Theravance Biopharma, Inc. (TBPH) Q2 2017 Earnings Report, Transcript and Summary

Ladies and gentlemen, good afternoon. At this time, I'd like to welcome everyone to the Theravance Biopharma Conference Call. [Operator Instructions] Today's conference call is being recorded. And now I would like to turn the call over to Alex Dobbin, Head of Investor Relations. Please go ahead.

Good afternoon, everyone. Thank you for joining our second quarter 2017 financial results conference call and webcast. With me on the call today are Rick Winningham, Chief Executive Officer; Renee Gala, Chief Financial Officer; and Brett Haumann, Chief Medical Officer. Following our prepared remarks, we'll open the call for questions. A copy of the press release and the slides accompanying this call can be downloaded from our website, or you can call Investor Relations at 650-808-4045, and we'll be happy to assist you. Before we get started, we'd like to direct your attention to slide 2 of the deck and remind you that this conference call will contain forward-looking statements, which involve certain risks and uncertainties, including statements about our product pipeline, expected benefits of our products, the anticipated timing of trial results and regulatory filings and expected financial results. Information concerning factors that could cause results to differ materially from our forward-looking statements are described further in the company's filings made with the Securities and Exchange Commission. And with that, I'll hand the call over to Rick.

Thanks, Alex. Good afternoon, everyone, and thank you for joining us. We're incredibly pleased with our progress to date in 2017, which includes the delivery of meaningful clinical data across multiple programs within our portfolio. In our call today, we will provide an update on our key programs and anticipated milestones outlined on Slide 3, including data from the first cohort of patients in our Phase 1b study of our intestinally restricted JAK inhibitor, TD-1473. I'll start with an overview of our recent progress, and then Brett will share some additional color. Then Renee will review our financial performance. Then we'll open the call for questions. I'll start with our Phase 1b study of TD-1473 in patients with moderate to severe ulcerative colitis. In a separate press release this afternoon, we released -- we announced results from our first cohorts of patients. With four weeks of treatment in a small group of patients, 1473 demonstrated encouraging evidence of localized target engagement. More specifically, we saw improvements in measurements of disease activity without evidence of systemic exposure and with a favorable tolerability profile, all of which is consistent with the desired target product profile of this drug. These early findings represent an important step in our JAK inhibitor program, which is compelling directional evidence of local biological activity of 1473 and the absence of systemic exposure that can lead to unwanted side effects. Systemic exposure is a key question we're seeking to answer in this study, given its bearing on the premise of our program. Data from the first cohort suggests that the concentration of 1473 in the blood in ulcerative colitis patients is low and consistent with what we observed in healthy volunteers. Today's results are in keeping with the distinctive design principles we sought for 1473, to reduce inflammation in the intestinal wall while sparing the body from systemic immune suppression. On the basis of the data generated from our first cohort, which Brett will describe in more detail, we've begun planning for a larger induction and maintenance trial targeted to begin in 2018. Today's announcement from our JAK program follows on the heels of recently announced positive results from other key programs at Theravance Biopharma. Just last week, we were pleased to share top line data from the Phase 2b study of velusetrag, our highly selective 5-HT4 agonist, which demonstrated statistically significant improvements in gastroparesis symptoms and gastric emptying in patients receiving 5 milligrams of velusetrag as compared to placebo. This was the first clinical evaluation of velusetrag's effect on the symptoms of gastroparesis, and this has provided us with a positive proof of concept. This is incredibly important, not just for the company but for patients with gastroparesis. For nearly 35 years, metoclopramide has been the only FDA-approved agent in gastroparesis, and it's associated with risks that limit its use to no more than 12 weeks. So certainly, the opportunity exists for new effective and safe therapies in gastroparesis, and we are now preparing to meet with regulators to discuss the next phase of development. Additionally, at the end of last month, we and our partner, Mylan, reported positive results for the long-term Phase III safety study of revefenacin, our investigational long-acting muscarinic antagonist, or LAMA, in development as a once-daily nebulized bronchodilator for chronic COPD. When combined with the data from two Phase 3 efficacy studies, these results support our planned NDA filing targeted for the fourth quarter of this year. This anticipated registrational filing will be a major achievement for 2017. Meanwhile, the revefenacin Phase 3b study in COPD patients with low peak inspiratory flow rate, or PIFR, is ongoing, and results are anticipated in the first quarter of 2018. As we've previously noted, the PIFR study is intended to support commercialization of revefenacin and is not required for the NDA submission later this year. I'll now turn the call over to Brett to speak to today's results from the first cohort of our Phase 1b study of TD-1473 and provide other program updates. Brett?

