Enanta Pharmaceuticals Inc (ENTA) 2018 Q4 法說會逐字稿

Good afternoon. My name is Deirdre, and I will be your conference operator today. At this time, I would like to welcome everyone to the Enanta Pharmaceuticals Fourth Quarter Financial Results Conference Call. (Operator Instructions).

I would now like to turn the call over to your host, Ms. Carol Miceli. Ma'am, you may begin your conference.

Thank you, Deirdre, and thanks for joining us this afternoon. The news release with our financial results was issued this afternoon and is available on our website.

On the call today is Dr. Jay Luly, President and Chief Executive Officer; Paul Mellett, our Chief Financial Officer; and other members of Enanta's senior management team.

Before we begin with our formal remarks, we want to remind you that we'll be making forward-looking statements, including plans and expectations with respect to our product candidates and financial projections, all of which involve certain assumptions, risks and uncertainties that are beyond our control and could cause our actual developments and results to differ materially from these statements.

A description of these risks and uncertainties is in our most recent Form 10-Q and other periodic reports filed with the SEC. In addition, Enanta does not undertake any obligation to update any forward-looking statements made during this call.

I'd now like to turn the call over to Dr. Jay Luly, President and CEO.

Thank you, Carol. Good afternoon, everyone, and thank you for joining us today. Our goal for the past several years has to become a leader in the field of viral infections and liver diseases for investing in and diversifying into other disease areas with significant unmet need and market opportunity, all without the necessity of additional dilutive equity financing.

I'm proud to say that we have accomplished many of our objectives. Specifically, in the past year, we have advanced all our wholly owned programs and currently have 3 Phase II clinical trials ongoing in RSV, NASH and PBC, and a clinical study of our first HBV candidate is expected to begin next year.

Enanta's scientific expertise has been validated by our partnered HCV protease inhibitor program, which has resulted in 2 approved products marketed by AbbVie, including MAVIRET, currently the leading HCV drug treatment in the world. Royalties from our AbbVie collaboration have provided a strong financial foundation to fund our wholly owned in-house efforts.

As we have significantly grown our clinical pipeline, we have kept our financial resources strong with ongoing royalties and a cash balance of $325 million, while continuing to discover new compounds and advancing our wholly owned programs in the clinic, which I will now discuss.

One of our most exciting near-term opportunities is EDP-938, our product candidate for respiratory syncytial virus or RSV. EDP-938 is potent non-fusion inhibitor targeting the N protein in RSV, and the only N inhibitor in clinical development today. EDP-938 works by inhibiting the replication process of the virus, and as a result has the potential to be more effective when used at later stages of RSV infection than fusion inhibitors.

Recently, we successfully completed the Phase I study of EDP-938 and are quite pleased with the safety and pharmacokinetic profile of this first-in-class compound. The Phase I results were presented at the 11th International Respiratory Syncytial Virus Symposium earlier this month and demonstrated that EDP-938 was generally safe and well tolerated over a broad range of single and multiple doses, with a good pharmacokinetic profile suitable for once or twice daily oral dosing.

An important observation to note is the main trough levels in the Phase I study were approximately 30x higher than the EC90 of EDP-938 against RSV infected human cells tested in vitro. In general, achieving drug concentrations had a significant multiple over the antiviral EC90, a measure of potency for the compound, is crucial to the success of progressing an effective antiviral drug.

In October, Enanta initiated a Phase IIa human challenge study in healthy adults inoculated with RSV. This randomized, double-blind, placebo-controlled, human challenge study will enroll up to 114 healthy adult subjects who will be randomized into 1 of 2 drug dosing arms or a placebo arm and dosed for 5 days. Primary and secondary outcome measures include changes in viral load measurements and changes in baseline symptoms. Given the relatively short dosing duration of this study, we expect to announce preliminary data in the third calendar quarter of 2019.

Moving to NASH and PBC. Enrollment continues for our 2 Phase II clinical trials in NASH and in PBC with EDP-305, our FXR agonist candidate. Both studies are 12-week, randomized, double-blind, placebo-controlled studies, evaluating the safety, tolerability, pharmacokinetics and efficacy of EDP-305 in subjects with NASH and PBC.

