Insmed Inc (INSM) 2017 Q1 法說會逐字稿

Good morning. My name is Chris, and I will be your conference operator today. At this time, I would like to welcome everyone to the Insmed First Quarter 2017 Financial Results Conference Call. (Operator Instructions) Thank you. Laura Perry, Investor Relations, you may begin your conference.

Thank you, Chris. Good morning, and welcome to today's conference call to discuss our first quarter financial results.

Before we start, let me remind you that today's call will include forward-looking statements based on current expectations. Such statements represent our judgment as of today and may involve risks and uncertainties that may cause actual results to differ from the results discussed in the forward-looking statements. Please refer to our filings with the SEC, which are available through the SEC's website at www.sec.gov or from our website for information concerning the risk factors that could affect the company.

Joining me on today's call are members of the Insmed executive management team, including Will Lewis, Insmed's President and Chief Executive Officer; John Goll, Vice President, Finance and Corporate Controller; and Roger Adsett, Chief Commercial Officer. For today's call, Will will provide a corporate update, and then John will review the first quarter financials. Following brief closing comments from Will, we will then open the call for questions.

At this time, let me turn the call over to Will.

Good morning, everyone, and thank you for joining us today. 2017 is undoubtedly the most significant year in our more than 10-year history as a company. It will be truly transformative as we reveal the top line results of our Phase III CONVERT study. Positive results would be an important milestone in our efforts to reach our goal of building a self-sustaining biopharmaceutical company that addresses the unmet needs of patients with rare diseases. Positive top line results from the CONVERT study have the potential to put Insmed on a new trajectory of growth.

The CONVERT study is investigating ARIKAYCE, liposomal amikacin for inhalation or LAI as a treatment for the thousands of patients who suffer from refractory nontuberculous mycobacteria lung disease or NTM caused by Mycobacterium avium complex or MAC. This is a potentially debilitating, pervasive and costly disorder with no approved therapies available in either the U.S. or Europe. Typically, the disease affects an older population and is associated with an increased mortality rate that is further complicated by multiple comorbidities in these patients. The best available information informs us that the 5-year mortality rate for NTM is approximately 25%. It is far higher for those patients who are refractory to current therapeutic approaches.

To remind you, LAI is a combination of the aminoglycoside antibiotic amikacin. Amikacin is encapsulated in a specialized liposome made up of DPPC and cholesterol, which enhances its potential to treat certain infections. LAI is taken directly into the lungs through a customized ultrasonic nebulizer. Following the achievement of our patient enrollment objective late last fall, we remain on track to report top line data as planned later this year. We expect that this will occur in the September time frame, plus or minus a month. CONVERT is the largest study in NTM undertaken today. It is a global study being conducted at over 125 sites in 18 countries. This is a rigorous and robust study that was informed by 3 things: Our encouraging Phase II study results, input from thought leaders in the space and discussions with regulators. When fully completed, we believe the data will inform the most thorough understanding of the treatment of NTM yet undertaken.

We have completed randomization of a total of 336 patients with 2:1 randomization in favor of the LAI treatment arm. Patients receive either LAI plus standard of care or standard of care alone. The primary endpoint is the proportion of patients achieving culture conversion by Month 6. Culture conversion is defined as 3 consecutive negative monthly sputum cultures. This is the gold standard for measuring culture conversion and is commonly used when evaluating patients with diseases like tuberculosis. The study is powered at a 90% confidence level to demonstrate a 15% difference between the treatment group and the control group. We've designed the study to mirror actual clinical practice such that patients who achieve culture conversion by Month 6 will continue in the CONVERT study for an additional 12 months of treatment. Patients who do not culture convert have the option of enrolling in the INS-312 study. INS-312 is an open-label study where patients will receive LAI plus guideline-based background therapy for 12 months. This extension study will allow us to evaluate several important impacts of the drug's use: Number one, the impact of remaining on the drug for longer than 6 months to help answer the question to patients convert after more than 6 months of therapy; number two, the durability of culture conversion after exposure and removal of therapies; and number three, the impact of LAI on other longer-term clinical objectives such as mortality.