Thanks, Rick, I'll start with our JAK inhibitor program, where our goal is to develop a highly differentiated treatment option for inflammatory bowel diseases, including ulcerative colitis. TD-1473 is designed to remain localized and only act within the gut wall, thereby, maximizing local anti-inflammatory efficacy and minimizing the systemic exposure that would otherwise lead to immunosuppression. The primary purpose of the Phase 1b study, outlined on slide 5, is to evaluate the safety and tolerability of 1473 administered once daily for 28 days and to assist the compound's plasma exposure following oral dosing. In addition, the study incorporates biomarker analysis and clinical, endoscopic and histological assessments to evaluate the biological effect of the drug. At this stage in the program, we are looking for compelling directional evidence across a range of measurements that the drug is having an impact on the disease and behaving in patients as we would expect. These measurements include biomarkers; clinical assessments, including frequency and bloodiness of stool; histological assessments from biopsy samples; and centrally read endoscopy scores. Taken together, these results will inform dose optimization for future clinical development. Today, we reported results from the first cohort of patients, which consisted of 10 patients receiving 80 milligrams of 1473 dosed once daily for 28 days and 3 patients on matched placebo. The number of patients is small, but the data we're seeing from the first cohort is promising, and it's been reviewed with a considerable measure of scrutiny and intensity on a patient-by-patient basis by both internal and external experts in ulcerative colitis. As we've noted on slide 6, data from the first cohort showed positive changes in key disease measures after four weeks of treatment. As noted in our press release today, systemic drug levels were low based on evaluation of plasma levels and consistent with what we saw in healthy volunteers. There was no evidence of systemic immunosuppression or infections, including no occurrences of zoster reactivation or alterations in total leukocytes, neutrophils or lymphocytes relative to placebo. In clinical assessments, all of which are based on the Mayo score, 7 of 10 patients on 1473 experienced a 1 point or more reduction in rectal bleeding subscore compared to 1 of 3 patients on placebo. 3 of 10 patients on 1473 experienced a 1 point or more reduction in endoscopic subscore compared to zero patients on placebo. As a reminder, endoscopies in our Phase 1b are read centrally as is a standard with larger-stage studies with other products being assessed in ulcerative colitis. 2 of 10 patients on 1473 showed evidence of mucosal healing compared to zero patients on placebo. We also find this to be noteworthy, given the relatively short treatment period and the level of improvement required to be classified as mucosal healing. Additionally, 2 of 10 patients on 1473 achieved clinical response as defined by total Mayo score compared to 0 patients on placebo. 4 of 10 patients receiving 1473 achieved clinical response by partial Mayo score compared to 1 of 3 patients receiving placebo. Biomarker changes were also encouraging with reductions in serum C-Reactive Protein and Fecal Calprotectin, both surrogate markers of active ulcerative colitis. And a reduction in pSTAT1 levels in tissue biopsies of patients treated with 1473. We're also assessing more subtle treatment effects that are not necessarily detected by the Mayo subscores. For example, one patient on placebo had an improvement in stool frequency from 3 to 2 per day, resulting in an improvement in partial Mayo score. In contrast, one patient on 1473 had a change in bowel frequency from 8 to 5, representing an important improvement despite the fact this was not reflected in the partial Mayo score. From a safety standpoint, 1473 was generally well tolerated with no moderate or severe adverse events related to study drug, no adverse events leading to withdrawal and no clinically relevant changes in blood laboratory parameters, including no alterations in NK cells or lipid levels relative to placebo. Obviously, we're very pleased to share this data with you today. As we've stated, at this stage in the program, we're looking for compelling directional evidence of local target engagement. And the totality of data from the first cohort supports this target product profile. For the two remaining cohorts, we've decided to run them in parallel as opposed to in sequence, and this portion of the study is underway. The second and third cohorts will receive 20 milligrams and 270 milligrams of 1473, respectively or match placebo. Understanding the effect of 1473 at both lower and higher doses is useful for dose selection in our next clinical study. But these cohorts will not be rate-limiting to our preparation for the next study. Enrollment in this Phase 1b study has proven to be slower than expected, driven largely by short duration of therapy with patients favoring participation in longer-term studies. Our intent for the next study, a formally powered efficacy study, is to assess the effect of 1473 as an induction and maintenance treatment. To that end, the team is now working in earnest to initiate a large induction and maintenance study in 2018. Turning to velusetrag on slide 8. We were very pleased by the data from the Phase 2b study shared last week, which showed a favorable efficacy and safety profile for the lowest dose tested, which was 5 milligrams. The study enrolled more than 200 patients with gastroparesis, split roughly 50-50 between diabetic and idiopathic gastroparesis. The study assessed three doses of velusetrag -- 5, 15 and 30 milligrams -- and placebo dosed once daily for three months. The study was powered to identify at least one velusetrag dose that was more effective than placebo in reducing the symptoms of gastroparesis. The primary endpoint was the placebo-adjusted change from baseline in a patient-reported symptom score referred to as the GCSI, or Gastroparesis Cardinal Symptom Index, after 4 weeks of dosing. The 5 milligram dose demonstrated statistically significant improvements in gastroparesis symptoms as compared to placebo with a nominal P value of less than 0.05. In contract, the 15 and 30 milligram doses did not show statistically significant symptom improvements versus placebo. All three doses of velusetrag showed highly statistically significant improvements in gastric emptying as measured by scintigraphy at the 4-hour time point on day 28. As we turn to slide 9, the positive response seen with 5 milligrams in the GCSI score was also seen with the Gastroparesis Rating Scale, or GRS, with statistical significance seen at weeks 1 through 4 and at week 12. As a reminder, the GRS is a proprietary symptom tool we're developing through academic collaboration and in alignment with FDA guidance. We were pleased to see consistent improvement across all individual subscales included in the GRS tool with statistically significant improvements in fullness, bloating and upper abdominal pain for the 5 milligram dose. In terms of safety, the 5 milligram dose was generally well tolerated with rates of adverse events that were comparable to placebo. Serious adverse event rates were low across all treatment groups. Three events on placebo, four on 5 milligrams, two on 15 milligrams and three on 30 milligrams. Of these, only one was assessed as likely related to study medication, and this event was subsequently found to be in the placebo group. Rates of early discontinuation due to side effects were lowest in the 5 milligram dose group with only 3% of patients withdrawing compared to 8% on placebo, 10% on 15 milligrams and 7% on 30 milligrams. In closing on velusetrag, the Phase 2b study provided us with some key new learnings on the effect of different doses on the symptoms of gastroparesis, including the fact that higher doses may increase side effects and counteract any symptom improvements being measured by symptom instruments despite all the doses increasing gastric emptying. The study has identified the 5 milligram dose as providing both symptom relief and meaningful improvements in gastric emptying. We look forward to meeting with regulators to discuss validation of the GRS tool and the next phase of development for Velusetrag. Now moving on to revefenacin in slide 11. We and Mylan recently shared positive results from our Phase 3 long-term safety study of more than 1,000 patients with COPD, the last clinical assessment prior to our planned NDA filing later this year. Results showed that revefenacin was generally well tolerated with no new safety issues identified. The data from this 12-month safety study built on our observations from the previous 3-month efficacy studies and suggested revefenacin has a favorable safety and tolerability profile when dosed chronically, either as a stand-alone therapy or when taken as an add-on to other COPD therapies, including combinations of ICS and LABA. As of today, there are no approved nebulized LAMAs despite a significant number of patients needing or preferring nebulized therapy for the treatment of their disease. Having achieved positive efficacy and tolerability data in our Phase III program, we and our partner, Mylan, believes that revefenacin is well positioned to address this important patient need. We remain on schedule to submit the NDA in the fourth quarter of 2017, which is the next step towards our goal of delivering the first once-daily nebulized bronchodilator to the COPD patient community. To complement our Phase 3 program, we're conducting a Phase 3b study in approximately 200 COPD patients with low PIFR. This study is not required for filing but rather is being conducted to better understand the needs of these patients and to support commercialization of revefenacin if approved. We believe that this patient group may benefit particularly from the use of nebulized therapy because they're not able to inhale with enough force to benefit fully from hand-held inhalers. We're pleased with our progress on this study and expect to have results in the first quarter of 2018. I'll now pass the call over to Renee to provide a financial update.