We anticipate sharing initial data starting with NASH in mid-2019. Recruitment in PBC has been challenging due to multiple clinical trials attempting to access a very limited number of available patients for this potential second-line therapy in an orphan indication. We will continue to provide further updates on PBC enrollment in the coming months.

In NASH, we're accumulating data and building on our extensive preclinical knowledge of our FXR agonist, EDP-305. Earlier this month, 2 preclinical posters from the lab of Dr. Bryan Fuchs at Massachusetts General Hospital were presented at the liver meeting. One poster demonstrated that in a rat model of cirrhosis and hepatocellular carcinoma or HCC, EDP-305 reduced serum markers of liver injury, fibrosis, ascites development, HCC development and mortality.

The second poster demonstrated reduced fibrosis progression in a rat model of NASH cirrhosis, suggesting EDP-305 may have potent anti-fibrotic effects in late stage liver disease. We have a long-term commitment to NASH. And we're identifying new follow-on FXR agonist leads and continuing our ongoing discovery work in non-FXR mechanisms, anticipating the likely need for combination therapy in this indication.

Lastly, I'm very excited to highlight our announcement of the selection of EDP-514 as our first development candidate from our HBV program. EDP-514 is a core inhibitor, also referred to as a capsid assembly modulator or a core protein allosteric modulator. These are a new class of HBV inhibitors that can disrupt the assembly and replication of this virus at multiple steps in the viral life cycle. Based on our preclinical data, we believe EDP-514 has best-in-class potential for the core inhibitor mechanism.

In 2019, we plan to present this compelling preclinical data and initiate a Phase I clinical study of EDP-514, which will include a Phase Ib portion in chronic HBV patients. In addition, we are exploring other anti-HBV approaches as we believe that it may be necessary to utilize multiple mechanisms for the treatment of HBV.

Enanta had a strong 2018 and I'd like to thank our dedicated employees who worked tirelessly to achieve our goals. I'd also like to thank the patients and their families that are participating in our clinical trials. And I look forward to providing more updates in 2019.

I'll stop here and turn the call over to Paul to discuss our financial results for the quarter. Paul?

Thank you, Jay. I'd like to remind everyone that Enanta reports on a fiscal year schedule. Our year-end is September 30 and today we are reporting results for our fourth fiscal quarter and year ended September 30, 2018.

For the 3 months ended September 30, 2018, total revenue was $67.2 million and consisted entirely of royalty revenue. This compares to total revenue of $75.9 million for the same period in 2017, which included $65 million in milestone payments. The increase in royalty revenue in the current quarter was due to an increase in royalties earned on AbbVie's $862 million in global sales of HCV regiments, including royalties on 50% of the $839 million of MAVIRET sales in the quarter.

Total royalty revenue for fiscal 2018 was approximately $192 million. I will remind you that AbbVie is eligible to earn annually tiered -- Enanta is eligible to earn annually tiered double-digit royalties on 50% of AbbVie's global net HCV sales of MAVIRET. This means that our average royalty rate for a given royalty year will restart at the lowest rate in our quarter ending March 31, and will be highest in our quarter ending December 31. In the quarter ending to September 30, our royalty rate reached the 17% tier, which is also the rate at which we would expect to earn our royalties in the quarter ending December 31.

Moving on to our expenses. For the 3 months ended September 30, 2018, research and development expenses totaled $26.9 million compared to $16.5 million for the same period in 2017. The increase was primarily due to greater preclinical and clinical costs associated with the progression of our wholly owned R&D programs in NASH, PBC, RSV and HBV. General and administrative expense for the quarter was $5.8 million versus $5.1 million for the comparable quarter in 2017.

Enanta reported income tax expense of $8.5 million for the 3 months ended September 30, 2018, compared to an income tax expense of $18.4 million for the same period in 2017. Enanta's effective tax rate for fiscal 2018 was approximately 23% compared to approximately 34% in fiscal 2017, primarily due to the U.S. Tax Cuts and Jobs Act enacted in December 2017. Enanta's effective tax rate for fiscal 2018 also reflects the impact of a noncash revaluation charge against deferred tax assets due to the reduced federal corporate income tax rate.

Net income for the 3 months ended September 30, 2018, was $27.4 million or $1.30 per diluted common share compared to a net income of $36.5 million or $1.86 per diluted common share for the corresponding period in 2017. Enanta ended the quarter with approximately $325 million in cash and marketable securities, an increase of approximately $31 million to our 2017 fiscal year-end balance of $294 million. We expect that these cash resources, as well as our continuing royalty revenue will be sufficient to meet our anticipated cash requirements for the foreseeable future.