We continue to remain blinded to the CONVERT efficacy data. However, we can update you on some aspects of the trial conduct thus far. As I mentioned, all patients have been randomized at this point. The pace of enrollment was studied throughout the course of the study, and as of today, we're only waiting for approximately 25% of randomized patients to complete the 6-month treatment period. The dropout rate seen to date continues to be consistent with our expectations. Additionally, there now have been a total of 7 safety monitoring board reviews thus far. The latest review, as with previous reviews resulted in a recommendation that the study continue without modification.

In the event of a positive outcome around the primary endpoint in the CONVERT study, we are positioned to move quickly to file for approval in the U.S. under Subpart H as planned. Subpart H enables sponsors to file on an accelerated basis, and subsequently complete post-market activities to verify and further describe clinical benefit. We continue to believe that positive results from the CONVERT study will provide sufficient data to support full approval. We also plan to pursue regulatory submissions in other geographies. And from a manufacturing and supply chain perspective, we believe we are in a solid position as we prepare for commercialization with 2 sources of clinical and commercial supply up and running.

We believe that LAI has the potential to dramatically change the way patients with NTM are treated, and we're excited to show results with you in the months to come. It is important to link the impact the medicine may have on patients to the value this represents to the payment entities that support these patients. A review of data from an insurance database has confirmed for us that NTM patients tend to be costlier than age-matched controls across multiple metrics. These measures include frequency and duration of hospital stays and emergency room visits. These costs drop dramatically if patients are able to achieve culture conversion. This drop in cost is largely driven by reduced hospitalization.

For context, the costs for NTM patients exceed the annual cost for asthma, COPD or tuberculosis. Moreover, in the first year following diagnosis, costs to treat NTM exceed those associated with lung cancer and idiopathic pulmonary fibrosis and are almost more than those 2 combined. Additionally, the mortality rate for these patients remains high at approximately 25% at the 5-year mark and higher still for refractory patients, exceeding that for patients with COPD and bronchiectasis. Clearly, there is an unmet need here and the survey of the insurance database suggests that a new therapy that achieves durable culture conversion could dramatically improve the treatment paradigm for patients and their treating physicians, while also demonstrating real economic value for payers.

Finally, disease awareness and physician education have remained a strategic priority for us. We have built a non-branded disease awareness campaign designed to help physicians understand the importance of an appropriate diagnosis. Through this campaign, we provide valuable information on how to recognize the signs and symptoms early in the treatment process to better manage NTM-1 disease. Our website, ntmfacts.com, shares patient perspectives that illustrate the daily challenges of living with NTM. We will be hosting a booth at ATS next month to raise awareness about the disease and the importance of proper diagnosis.

Let me now turn to an update on our next highest clinical development priority, INS1007. This is a reversible oral DPP1 inhibitor we licensed from AstraZeneca last October. This Phase II ready asset complements our LAI program, directly leveraging our expertise in the rare pulmonary space. We will also be targeting non-CF bronchiectasis as an initial indication. This is a rare chronic pulmonary disorder in which 40% of patients are believed to also have structural lung disease. Similar to NTM, there are no FDA-approved treatments for this disease. Patients with this condition experience a vicious cycle of bacterial colonization, inflammation, airway destruction and abnormal mucus clearance. The current standard of care involves utilizing antibiotics to manage the infections and exacerbations rather than targeting the underlying inflammation.

As a novel oral inhibitor of DPP1, 1007 impairs the activation of 3 neutrophil serine proteases that reside within our neutrophils: Neutrophil elastase, proteinase 3, and cathepsin G. When they are released in excess of endogenous inhibitors, they become destructive and trigger excessive mucus release and pro-inflammatory events. This then contributes to lung matrix destruction and inflammation. We're working to finalize our plans for a Phase II study, which we expect to initiate in the second half of this year. Key next steps include a meeting with FDA where we'll discuss our proposed study design and development program in non-CF bronchiectasis. We look forward to sharing more about this program as it advances.