Thank you, Brett. Revenue for the second quarter of 2017 was $3.5 million, primarily related to U.S. net product sales of VIBATIV. Research and development expenses for the second quarter of 2017 were $42.9 million as compared to $32.1 million in the same period in 2016. The increase in R&D expense is primarily attributed to costs associated with the progression of our key programs. Second quarter R&D expense includes $4.9 million in noncash share-based compensation expense. Selling, general and administrative expenses for the second quarter of 2017 were $24.3 million as compared to $20.3 million for the same period in 2016. The increase in expense is primarily due to employee-related costs, external expenses related to G&A and share-based compensation, partially offset by a reduction in external expenses related to commercialization activities. Second quarter SG&A expense includes $5.5 million in noncash share-based compensation expense. We remain in a well-capitalized position with $498.3 million in cash, cash equivalents and marketable securities at the end of the second quarter. For the full year 2017, based on current plans and expectations, we project our net operating loss, excluding share-based compensation, to be in the range of $205 million to $215 million. This guidance is above previous full year guidance of $195 million to $205 million, primarily driven by our decision to accelerate spending associated with the next phase of development in the JAK inhibitor program. Finally, I'll remind you of our economic interest related to the GSK respiratory program. GSK will pay upward-tiering royalties ranging from 6.5% to 10% on worldwide net sales of the Closed Triple in COPD, if approved, and Theravance Biopharma holds an 85% economic interest in those future potential cash flows. The Closed Triple is currently under review with regulatory authorities in both the US and the EU. Based on time lines provided publicly by GSK, we could expect approvals in both regions before the end of the year and could begin receiving cash flows related to this program as soon as early 2018. GSK is responsible for all development and commercialization costs related to the Closed Triple with no cost being borne by Theravance Biopharma. GSK's ongoing investment in clinical studies for the Closed Triple includes the Phase III impact study in COPD, which is expected to read out later this year. This study in 10,000 patients with COPD is designed to assess the impact of the Closed Triple on exacerbations of COPD. Additionally, in late 2016, GSK initiated the Phase 3 CAPTAIN study to assess the Closed Triple in asthma. This study is expected to read out in 2018 and, if positive, would be followed by regulatory filings also in 2018. Now, I'll turn the call back over to Rick.

