in our study with individual levels up to 300. On the next 2 slides, we will provide you with the post-hoc analyses using baseline CRP to better understand the signal and trends that were noted in the ITT population for time to recovery and mortality outcomes. On Slide 12, the side-by-side figures show a baseline CRP level of less than 150 on the left including 171 patients and greater than or equal to 150 on the right including 30 patients. These analyses include all patients for which we have baseline CRP values and address 96% of the ITT population. The baseline characteristics are balanced between both groups. The CRP cutoff of 150 was based on seminal work from the NHS trust, which was published in Lancet Rheumatology, stratifying severity in hospitalized patients. In patients with baseline CRP less than 150 in our study, we noted there was an improvement in time to recovery in favor of nezulcitinib with a p-value of 0.02. In contrast, the right-hand figure shows there is no difference in patients with baseline CRP greater than or equal to 150. Following the results of improved time to recovery on Slide 13, we show the mortality analysis by CRP level. It's important to note that all of the deaths that occurred in the study are incorporated in this analysis. The figure on the left shows that in patients with a baseline CRP of less than 150, there was an improvement in the 28-day mortality rates, 1.2% in the nezulcitinib arm versus 10.6% in the placebo arm or 1 death on the nezulcitinib arm versus 9 deaths on the placebo arm. There's a clear separation of the curves, showing an improvement in time to mortality with a p-value of less than 0.01. The hazard ratio of 0.1 also corresponds to a significant reduction in the likelihood of death in favor of nezulcitinib. The figure on the right shows that in patients with a baseline CRP of greater than or equal to 150, the curves overlap, indicating there is no difference. Of note, the mortality rate in this group was approximately 30%. Complementing our efficacy results, Slide 14 summarizes the key findings in the study that further inform the overall clinical safety profile of nezulcitinib to date. First, nezulcitinib is well tolerated. When we look at the overall adverse events, these occurred in 34% of patients treated with nezulcitinib relative to 41.2% of patients who received placebo. And for serious adverse events, these occurred in 9.7% of patients treated with nezulcitinib relative to 15.7% of patients who received placebo. In terms of other safety findings relevant to the use of the JAK inhibitor in patients with COVID-19, adverse events of liver abnormalities or liver disease occurred in 9.7% of patients treated with nezulcitinib and 7.8% of patients on placebo. The frequency of patients experiencing serious infections was limited to 1 patient or 1% of those who received nezulcitinib and 2% of those who received placebo. As for events of venous thromboembolism, or VTEs, which have been associated with the use of systemic JAK inhibitors as well as observed in patients with COVID-19, there were no instances of VTEs in patients treated with nezulcitinib. In contrast, VTEs were reported in 4.9% of patients who received placebo. Based on our Phase II study results in patients with COVID-19, we're reassured by the overall safety profile of nezulcitinib and we're excited about the potential for utility of benefiting patients with acute lung injury. Slide 15 summarizes the key takeaways from our Phase II nezulcitinib trial in hospitalized patients with severe COVID-19. As reported in the literature, the heterogeneity of clinical practice across different geographies contributes to the challenge of meeting progression endpoints based on the ordinal scale in COVID-19. In contrast, mortality is an objective outcome that does not change across geographies. This is consistent with results from our study showing a clinically meaningful impact of nezulcitinib on mortality and time to recovery in patients with severe COVID-19. Although this 200-patient study was not powered to demonstrate a difference in mortality, we saw an encouraging signal. In addition, the utility of an established biomarker, like CRP, may complement the traditional clinical status ordinal scales in identifying a target population with acute lung injury that may best be treated with nezulcitinib. With that, I'll turn the call back to Rick Winningham for closing remarks.

Thanks, Rick. This is the first Phase II study of an inhaled JAK inhibitor in COVID-19 patients, and the standard of care in our study included 99% of patients on systemic corticosteroids. While the study didn't meet its primary endpoint, we're encouraged by the trends that we saw in time to recovery and the numerical reduction in mortality. As Rick discussed, these clinically meaningful endpoints were more pronounced in the post-hoc analysis of patients with CRP less than 150. We believe these results demonstrate the potential for nezulcitinib, the first inhaled pan-JAK inhibitor for the treatment of acute lung injury. The execution of this program brought the full complement of our scientific and technical expertise and applied it to the clinical program and in parallel used our experience in inhaled products to manufacture the product at scale. Before closing, I'd like to acknowledge how grateful we are to the patients and their families, our research partners, the clinical investigators and our team with Theravance Biopharma for their important contributions to the study. I'll now hand the call back to the operator for questions.

[Operator Instructions] We'll have our first question from Geoffrey Porges with SVB Leerink.

This is Anna Baran on for Geoff Porges. Two questions from us. First, what are your findings suggest about the activity of the inhaled JAK approach for other inflammatory lung diseases? Do you have evidence from this that the drug was absorbed and pharmacologically active? And then second, what are the hypothesis or hypotheses related to why there might be a difference in effects between the CRP low and the CRP high group and specifically why there'd be an infection with CRP low group?