Beyond non-CF bronchiectasis, 1007 has also shown potential in multiple disease states, and we are reevaluating the potential for Phase II studies in additional indications. The Insmed team is also working to advance a number of other programs in rare disorders. The most advanced program from our internal pipeline is INS1009, a nebulized prodrug formulation of treprostinil. Phase I data from this clinical candidate is encouraging and we continue to believe that this program holds tremendous value. We are continuing to evaluate our options to further advance this development.

We are at a pivotal point in our evolution at Insmed and are excited about the year ahead. We look forward to sharing top line results with you as they become available. Our regulatory activities for LAI continue in the U.S. and other geographies as we also focus on building awareness around NTM lung disease. Investments in our pipeline beyond LAI continue with the advancement of 1007 and our plans to bring that program into Phase 2 development in the second half of this year.

With that, let me introduce you to John Goll, an important member of the Insmed team. John serves as Vice President Finance and Corporate Controller. He has been with Insmed for more than 3 years and has over 20 years of financial experience within the pharmaceutical industry. He provides us with continuity and managing our finances and will continue to ensure a smooth transition as we near the end of our search for a new CFO.

With that, I'll turn it over to John for this quarter's financial update. John?

Thanks, Will. Good morning, everyone. Turning now to our quarterly results. This morning we reported a net loss of $37.4 million or $0.60 per share compared with a net loss of $33.5 million or $0.54 per share for the first quarter of 2016. Research and development expenses increased to $22.3 million compared to $20.5 million in the first quarter of 2016. The increase in research and development was primarily due to an increase in headcount and related expenses. First quarter G&A expenses were $13.7 million versus $12.5 million in 2016. The increase was primarily due to higher expenses related to our pre-commercial activities for LAI. As of March 31, 2017, we had cash and cash equivalents of approximately $126 million and our cash-based operating expenses for the first quarter were $31.2 million. Note that we have now added a reconciliation in our press release of our GAAP operating expenses to our cash-based operating expenses.

With regard to our cash guidance, we continue to expect cash-based operating expenses to land within the range of $67 million to $77 million for the first half of 2017. This range primarily reflects spending for the CONVERT study, the follow-on 312 study for those NTM patients that do not convert, the continued regulatory and pre-commercial development of LAI. The estimates also include expenses related to INS1007, including clinical inventory purchases as well as preclinical and clinical start-up activities. As we've mentioned in the past, we will remain disciplined in our use of capital, and we will ensure our 2-priority programs, LAI and INS1007, are appropriately resourced.

With that, I'll turn the call back to Will.

Thank you, John. The Insmed team is committed to developing what we hope will be the first approved drug for a globally prevalent rare disease with no approved treatment options currently available. We believe that for patients struggling with NTM-1 disease, LAI had the potential to provide welcome relief to a potentially debilitating and deadly condition. We're also excited about our emerging pipeline of therapies to address unmet needs in other rare diseases. 2017 is shaping up to be a transformational year for Insmed, and we are looking forward to updating you on our progress. We're planning for an R&D day this summer, so please stay tuned for details on that event.

With that, I will hand the call back to the operator to begin the Q&A. Operator?

(Operator Instructions) Your first question comes from Josh Schimmer with Piper Jaffray.

As we're getting closer to the CONVERT data, there's a lot more interest from investors. Well, I was hoping that to start, maybe can you just review the commercial opportunities for ARIKAYCE, the patient demographics for the particular subset that you're exploring? I have 1 quick follow-up after that.

Sure, Josh, and thanks for the question. For that, I'm going to turn it over to Roger to answer, our Chief Commercial Officer.

Yes, thanks, Will. It's a great question. It's one of the areas that we've been spending quite a lot of time over the past few months trying to really understand what the opportunity looks like. We've been getting a lot more granular in looking at patient opportunity and mapping those patients across the globe. We're encouraged by what we're seeing. I think we'd like to give you more commentary on that at the session that we have that Will alluded to coming up in the summer. But we definitely are encouraged by what we're finding today.

Is there any way to get a little bit more granular than encouraged in terms of number of thousands of patients that you've been able to identify or the literature would suggest?