Thanks, Renee. Moving to our commercial infrastructure. The branded antibiotic space has been challenging in 2017 primarily due to the introduction of multiple suppliers of generic daptomycin and its effect on the outpatient market. We continue to focus our label expansion strategies and optimizing our operational effectiveness prior to the potential launch of revefenacin. Our ongoing Phase 3 study in primary bacteremia is progressing. Additionally, our recently completed patient registry study, or TOUR, has provided and will provide valuable information about the use of VIBATIV in real-world clinical settings, including reports of positive clinical responses in patients with bacteremia, endocarditis, osteomyelitis, skin and respiratory infections. We presented 15 TOUR-related abstracts and posters at conferences to-date, and additional publications will detail new findings from the registry over the remainder of 2017. We see the acute care setting as a very important inflection point in patient identification. We look forward to the potential to leverage this organization for the promotion of revefenacin. We're in an exciting period of progress at Theravance Biopharma, driven by the breadth of our portfolio, the productivity of our internal R&D engine and a robust business model. Turning to our list of achieved and expected upcoming milestones on slide 12. Earlier this year, we reported positive findings for our multiple ascending dose study of 1439, our second NEP inhibitor to be evaluated in a Phase 1 program. Results were in line with what we saw on a single-ascending dose study supportive of our target product profile, specifically 24-hour target engagement, nonrenal clearance and a favorable safety and tolerability profile. In the last few weeks, we've reported promising clinical findings in three of our key programs -- revefenacin and the treatment of COPD, positioning us to file for our revafenacin NDA in the fourth quarter; velusetrag in gastroparesis; and 1473 in ulcerative colitis. The data on 1473 is both exciting and important, not only for paving the way for an induction and maintenance study in 2018 but also in providing us with greater confidence in our ability in research to predict local treatment effect in tissue without relying on systemic drug concentrations in the blood. Turning to the remainder of 2017. In addition to the TOUR publications, we expect data from the Phase 3 IMPACT study of the Closed Triple and its potential approvals in the U.S. and EU as Renee had mentioned previously. Turning to 2018, the Phase IIa study of TD-9855 in neurogenic orthostatic hypertension is ongoing. After seeing encouraging responses in a majority of patients enrolled in its single-ascending dose portion of the study, we're progressing into a 20-week open-label extension to assess durability of effect in treating NOH patients. We intend to seek an expedited development pathway for this debilitating orphan condition, and we'll provide updates on this study as we get closer to data. Additionally, we look forward to the following near-term milestones, data on the remaining cohorts of the 1b study of 1473, data from the 3b study of revefenacin in patients with low peak inspiratory flow rate, and finally, the Phase 3a CAPTAIN study and the Closed Triple in asthma patients. In closing, we're proud of the entire Theravance Biopharma team and are incredibly excited about both our progress to-date and the opportunities that we have ahead. We believe our programs represent valuable and differentiated product opportunities with the potential to have a meaningful impact on patients' lives, change how serious diseases are treated and create meaningful value for shareholders. I'd now like to turn the call over to the operator for questions.