Yes. Thanks for the questions. We've got evidence that, in fact, obviously, we've done population PK on the patients in the study. We achieved absorption overall as expected in the study. I think nezulcitinib performed overall in the study, I think, as expected as an inhaled drug -- as an inhaled drug would. I think it does because of the effect that we've seen, it does bode well for the study of this drug in other inflammatory conditions that caused acute lung injury. And for the second question, I'll turn it over to Rick.

Yes, that's right. And with regard to the different CRP cutoffs, here, in this case, we actually followed the science. This wasn't something that we looked at after the fact and just saw that there was an impact on CRP. This has been reported in the literature in a number of cases. As I mentioned, as I was going through the presentation, there's a paper in Lancet Rheumatology that describes the cutoffs of less than 150 or greater than 150. And what's clear in that longitudinal analysis is that in patients with greater than 150 CRP, these patients may have 4 clinical outcomes and may be more challenging to treat. So this is a great learning for us, again, following the science. And it may turn out that these particular patients need more than 7 days of dosing with nezulcitinib. So this is something we're still working to understand, and we'll take this information and apply it to our next stages of development.

Our next question comes from Liisa Bayko with Evercore ISI.

I'm wondering if you can kind of talk through what you've seen with the kind of inhaled JAK and sort of reconcile that with what we've seen with other JAK inhibitors in COVID like tofacitinib for example? And then also I wanted to understand what some of the next steps for the program are that you're thinking.

Sure. Well, to take tofacitinib since it was published in Mainland Journal of Medicine last week, I think the -- if you look at the population study,between the tofa population and the population of nezulcitinib, these populations are different. The tofa study included a number of patients who were Category 4 at baseline versus all of our patients were either Category 5 or Category 6. The other point is that we did include patients that were on non-mechanical ventilation at baseline, and I think noting the severity overall of the patients that we included in our study. So I think what you can see, whether it's the baricitinib experience or the tofacitinib experience or highlighted with the nezulcitinib data set is that there is a role for JAK inhibitors. And I think one of the things that we're pleased with on the nezulcitinib study is that it's really looking again at the CRP cuts that Rick talked about that the role of nezulcitinib at 7 days at the dose that we had dosed, which is 3 milligrams daily for 7 days with a 6-milligram loading dose on day 1, seem to be effective in that particular patient population based on the post-hoc analysis that we've done. Rick?

Yes. Just a couple of things to add on what Rick said. Maybe I'll focus on baricitinib. So importantly, the overall 28-day mortality rate that we saw in our study, I mentioned, was 12.5% in the placebo group. And that represents both clinical status score 5 and 6 populations who, by the way, are treated with systemic corticosteroids. So a large proportion of our patients had dexamethasone, limited use of remdesivir. But compared to the recently published paper with baricitinib in the COV-BARRIER study, this is around 1,500 patients, the overall mortality rate in these subgroups is quite similar. I think theirs was about 13%. So that speaks to the enrollment and performance of our study. Now with regard to what we do know about tofacitinib, baricitinib, ruxolitinib, it seems clear that JAK inhibitor certainly are having an impact in COVID-19. Now some of these end points that include the NIAID ordinal scale are challenging to achieve. And we've seen that in our study, but we've seen that with others as well. But what is consistent is that there are favorable outcomes on top of systemic steroids with regard to the objective endpoints of mortality and time to recovery. With regard to future indications, I think it's exciting that things are turning around with the vaccinations, but we can agree that COVID-19 is here to stay. And whether it's SARS, MERS, influenza, parainfluenza, we know that we are able to -- the results of the study suggest we're able to impact mortality in these particular patients. So we believe that next steps are nezulcitinib as a potential therapy for acute lung injury and, as we've discussed before, potentially more chronic indications like delay or prevention of lung transplant rejection.

Okay. Could you -- I know you've kind of just talked about this. But as you think about development, do you think you'd be kind of looking at certain subpopulations like those with low therapy? And maybe if you can just give a little more color again around why it just seems counterintuitive as you think about low therapy versus high therapy in the mechanism of action of JAK inhibitors.

Yes. I think, number one, I think the cutoff of 150, even these patients that are in the range of 75 to 150 or 50 to 150, these are patients that are in inflammatory states in the lung, either with their lungs and high inflammation overall. So I think it's not -- to cut at 150, you're probably looking at severe. And then critical patients, they're above the category -- above in that category of greater than 150. So those -- I think what we see is that clearly the patients that are less than 150, they have better outcomes overall. But even in the instance where they're all -- nearly all of them are on systemic corticosteroids, it seems that we were able to add something and benefit in the use of nezulcitinib at the dose of duration that we studied, likewise on greater than 150s that seemed -- that we worked in the dose and duration study. But I think that's relative to those iteration perhaps with nezulcitinib because of the severity of the underlying or criticality of underlying inflammation. Rick?

Yes. And Liisa, I think just important to remember that in our less than 150 CRP cutoff, that did represent 85% of the patients enrolled in our study. So this is not a small subset of patients. CRP is a well-established disease surrogate biomarker and is known to react in a very short time subsequent to inflammation. So again, this is an area where we followed the science here. This isn't something that just fell out of our data set. There's a number of reports on this.

Our next question comes from Anupam Rama with JPMorgan.