I think that -- I think what we previously guided is we feel very comfortable with that. And I think that before we commit to anything more than that, we'd like to just complete that analysis and give you a further update in the summer time.

Yes, and Josh, just for the benefit of those on the phone that haven't heard that previous guidance, again, the NIH came out with an estimate that the prevalence in the U.S. is about 181,000 patients, it's just U.S. numbers. We had previously discussed a diagnosed prevalence of around 55,000 patients and of that when we target, as this study does, severe refractory MAC induced NTM patients, who've been on background therapy for a minimum of 6 months, that profile and if we were to get that kind of a label, we estimate that the addressable market there would be about 10,000 patients in the U.S. So obviously, the exciting part of this is the opportunity to go beyond that at some point in the future. And that is certainly, as we dissect the market, we're looking both at our current opportunity informed by this trial and then the potential for the future.

Got it. And then, while you had touched on demonstration and the economic value of the product, do you expect that any of the clinical trial endpoints in CONVERT will provide direct evidence of economic value? Or is it going to have to be extrapolated from evidence that you've assembled for patients who culture convert? And then maybe you can kind of elaborate on the economic value that you've been able to show for those patients. And is that value extrapolated then to patients who convert and then relapse and then maybe can go back on their case?

Yes. So obviously, we're going to be doing a lot of work around the value for money story, which has been our mantra from day 1 here long before it was the topic of the day in the media. And I think what we have found from the work that we did in the distant past is that culture conversion results in a drop in hospitalization. And that correlation, while not extracted from our clinical data, nonetheless suggests that given that is our primary endpoint, we're focused on the right outcome measure for the benefit of insurance companies to understand that they're getting value for money by achieving that endpoint. We'll look at this extensively in our Phase III study obviously, but I think finding that information that these patients are actually surprisingly expensive that they're spending much, much more time in the hospital than age-matched controls and that that data comes from an insurance company database, it's not something we went out and researched. This is to clarify for everybody on the phone what we paid an insurance company to go in and look at the data and tell us what it cost them to treat an NTM patient and then how that has changed if the patient culture converts. And if you look at our investor relations deck which is available online, you will see that data spelled out in summary form in one of our slides and the reference to the poster that captures this in the publications behind it. So I would encourage everyone to turn their attention to that because what it frames out is a very compelling value for money story. I don't know, Roger, if you want to add anything to that in terms of where we're putting our time and efforts right now.

No. I think that captures it very well, Will. I think that one of the interesting things around this effort is partnering with the insurance companies and actually taking a look at their data is actually very valuable and educational for payers as well. So I think that raising -- the opportunities to raise the awareness of the expense of this diseases and the fact that we didn't come with a solution at the appropriate time, I think will be very well received.

Your next question comes from Adam Walsh of Stifel.

Three here. First one is, Will, can you discuss how we should think about the potential performance of the standard of care arm in the Phase III CONVERT study? And then number two, how we should think about the secondary endpoint of 6-minute walk tests? Is that something that the FDA has said to you is of critical importance to get the drug approved under Subpart H? And then number three, can you talk about a little bit about the timing of the filings in other geographies beyond the U.S.?