In terms of guidance of fiscal 2019, we expect our research and development expense to be between $135 million and $155 million, our general and administrative expense to be between $27 million and $33 million and our effective tax rate to be approximately 21%. Further details are available in our press release and will be available in our Form 10-K for the quarter when filed. I'd now like to turn the call back to Jay.

Now on our sixth year since we became a public company, I'm proud that in that time we've had 2 outlicensed products approved as part of cures for HCV, a major disease worldwide. We have built our internal programs in RSV, NASH, PBC and HBV from scratch. We've discovered and brought 2 clinical -- to the clinic 2 new compounds from these programs. And we are developing more compounds that we plan to bring to the clinic starting in the new year. And all of that work is still being funded through non-dilutive financing.

I'll now turn the call back to the operator to open up the lines for Q&A. Operator?

(Operator Instructions) Our first question comes from Brian Abraham -- I'm sorry, Abrahams with RBC Capital Markets.

I guess, my first questions are on the RSV program and 938. I was wondering if you could -- as we look towards the results of the challenge study and then the development path beyond, can you talk a little bit about your views on the predictiveness of the challenge study for a real-world trial, given some of the differences -- the intrinsic differences there? And maybe what this unique class might offer in that regard? And then, as you think about next steps, I'm sort of wondering what groups of patients you might be considering investigating the drug in subsequently? And how you think about prioritizing that? And then I have a follow-up on hep B.

Sure. Thanks for the question, Brian. So I think when it comes to the translation of the challenge study, which is a Phase IIa study to Phase IIb. Clearly, the things that we're looking at are whether it's Phase IIa, IIb, Phase III, whether it's on the market, the things in this indication I think you really have to look at are the fundamentals of the virology and the targeting of the target organ, which in this case is the lung. So Phase IIa, obviously, we're looking in humans who have been infected with the RSV virus. We're setting their dosing, people with 1 of 2 doses, which we, I think, have picked pretty carefully. And what that really allows you to do is in a reasonable real-world setting, although it's not a perfectly real-world setting, look at what the drug does on viral load during the time course of the infection. We'll be looking at the viral kinetics extremely carefully from the time we start dosing folks.

And out of that we'll also actually be measuring signs and symptoms. So this information, which is a fairly standard Phase IIa, I think, actually is some of the best information you can do in a reasonably controlled environment before going out and then looking at one of the other patient populations, which is another part of your question, and I will get to. So in terms of that translation, there aren't many examples of it, quite honestly. I think the main ones that I think people can point to in terms of fusion or of basically the challenge studies that have moved on to Phase IIb populations are Gilead's fusion inhibitor, which had some issues, I think, at that next age. It's our belief that a lot of that might have to do with the mechanism rather than the translation of going from a Phase IIa to a IIb. The other opportunity that we did have, but I'm not sure that we do have anymore, to see that translation occur is with the J&J, Alios nuc that was stopped in Phase IIb trials. Phase IIb trials were stopped on that -- within the last several weeks. I believe that was a preclinical finding that led to the stop there.

So those would have been a -- that would've been a very interesting study to see the data result from because that was a non-fusion approach, slightly different than the one we're taking. But importantly, not a fusion inhibitor. And -- so there's a -- the world's really lacking in this type of data. But I think fundamentally, if you think about the science and the scientific underpinnings in this study having good pressure on the virus, having good activity against clinical isolates, and we've got a lot of good activity against many different clinical isolates from a variety of territories for both type A and type B. I would say there's a very good chance of translation. I would say it's probably much higher than in many Phase II settings where you make that sort of a change.

So the other thing I think you get out of this is safety profiles. As you know, we looked in Phase I for a 7-day treatment duration, and we had a very clean safety profile. So we don't expect anything different in that in Phase II and beyond. But we'll certainly get a look at that at 5 days of treatment in that study. So that's another good thing. So we'll be, again, targeting the right organism at the right tissue and the right species in this study. And I think that goes a long way.