[Operator Instructions] Our first question comes from Geoff Porges with Leerink.

Appreciate the disclosure of the discontinuations on velusetrag. Can I just follow up on some of the comments on 1473. Wondering if you have the mean change in the partial Mayo in the active and the placebo. I know the numbers are tiny, but I think people a lot of people are looking at that. Secondly, Brett, could you comment on what we’re seeing on the endoscopies, particularly whether there was any difference in the effect seen in the upper bowel versus the lower bowel. Is the dispersal of 1473 and the effect of 1473 consistent across the colon or does it vary by location? And then lastly, or second to last, actually, why 270? Now you've seen that you're not getting any systemic absorption. You don't appear to be getting any systemic effects. Why wouldn't you be taking several steps up? Maybe 160, maybe 240. Why go all the way to 270? It looks as though you made up your mind that 80 as the dose. And then lastly, could you talk a little bit more about the induction maintenance? Is it safe to assume that that's going to be a several-hundred patient study with a 12-week duration and some sort of follow-up? Is that the sort of thing that you have in mind? I know you don't have a firm study design yet, but it would be helpful if you have any color on that. Thanks.

Geoff, this is Rick. And I'll let Brett handle most of these. That was a pretty rich set of questions. So, Brett?

Thanks, Rick. And Geoff, thank you. In fact I'll tie your first and your fourth questions together. You asked about changes in partial Mayo, and then you also asked about the design of the induction study. And I think those answers are linked in some way because, as we reported on before, the purpose in this initial Phase 1b study with a small number of patients is to look for evidence of directional effects. So that's consistent biological effect or compelling directional evidence that we're seeking really is best reflected in responder rate. So that's why we reported on the responder analysis that we've described. You're absolutely right that in such a small population, going to formal analysis of mean changes in partial Mayo was never the intent. We were always looking at broader patterns of performance. And in fact, that's what we've reported today. It does bring me on, though, to the intent with the next study, which, as you rightly described, would be a much larger piece of work. It's a formally powered study looking at induction. You'd mentioned 12 weeks. In fact, the convention that we believe we would mirror is that of tofacitinib with 8 weeks induction rather than 12. But our focus is also to extend the duration beyond induction to maintenance. Now this is, obviously, still subject to our interactions with regulators, who will be enormously helpful in defining and agreeing what the scope of that work would be. But really, again, to reiterate, the value in what we've seen here, this compelling directional evidence that we're disclosing today gives us confidence to be planning in earnest for the next study. It relates also to your second question, which - oh, beg your pardon, your third question, which is around doses, and I would say that our selection of a broad range of doses in this study is intentional. This is still what we would deem to be dose-range finding where although, you're right, we're very pleased with what we're seeing at 80 milligrams, we remain interested in the extreme ends of the envelope, so to speak. So at 270 milligrams, if we continue to see a benign safety profile and improvements as we reported here, then it tells us we're at the top of the dose response around 80 milligrams. Similarly, if we're done at 20 milligrams and see no effect at all, that's enormously instructive for the next phase of development because we know we don't need to be that low in order to identify a subeffective dose. So I think the reasons for picking such a broad range at this point are still compatible with this phase of development. It is early, but it is intended to be initial dose-range finding and will be instructive I think to informing the dose selection for that next study. Your last question was around endoscopy, and you'd asked about differences between upper and lower intestines. Excellent question. Actually, we had allowed patients to come in with ulcerative colitis that could have been either in the descending portion of the colon or it could have been entirely throughout the colon. Now the spectrum of disease the patients have often vary. Some patients with ulcerative colitis only have their disease in the lower part of the large intestines. Others have it distributed throughout. What's important to emphasize in this study is that wherever the inflammation was for a particular patient was whether biopsy was conducted before therapy and then also after therapy of 4 weeks. In other words, we measured exactly where we were biopsying the inflammation and then went back to that site. So for some patients, it would have been in the lower intestine and where we'll have pre and post dose assessments. Those are reflected in the endoscopic measures as well as the biopsy samples, and some patients would have had it in the upper end of the large intestine, and that's where we would have biopsied again at the beginning and end of therapy. So hopefully, I’ve answered your question. It did vary according to the patient's individual disease state.