Sure. Adam, thanks for the 3 questions, which I hope I got written down accurately. The standard of care arm in Phase III, I think you put your finger on a very important element of the study. As you know from Phase II, we have an understanding that standard of care for severe refractory MAC NTM patients converted about 5% of patients in the time frame that we examined, which was roughly 3 months. In this study, we expect the same performance. But it's an interesting question because nobody has ever studied the standard of care that's used. Indeed, the guideline therapy that was generated for use by specialists in this field is not based on a clinical study, it's based on the best opinions of key opinion leaders. Now I would support that those folks are pretty good in their judgments. But there isn't a lot of clinical data behind it. So this study not only examines ARIKAYCE in a very thorough way, but it also examines the standard of care and its value in a very thorough way for the first time. So to put a finer point on it, we are 90% powered to show a 15% difference between our drug plus standard of care versus standard of care alone. And we've communicated to the market that we believe standard of care should do about a 5% conversion rate and that our drug consequently should do about a 20% conversion rate after 6 months of therapy using that triple negative culture as the standard by which we make that judgment. But I think understanding standard of care here will be very important. And the read through would be it's more attractive to us, the more effective our drug can be relative to standard of care. So to be transparent, if standard of care is very successful, our drug, because we're randomized 2:1, should also be equally successful plus whatever we can bring to the equation. But if standard of care is very poor, it's really going to raise the question of the utility of that regimen of drugs because it's very difficult for patients to tolerate, and you are exposing them to antibiotics that aren't effective and effectively promoting resistance over time. So think of these patients, they come in 6 months on standard of care, then they get a minimum of another 6 months. If less than 5% of them are converting after 1 year of exposure to standard of care, folks are going to start to wonder what the value is of that regimen. So I think there is real learning to be had from that arm in the study. Second question was a 6-minute walk test. And that endpoint, look, there isn't a lot of study that has been done in NTM. We added that as one of the endpoints. I think given that this is an open-label study, although we treat it as a blinded study for purposes of our understanding of what's going on, that really is a very difficult measure for the FDA to feel like they can lean on. I think it was interesting what we saw in Phase II that patients who culture converted seemed to do better on 6-minute walk. But a lot of that is driven off of the baseline health of the patients, how well they do in this study. So I don't think that that's going to be a controlling endpoint, but it will be interesting. I would just add that in our payer ad boards, they're not particularly focused on 6-minute walk test as being an important measure for their assessment, they're really focused on culture conversion. So I hope that's responsive to that question. And then the third quarter was timing of filings in other parts of the world. We're in a unique situation at the company because we went very early to PMDA in Japan and had them tell us that the design of this study would be adequate for registration and approval in Japan and we added a PK analysis of a subgroup of Japanese patients. So earlier than the vast majority of companies at our stage of development, we are in a position to file in Japan at the conclusion of the study and our expectation is that we will do that. I think it's fair to say that, while Roger and his team are still early in the work, we've got a pretty good understanding of the scope of the opportunity in Japan and it's quite compelling. So I think that that is a market that I would promote to people's attention as being far more interesting than we first thought it was going to be and that's a byproduct of both the work that Roger and his team are doing as well as our experience in the clinical trial. Europe, as you know, we already have a strong operation up and running there. So I think collectively, we feel we're going to have access to global regulatory submissions on the basis of our 212 and 312 data. Hope that answers those questions.

Your next question comes from Ritu Baral of Cowen.

Well, my first question -- I have a CMC follow-up after this. But my first question has to do with the percentage of patients that will have completed the 6 months additional treatment period after first culture conversion. What percent of -- approximately, of course, what percentage do you think will have completed that period at around September-ish data release? And as that is likely to be, I think, perspective data, what could that follow-up data have for -- what sort of implications could it have for labels and claims and other data generated?

Okay. So I'll take this question, and then I'll come back to you for your CMC question. We don't -- I don't have a good estimate for you of how many patients will have completed additional exposure. I think just to frame it out for everyone on the phone, the way this study is designed, patients get 6 months of either standard of care or standard of care plus our drug. And then from the time of their culture conversion, the first of the 3 in succession is where we begin the additional 12 months of treatment that continues and that's per guideline therapy. So patient comes in, they culture convert in Month 12 or Month 4 for example, they would get an additional 12 months of therapy after that. So that the totality of their treatment period under guideline therapy would be 16 months from start to finish. We then track them for 1 year off all drugs to see a number of measures, including durability of the culture conversion if that's achieved. So that's the framework. And I think, Ritu, the question you're trying to get at is how much of that information will be available in around the September time frame that looks into that second part of the study where patients continue on drug. And indeed as of today, we obviously have some patients who've completed the entirety of the study and that information will be available for our filing. I think at this point, I wouldn't want to set expectations and say, our ambition is going to be to quickly turn out the top line result. I don't know that we're going to be in a position to add anything to that because we're, obviously, going to want to look at that very carefully before we put it out publicly. And I'm not sure that the top line results will be the right forum for that. But shortly thereafter, we would hope to have a peer reviewed setting where we could put out more information about both the initial treatment phase and then the subsequent phase of treatment whether it's in a continuation of 212 or the 312 study.