In terms of which patient populations we would think about targeting -- I'll remind you, I think you already know. But the top 3 are looking at children; the young are vulnerable to this infection, almost every child before age 2 is infected. And then the elderly, it's very common in elderly people who become hospitalized or just otherwise, elderly people who have greater susceptibility to respiratory tract infections. And then the third class would be in immune compromised settings where these patients are incredibly vulnerable to RSV infections, and the consequences can actually be quite severe.

So just in terms of prioritizing those, I expect, over time, we will be in all 3 of those studies or in all 3 of those patient populations in -- with different studies. But I think the prudent path that we're likely to be taking is starting with an adult population. And we'll have more to discuss about that in early 2019.

That's really helpful. And then just a follow-up on HBV. If you could maybe expand a little bit more on the differences and similarities between 514 and 367 with respect to things like genotype coverage, potency, selectivity, metabolism, and then how you think about -- or how the recent FDA draft guidance on hep B development might influence your development path and which populations, naïve, monotherapy combinations, you might think about exploring with that down the line?

I think that was 13 questions. So with EDP-514, what we have stated earlier this year, when we put out the prototype 367, that we were going to give ourselves a few more quarters, work through substantial part of the rest of the year trying to see if we had something in-house that was even better than 367, which was a very well-rounded molecule. And without getting into all those items point by point today, we do plan to present this data at a scientific conference in 2019. We'll have opportunities to put out a lot of detail around it. But what I can say today is that we believe it's an extremely good capsid inhibitor. It is better than 367. So if you use that as a benchmark virologically and some other dimensions that we haven't put all the data out in, and some of the DMPK parameters and just -- from pretty much every vantage point, we continue to drill down and optimize. So 514 is a very strong candidate. But in particular, we are pleased with the DMPK profile and the virologic profile are very strong.

Related to the clinical trial design, again, as we get closer to this, and I would expect earlier in the new year we will lay out more details about the timing, when we'll be in studies and what those studies will look like. Although, as I mentioned a bit earlier, it will even in Phase I in the Ib portion, we do plan to get into HBV patients. So stay tuned on that.

And our next question comes from Yasmeen Rahimi with Roth Capital Partners.

Few question on RSV and then one question on NASH. So Jay, can you kind of walk us through of the dose selection in the RSV challenge study, given the phenomenal PK Phase 1 data? It might be you had quite a lot of freedom to push the doses even higher. And then secondly, RSV challenge study seemed quite simplistic in theory, however it is quite finessing in getting it correct. So are there elements of the RSV challenge protocol that you and the team have sort of optimized and that have been contributing to the failures in other trials? And then I have 2 more follow-ups.

Sure. Well, regarding to the failure of others. I mean some people have failed at this stage. There are examples of fusion inhibitors that weren't very potent and they didn't work in the challenge study. And then there were examples where people moved -- were successful in challenge and then moved on. We already discussed that I guess on the last question, there's just not a lot of data points out there. So fundamentally, what we wanted to do is pick strong doses. Again, I think the challenge study lends itself to so much exploration it'd be pretty easy to get in there and get bogged down in a million different variations on exploring dosing dynamics as opposed to coming up with a solid set that we believe we've powered very strongly for the 2 endpoints that we have selected. And so the doses that we chose were at the higher range. So we explored in Phase I a full range of doses. And we went in with high doses from the Phase I really so that we could press the multiple of an EC90 really hard.

I mean, other people who have come in and do -- done these challenge studies might have a three or a fivefold multiple. What we did was we picked doses that we think based on exposures of looking at trough levels, either at 12 hours or at 24 hours, where we are going to have around 30x the EC90. The other -- so that's we think is tremendous amount of pressure. The more pressure you can put on the virus, more pressure there is to kill it. Kill it quick to avoid resistance and so forth. We think time is actually very important in RSV. So the sooner you can treat, probably the better. And the sooner you can get to steady state, probably the better. And the sooner you can get a significant multiple of the EC90 hammering that virus, probably the better.

So we went in and we looked at human primary cells. These are human bronchial epithelial cells. We looked at the Memphis Strain of the virus, the M37 strain, that's the one that we're actually using in the challenge study. And we did very careful virology to understand that. And then looked at all the PK curves, et cetera, et cetera, from our Phase I study and came up with those 2 doses. So the first dose, as you probably know, is 600 milligrams once a day. The other comes in with a loading dose of 500 and then follows it with 300 BID. And even without the loading dose, just looking at the BID dosing, we know we get very substantial trough levels and high multiples against the EC90. So we're using a very, very relevant human cell type and great PK. So I think it'll be an extremely exciting data set to get. We expect, again, the timing on that is in calendar Q3.