So, Brett, was there any difference in the pattern of response between the patients who had the relatively localized disease and the patients who had the more extensive or pancolonic disease?

We were impressed by the overall improvements that we've seen, so although we're not reporting on individual patient responses today, the improvements that we're seeing and bear in mind, this is not just on endoscopy but broader, looking at biomarkers, looking at clinical response, were still consistent with what we were seeking to observe, which was this consistent improvement in patients on active therapy.

Thanks Geoff, I'd just also to add to that, I think we're terrifically excited about the 80 milligram dose because we were able to deliver this dose and we were able to get signs of activity. And it really gives us an opportunity to, as Brett said, to explore the dose range given that the PK at 80, as we've noted, is right in line with the healthy volunteers, and that was just great news for us that there wasn't additional absorption of the drug into the bloodstream and, therefore, gives us several degrees of freedom with regard to going forward on optimizing the dose. So next question, operator?

[Operator Instructions] Our next question comes from Umer Raffat with Evercore.

I guess just to start off, Rick and Brett, it seems you guys are not disclosing the average partial Mayo score. Maybe just to start off, I was curious why that is.

So this is Brett. I think the response is very much as I'd mentioned, suggesting this Phase 1 study, as we reported, I think, several times, our purpose is not to focus on one endpoint or to do a statistical assessment of one clinical endpoint. As attractive as that may have - may be, particularly because we've seen that with drugs like tofacitinib - those are subject, obviously, to much larger studies in the evaluation of balanced assessment between active and placebo. And so in the study of this nature, in this early phase of development with only one cohort of data, our view has always been to look at a more holistic approach, a balance of evidence as we've described. And I think we've been very consistent in describing this as we've spoken about this study. So it's on that basis - we're not preoccupied with one symptom endpoint and indeed not in the position to be able to power for any evaluation of any endpoint, including partial Mayo. But what we have reported as you've seen is the partial Mayo response rate, and so we've included that in the list, and it needs to be seen in the context of the other improvements that we were seeing, for example, the improvements on endoscopy, the improvements on biopsy markers and then perhaps most importantly for us is this absence both of drug in the blood and of systemic consequences of that because that's what sets us apart from tofacitinib. This distinction in our minds is not just that we're seeing local effect. It's that we have absence of systemic effect. And drug levels that are entirely consistent with healthy volunteers really reaffirming that patients with an actively inflamed mucosa don't have high levels of drug, which we may have - people may have speculated might have been the case. This study has really helped to confirm that the drug levels are entirely in keeping with we've seen with healthy volunteers.

Okay, got it. If I may, I did want a follow-up outside of the actual average. Perhaps maybe, Brett, could you clarify how you defined response on total and partial Mayo score? And also, was there any overlap in the 7 patients that had a 1 point reduction on rectal bleed versus the 3 patients that had the endoscopic reduction? And also, I noticed on the partial Mayo, there was no commentary on the stool frequency or the PGA for that matter, so maybe just some color there. And then finally, any commentary on baseline.