And what -- I mean, what sort of key either analyses from that follow-up period could be important for either the label or payers do you think even if we don't have it at [September]?

Yes. No, that's a great follow-up. So I think what's really going to be important is the first -- the answer to the first question is going to be, patients are exposed for 6 months and a certain percentage are converted. What of those who don't convert in that first 6 months, but continue on drug, do they continue to convert? Do we see by Month 9 or indeed by Month 12, that even more have converted? Because that feeds into the logic of continuing treatment beyond the initial 6-month treatment phase that we've identified in this study. And that could be quite important for duration of therapy to achieve that first culture conversion. So I'll be very interested to see that. I think then the other thing that will be very important is the durability of the conversion. We are going to be watching patients for a full year off of all drug. And as we saw in Phase II, we managed to get these patients who had been very resistant to any culture conversion. Some of these patients had been on drug for years and never culture converted. We culture converted them. And then 1 year off drug, they remain culture negative. So that's quite a surprising and positive finding. And we hope to see that, again, in this study. Will those patients show up culture positive after they're converted? It will be very important for us to examine the underlying infection. We're going to be DNA typing these samples to ensure that we know whether the patient has relapsed. In other words, we didn't completely eradicate the infection or it's a reinfection from a new species, and therefore, would be subject and available to retreatment by LAI. In the Phase II study and the experience to date, we expect about 1/3 of the patients will be reinfected with a new strain of NTM that we would then have to retreat.

Got it. And then my CMC question, can -- you mentioned that you have 2 sources of API, I believe. Can you take us through what your facilities are for fill and finish? And whether all of these facilities are GMP certified and potentially when they've last gone through inspections?

Yes. And so what I'll say to that right now is, we're going to be taking a deeper dive on that at our Analyst Day, because we want -- there's obviously a lot of detail around CMC issues. As we all know, 40% of complete response letters last year were CMC-based. We've spent the last several years under the direction of Don Nociolo here at Insmed really getting our facilities to a top standard. We now have 2 sources of supply that's a creation of drug product. And we have also qualified all the way back through our supply chain, and not to beat the point to death, but we know the farm field in China where the soybeans are grown that give rise to the kanamycin that gives rise to the amikacin sulfate, which is the raw material used to create the amikacin that we use. We have second sources of supply or inventory built up all along that supply chain. Similarly, the DPPC and cholesterol sources that are used to create the liposome have either been inventoried backed up or a second source of supply created. So I think we have really taken to heart the importance of CMC and invested in it heavily over the last several years with the result that we now have 1 source of supply at 50 liters scale, and then second source of supply fully automated at 200 liters, both of which are creating product as we sit here today for our clinical trial. So I would say never rest, but we feel very good about our CMC situation. I hope that's responsive to your question and for the details on fill and finish and the other aspects that are important of supply chain, I would draw your attention to our July meeting where we're going to have tech ops share some of that information.

Your next question comes from Joseph Schwartz of Leerink Partners.

So within MAC NTM, there seems to be some high geographic subspecies variability as you look at the strains that are distributed around the world. And so given that the last study was performed only in North America and CONVERT is distributed across 18 countries, as I recall. What do we know about how well ARIKAYCE can work across different isolates based on the MICs in vitro? Or is there clinical data comparing the natural history for patients who are treated comparably, but have different subspecies strains?

Well, so it's a good question. I think you're getting into the weeds, as we should. I think the first thing I would observe is, it's one of the benefits of being in an orphan condition, the key opinion leader network is global. So all of the dialogue that surrounds the analysis of the different subspecies that give rise to NTM is one that happens at every one of the conferences, whether it's ATS or ERS or what's been kicked off now is an important conference over in Japan to look at these diseases and what gives rise to them. We are not concerned by the variability that is suggested by the geographic growth or outreach from this study. In fact, we spent a lot of time looking at exactly the question you're asking before we ever went international. And I think we feel confident and I'm not going to go on to the -- more of the details of that other than to say that in looking at the subspecies that give rise to NTM, remember that we are focused exclusively on MAC in the study and that the responsiveness of that particular avium complex to different treatment regiments has been examined. And so that has informed the design of our Phase III study. So we feel very good about the performance of our drug in all of these different settings. And I would go further to say that, not just the underlying bug that gives rise to the infection, but the various ethnicities that are involved and the behaviors that sometime append to those kind of variabilities have also been considered in the design and the administration of the study.