Just you mentioned in the press release meeting with regulators, what are the time lines to that? And then just a patient accounting question. The ITT was 210 patients. The post hoc for CRP was 201 patients. Were there 9 patients that didn't kind of get their labs done? Or how should we be thinking about that?

Rick, do you want to take him through the reconciliation of 201 to 210?

Yes. Thanks, Anupam, for your question. That's exactly right. What you said is that we had baseline CRP levels in 201 of those patients. So we reported the results in the post-hoc analysis for all patients for whom we had a baseline CRP level.

And then 205 patients were actually dosed and studied. There were 5 patients that were randomized and not dosed. On the online -- On the -- so obviously, we're -- this data is very new to us. We will wrap up the data packages and take them into regulators in both U.S. and Europe. And interactions with other government entities that we've been talking to over the last year or so and lay out a plan based on those conversations with FDA and Europe.

Our next question comes from Douglas Tsao with H.C. Wainwright.

Just curious, when you look at some of the other JAK inhibitors that have been looked at in COVID, did they see the same sort of correlation or sort of was the sort of efficacy correlated with C-reactive protein in the same way that you have -- I don't remember seeing anything suggesting that there is such a strong relationship between those levels.

Rick, do you want to start? I can...

Yes, sure. So thanks, Doug, for the question. With regard to the other JAK inhibitors, the reports that are in the literature, they have not reported anything about the CRP cutoffs as we have. There is a recent publication on lenzilumab that used the same CRP cut, so that would be an interesting thing for you to look at, but not with regard to the other JAKs. And I just think one of the other things, since we've been talking about the other JAK inhibitors today, I just want to make the point. With regard to a lung-selective JAK like nezulcitinib, we believe the differentiation here is designing a drug that can specifically target the organ of interest here, specifically targeting the lung, which is the primary of injury and cause of morbidity and mortality with COVID-19. So I just wanted to make that distinction. Go ahead, Rick.

That's good, really nothing to add there. I think we've looked at CRP in addition to ordinal scale on the other studies that we referred to on JAK inhibitors to date, have looked at ordinal scale and progression or not from ordinal scale.

Our next question comes to [ Eva Privitera ] with Cowen.

Could you elaborate on the evidence you see that nezulcitinib is achieving effective target engagement? What percentage of JAK inhibition is being achieved here?

Okay. Rick?

Yes. So with regard to specific target engagement in the study, Eva, it's really based on what we reported today regarding the improvements in efficacy measures around 28-day all-cause mortality and time to recovery, both in our intent-to-treat population as well as in the CRP cutoffs. But as we reported previously, both in a Phase I study in healthy volunteers and then especially in the dose-finding part of this study, we did also have an impact on mortality in that dose finding part of the study, but we saw a change in a number of biomarkers that are associated with target engagement with a JAK inhibitor. So taken together, the totality of evidence suggests we are achieving target engagement. We've also done a lot of preclinical work where we've done modeling to demonstrate the proportion of target engagement that we see in those preclinical models. And that's largely what helped us to decide on the 3-milligram dose to take forward into this part of the study.

[Operator Instructions] Our next question comes from Joseph Stringer with Needham & Company.

Two quick ones. Acute versus chronic in future sort of lung indications here. Is there any thing about the data today that would perhaps a safety profile in -- the 28-day study, but give you increased confidence that the drug could be effective and safe in more chronic lung indications? And secondly, on CRP, do you think that this will be a key factor in the sort of stratification of patients or the type of, for example, acute indications you would select going forward? Or is the CRP data that you shared today sort of more case specific to COVID patients?

Great questions. I think the CRP, I think, is broadly applicable for us in looking at conditions of acute lung injury. We haven't spent much time on the safety evaluation. But obviously, we're very, very pleased with the safety of nezulcitinib in the study. It really comes through as a well tolerated and not adding incremental safety concerns on top of baseline and on top of really standard of care. And again, as we said, 99% of patients had systemic corticosteroids here. Rick, anything to add?

Joey, just a couple of things on CRP. Predicting the endotype with regard to disease markers like CRP, it's still an evolving clinical science. But like Rick said, we do feel that these new data, the CRP and nezulcitinib, will help us going forward in our understanding and predicting these biomarker response relationships. With regard to acute versus chronic, so unlike a number of other agents or medicines that have been studied in COVID that are drugs that have been studied in thousands or tens of thousands of patients, we started last summer with a drug that had not been in clinical development. So we moved very quickly, as you know. And we took an approach that was set out to understand the benefit/risk profile. So at this point, with the totality of data that we've seen to date, we're very pleased with the safety profile of nezulcitinib. So we are looking forward to thinking about future indications, both acute and chronic for this agent.

It appears we have no further questions on the phone. I'd now like to turn the conference call back over to Mr. Winningham. Please go ahead, sir.

Okay. Thank you very much for joining us this morning. As you know, at Theravance Biopharma, we've got a very busy few months ahead of us. We're pleased to announce the nezulcitinib data today, and we look forward to updating you on the remainder of the programs as the summer progresses. Have a great day.

This concludes today's conference call. We thank you for your participation. You may now disconnect.