Great, thank you. I've jotted down your comments. I'll try and capture those. And I think what you'd asked about was, obviously, the overlap with patients, so this was around people who'd had rectal bleeding as well as seeing effects on other elements. And it's fair to say, without going into the specifics because we're describing the overall effect, that there are patients who have overlaps of this various elements as you would expect. So in order to qualify, for example, on a total Mayo, and you'd asked about how they were defined, it is actually referenced in our press release as to how these definitions are specified. And they are consistent with the way that tofacitinib has evaluated its Mayo assessment. So without going into the descriptions because they can be fairly convoluted technical descriptions, I would refer you to the publication that we've quoted in our press release because that fully describes the definitions that we were using in our own assessments. But again, back to the point of overlap. There are patients who have both improvements in their rectal bleeding as well as improvements on their endoscopy as you would expect, and these patients then are contributing to the improvements that we see at holistic levels with measures like the full and partial Mayo. Stool frequency, it's a very good point, Umer, I mentioned in our script here, but it's not in the press release, that we are looking at more subtle changes within, for example, stool frequency. One of the things we've recognized with instruments like the Mayo is that the subscore for bowel frequency is relatively insensitive if you're moving from a very high frequency of bowel motions down to a less frequent and the example that I cited, a real one, actually. We saw this in the study. One of our patients on placebo moved from a bowel frequency of 3 to a bowel frequency of 2. Now that may be viewed as a fairly benign change, but actually, by way of the definition of the Mayo score, this patient actually moves an entire category in shifting from 3 to 2. In contrast, one of the patients that we had on our active on 1473 was quite severe. Actually, this patient had an average of 11 additional bowel movements per day. And we were able to move that down to 5. So that was a big change, and that change, unfortunately, still falls within 1 category. So that would not be picked up on partial Mayo. But it's those sorts of deltas, those changes that remain of real interest to us. Because when we were looking at overall biological effect, we still think those are important. Now in larger studies, we will have to accept that the partial Mayo, with its limitations or even the full Mayo will be able to detect these effects in larger numbers of patients. That's the purpose of running larger studies. But again to reiterate in the small study, we really are interested in some of these more subtle effects that may not all be expressed in some of these instruments. You mentioned PGA as well. This is the physician global assessment. We haven't deliberately omitted that at all. It's actually a subscore within the partial Mayo and the full Mayo. I think it's fair to say that the current view within the community is that the PGA isn't as strong a subscore as some of these others are. In fact, there are now moves afoot to question whether the PGA should be included in further studies or in future studies, and in fact, the FDA is weighing in on this. So although we are - we saw improvements in PGA, we're putting less emphasis on this because we think that the community in the future will be placing less emphasis on the PGA itself. I'll pass over to Rick to see if you had any additional comments, Rick.

No. Umer, I'd just like to just reconfirm one more time the underlying purpose of this study. The underlying purpose, we were working to get directionally compelling evidence of target engagement in patients with ulcerative colitis. And I think the - we talked several times publicly about the importance of looking at the totality of the data to develop a perspective on compelling directional evidence. And I think with the reductions in rectal bleeding and the endoscopy improvements and seeing some mucosal healing, this was just very, very encouraging to us and probably more to me than any other factor. We achieved that without seeing anything happen differently in these patients than happened in healthy volunteers relative to absorption. And we were meticulous, I would say, in terms of really measuring the effect of any drug in the blood, and we just didn't see anything. And that was quite gratifying in light of the biological activity in ulcerative colitis that we did see.

Rick, if I may, are you aware of any data on the 20 and 270 milligram?

The only data that I'm aware of is that we're enrolling in the next 2 cohorts.

Our next question comes from Brian Skorney with Robert W. Baird.

This is actually Neena on for Brian. So I just had a question about the enrollment time lines for the Phase 1b study of 1473. I know you just mentioned that you're enrolling the other 2 dose cohorts. Now are you having any trouble enrolling patients? And can you walk us through the timeline and what that might mean for enrolling a larger study? It just seems like the timeline here is a little bit longer than we had expected.

Thank you, Neena. This is Brett. That's an excellent question, and we alluded to it in the script. But let me expand on that. I think it's fair to say that our enrollment rate in this 4-week study has been slower than we were expecting. It really appears to be predominantly fueled by the fact that patients in considering their choices when they enroll into a study are presented with the option to join our 4-week study or to go into much longer studies, including induction and maintenance studies that are offering minimums of 6 months therapy, in some cases, much longer, a year or in fact even longer through the use of open-label extension studies. And this is a practice which has become something of a convention in the treatment of patients with ulcerative colitis. The patients in clinical trials expect to be able to have ongoing therapy options through the clinical program. And so I mentioned that because it has not only hampered our enrollment in a 4-week study, but I think it's also educated us that enrollment in larger studies will overcome this. And really, that's the focus with our next piece of work. I don't believe to answer your question directly, that we'll see similar impediments to the enrollment in our next study because there we're offering induction and maintenance. The patients will have the opportunity to be treated for a longer period of time. But I do note that, actually, certainly with this 4-week design, one of the unintended learnings, I think, for us has been that short-term treatment is really not an inducement for patients to participate in clinical programs.