Okay, great. I'm just trying to think of all the known and unknowns here. One other one could be, do we know if patients in CONVERT have similar degree of disease severity in terms of their sputum counts as the patients who are enrolled in TARGET? And then when you look at TARGET, do you see any baseline effect, such as ARIKAYCE converting fewer people with higher sputum counts of NTM bacteria than those with lower sputum counts?

Yes. So this goes back for the benefit of those on the phone to the primary endpoint of the Phase II study, which was the reduction in bacterial density that was examined at the time. It was felt that that endpoint would be predictive of future culture conversion because at the time and the shot period that we are administering our drug, nobody believed we'd actually accomplish any culture conversion. And the unexpected but positive outcome of achieving culture conversion in so many patients is what led the breakthrough therapy designation and our pursuit of that as the primary endpoint in Phase III. I think what we've learned from Phase II was that, that primary endpoint was ill-advised. It was not particularly correlated with any outcome. It had not been used in a clinical study before. And subsequently, we do not feel it's beneficial to even examine bacterial density. This was a 7-step scale that was sort of created for individual patient examination, but really not meant as a comparative tool across patients given the inherent variability of sputum. Moreover, the actual ranking of the sputum density does not correlate to any clinical benefit that anyone's established. So I think we have to set that endpoint aside as an informative tool and accept that it was probably too broad and too blunt to be helpful for us. And really just hold ourselves to the higher standard of culture conversion. The goal of treatment in these patients is eradication of the infection, that's what culture conversion represents, that's what we are able to demonstrate in Phase II, and that's what we're looking at in Phase III.

But do we know if ARIKAYCE is able to convert people to negative easier if the sputum counts are lower, as opposed to if they were higher?

We don't because the challenge of sputum density is when you take a sample to get into the finer point, you don't know whether that sample is actually representative of the density within the lung, right. You're getting a sample, there's inherent variability in sputum sampling, you will hear that over and over again. And so bacterial density as a measure gives a false sense of confidence that you've got a good read on the bacterial density inside the patient's lung. That's why we set it aside, it's not that there isn't a point of interest around your question, it's just that there's no reliable way to measure it.

Your next question comes from Liisa Bayko of JMP Securities.

Just curious if you can give any more color on discontinuation rights, I know you said they were in line with expectations. Maybe if you can't say specifically, maybe remind us of what you were expecting?

So yes, it's a good question, Liisa, and thank you for it. We're not talking more specifically about numbers here and I know that's disappointing. But I think that the important thing about this is that we remain within our expectations, and we'll look forward to providing greater detail in the future. I will just comment that one of the things that holds us back is you never know how the behavior of the last bolus of patients is going to be. And we don't want to get ourselves caught out by making a comment about where we are if it then changes because for some reason the last group of patients has a higher dropout rate for whatever reason. So that's the explanation for why we're not providing more color. Your specific question about what percentage do we expect? I think what we saw in Phase II was around 25%, 30% of patients dropped out over the duration of the study. And I think when we talk about meeting expectations, we want to be within reach of those kinds of numbers and we won't know where we are specifically until all patients are through the study.

And the discontinuation rate pretty consistent across arms?

And you mean with regard to Phase II? Or are you trying to ask me something about Phase III again?

Well, I guess I'll stick to Phase II because you probably don't want to comment on Phase III. But if you do want to comment on Phase III, please go ahead.

No, I think we're going to hold off on Phase III, again, for the reasons previously said. And in Phase II, you do see a higher discontinuation in AE rate as you would expect in the patients who were administered LAI. Consistent with other inhaled antibiotics, you find the experience of breathing in drug can be unsettling the people who are not used to it. And once they have used it for a period of time and pushed through the initial discomfort, the AEs drop-off to almost nothing and that's consistent with other inhaled antibiotics like tobramycin and CAYSTON and others. In our case, we saw that in Phase II, and so there is a higher discontinuation rate in the treatment arm. We would expect to see that here to be sure, but the question will be will it be within the reach of that previously mentioned number?

Okay. And then just curious about the 6-minute walk test. What's the correlation between culture conversion and change in 6-minute walk improvement? I guess, I would say the 6-minute walk. And what's the timing of that? Is it -- does it hold, like if you culture and convert, you then should kind of immediately see a benefit in 6-minute walk? Or what's the right way to think about the connection between the 2 in terms of both just actually improvement and also timing?

Well, your question actually raises the very issue that makes us uncomfortable, which is that we don't have enough data to really make definitive answers to those inquiries. We do know that we saw, post hoc, a good correlation between culture conversion and improvement in 6-minute walk test at the time that, that was measured, which is base line to 3 or 6 months out. And that, that appeared to relate to patients who had culture converted. So we will be more explicitly exploring that in this study. But again, we've got a lot of variables in operation here. That was a post hoc analysis, and in this situation, we don't know if we'll see that same behavior and evidence. And a lot of it will depend on the characteristics of the patient at baseline. Patients are very healthy relatively speaking in terms of their ability to walk at baseline. Will we see the same degree of improvement? If they're very sick, do we see a more dramatic degree of improvement? Those are the nuances that we'll be looking at to try and understand how and why 6-minute walk test may be a valuable measure.

There is time for one more question and it comes from the line of Josh Schimmer of Piper Jaffrey.

Just want to continue the line of questioning from Liisa and Joe. How do we think about making a case for patients who don't culture convert to stay on therapy? And if we can't make a strong case for those, how do we think about this market dynamic? Do kind of most patients lined up on therapy for 16 months and then maybe 1/3 of them kind of turn into a chronic relapsing remitting type control state? Or just kind of address those 2 points.

Yes. I think the Phase III study is going to give us a lot of the answers to that question, which is what happens to the patient after they've been on drug for 6 months versus 3 months versus 9 months. Once we know that and can extrapolate that line, I think we'll be in a better position to comment how long patients should be on drug. It's an interesting question because if you speak to KOLs now, as I know you have, what you will hear is that most of them say, when they entered the clinic with NTM, they don't ever leave. And so they give them this cocktail of drugs for a long period of time and they rotate different permutations to try and make some improvement and achieve that culture negativity in the durable way. At the PFDD meeting that the FDA held, where they examined what it meant to have NTM, there were patients who were standing up talking to them about how they've been drug for 10 or 20 years and had never been culture negative. So we don't know how long patients will be on drug. I think the clinician's instinct is going to be to continue to treat. That's currently the practice right now. I think what we need to tease out is how do we reconcile that with the payment community and ensure that we're getting the right outcomes that they need to see. I don't know, Roger, if you want to add anything to that?

No, I think that's exactly right. I think it's pretty clear cut with patients who achieved the culture conversion and then an additional 12 months of therapy according to the guidelines. Clinicians, I think, while they desire to have culture conversion as a hard endpoint that they'd like to achieve, keeping the infection in check is also a goal that they pursue and that's why as long as the patient will continue to tolerate the therapy, they're willing to continue to treat them in order to achieve that goal. As we engage with the Bayer community and talk about what success looks like for those patients and as we get more economic data and get more information from our Phase III trial, I think that picture becomes much more clear as to how physicians should be thinking about treating those patients who don't achieve that immediate culture conversion, but are seeing some sort of a benefit from continued treatment.

And that goes back to the whole CF model, right, the inhaled antibiotic that is not ever intended to eradicate the underlying infection, it's taken to keep it at bay. And so I think what we're trying to do is figure out how this model is going to operate, and obviously, data will help us inform that -- answer that question.

This concludes today's question period. I now return the call to Mr. Will Lewis.

Thank you again for joining us. We look forward to meeting and speaking with you in the months to come. Have a great day, everyone.

This concludes today's conference call. You may now disconnect.