The only thing - this is Rick. The only thing I would add is that, of course, we were enrolling patients with moderate to severe disease. We were enrolling them on a new therapy, obviously, without evidence that we were going to have biological activity, certainly, because we just tested in healthy volunteers. Now I think we have a level of evidence, of target engagement in these patients. And importantly, I think, again, as I've said and Brett said, we have no systemic exposure that's greater than the small amount - very small amount we saw in healthy volunteers. So we look forward to pushing the 1b along and preparing to go into the induction and maintenance study.

Our next question comes from Louise Chen with Cantor Fitzgerald.

The first question I had here was on the second and third cohort. Are you going to be looking at similar endpoints to what you showed in the first cohort and that means there won't be any mean change in partial Mayo score included there? And then on the induction and maintenance study, what are you going to be exploring there in terms of endpoints? And how many patients do you plan to enroll? And then last question is just on TD-9855, why you move that study out in terms of the readout into next year? And then also the advantage of your 20 weeks of treatment versus what you see on the market today.

Sure. I'll take the 9855 question then give the JAK question to Brett. So 9855, the key here for us is to show duration benefit. If we were to get patients out with the responses at 20 weeks, that would be a remarkable duration benefit for patients with neurogenic orthostatic hypertension, given the current label drugs only have activity for 2 weeks. So we're hoping to see an extended duration out to far beyond what existing medicines have seen, and that's the reason really for pushing it out into 2018 is expanding it this out to 20 weeks. I did mention in the script, we'll be seeking an accelerated pathway for this orphan disease. So back to Brett on the - on what we're looking for in the next 2 cohorts in the JAK program.

Thanks, Rick. Louise, great questions, actually. For your first question, are we looking at the same endpoints in the second and third cohort? The answer is yes. We'll continue to use the same approach that we've applied in the first cohort because that will give us a frame of reference to compare against when we look at the low dose compared to [Technical Difficulty] dose and then, of course, the high dose. Bear in mind that our overarching purpose in this is still to look for that totality of evidence, looking at all endpoints to see whether we're seeing that consistency of response and by extending it now to the three doses whether there any patterns that tell us that higher or lower doses will be less or more beneficial. So we find this all. We believe this is still very instructive to the next piece of work. That brings me onto your second question. The induction and maintenance studies are going to be much like you've seen with competitors in this class, with other people who've set the precedent. So the purpose of new program is particularly instructive to us. We're interested in the 8-week induction. We note that they looked at things like partial Mayo as a driver of their powering, and of course, we would look to do the same so that it creates a frame of reference. I will note though that the industry standard and the regulatory view is shifting a little bit in terms of the rigor of the partial Mayo and whether there are alternates with some changes on the partial Mayo. And I mentioned earlier the main one that's being debated is the utility of the PGA, the physician global assessment. And so although standing here today we see partial Mayo as a really important element of that study, I couldn't today confirm whether that's likely to be our primary endpoint because really that's the purpose of engaging regulators as we begin to plan for the study in earnest now. So it is our intention fueled by the confidence we've seen here to engage with regulators to define the endpoints, the duration, the extension from induction into maintenance, which is a purpose we are focused on with that study. You also asked about sample size. And again, that would be really finalized when we're talking to regulators about endpoints and powering. It often stems from that. But I think it would be fair to say today that you should expect to see hundreds of patients in that study as opposed to the very small numbers we have here because if we're pursuing those endpoints and looking at mean changes and deltas, we need the power to be able to detect that. And for that, we would need hundreds of patients. So expect it to be a larger, more significant piece of work.

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

Thank you very much, operator. I'd like to thank everyone for participating in our call today. And thank you for the energy over the past several weeks as we've read out three very important studies to the company's future. We look forward in the upcoming weeks and months keeping you updated on our progress. Thank you.

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