Great. And then 2 quick questions. Are there any regulatory obstacles that are unique in RSV development in children versus taking it into the elderly or the immunocompromised?

I'm sorry, are there any regulatory what?

Obstacles. Is there any elements of development that's unique that you have to show for before you can take into children versus taking it easily or into the elderly versus immunocompromised?

Nothing in particular. I mean, it's a standard regulatory development path for peds in terms of dose selection, et cetera. So nothing special. I mean, it's -- peds are peds. I think just from our perspective, what we're wanting to do is just get more adult experience before heading into the peds. I just think it's a prudent course. I mean, you do have to think about other considerations once you're -- in infants you have to play around with you doses to get them right. Children are a little bit of moving targets during their development time frame. So from that perspective, I don't want to suggest that dosing infants is the same path it is for healthy -- or otherwise healthy adults. The path is a little bit different. But we're going to get -- we've got good experience in Phase I, we're getting even more in our well powered challenge study, then we'll get Phase IIb going on in an adult population or populations and then we'll gradually phase in the children as well.

And our next question comes from the line of Liisa Bayko with JMP Securities.

I wanted to just drill down a little bit more on RSV. What have you done in your preclinical setting to sort of test the sort of initiation of therapy? Because I think that's one big thing, how quickly do you find out you have RSV in reality? And how soon do you need to start to treat to see an effect? And maybe you can speak to what you have observed in any animal model testing in that regard sort of when -- starting up therapy and how that impacts the outcome, if at all?

Sure. So as I mentioned earlier, the -- with any viral infection, the sooner you get to treatment, more likely the better the outcome is going to be. So it's going to be a spectrum. And what do we know about the virus? Generally speaking, in a typical viral infection, you have the infection, the virus will build up over the course of a few days. It'll sort of peak and then have a sort of a slow decline, will come down probably over the course of around 10 days or so. Now, near the late stage of that course is when other things -- if you haven't treated the infection or if the infection's particularly severe, you're going to get an inflammatory component towards the back end of that time period. So the real goal here is to get the drug on board as soon as possible and sort of deal with it when it's primarily a viral infection rather than a viral infection and an inflammatory sort of a syndrome cascade of sorts.

And so that's why -- so to answer your specific question, we don't have perfect animal modeling for this. We did a limited amount of animal modeling to sort of duplicate what some others had done in a non-human primate model, this the African green monkey model. And the dosing in that, I would say, while efficacy in that model has been predictive of success in human challenge studies -- human challenge study that we're doing right now sort of picks that up and carries it the next step. So again, we infect people. We let the viral titers climb very high, and then we come in and intervene after a few days. And after, I think, no more than 5 days. So it's in that sort of a time frame that you have a window.

Now when patients present, they've likely been infected for a few days because the viral load builds up and then you become symptomatic and then you take action upon being symptomatic. And then it's -- the day you become symptomatic, the day after, the day after that. When you present to a physician, that's going to be a slightly different situation at each one of those days because -- again, the further you get from the initial infection on that continuum, the harder it is likely to be able to treat.

Now the reason we focused on non-fusion inhibitors is because we believe that the maximum efficacy with a fusion inhibitor is going to be expressed very, very, very early in the course of the infection, as soon as you possibly can get it in there. Because once the cells become infected, you're no longer helping that cell. In fact, you'll turn that cell into a replication factory, and you may stop the infection of other cells. But at some point, you get such a viral burden going on that it's later and later it's too late for that. So what we want to be able to do with our non-fusion approach, which is the N protein inhibitors, is to go in and sort of address, no matter where you are on that cascade, you're going to have infected cells and we can get in there with our drug and shut down that virus pretty quickly.

So what we're hoping to have is an efficacy at a wider berth of days and also starting at a later day than you might with some other mechanisms that target purely viral entry. So I don't know if that's helpful. Again, I think there is only so much modeling you can do in a non-human species. I think we did the right experiments to build confidence that the next step should be as de-risked as we could reasonably make it and then just go in and do the human experiment.

Okay. That's helpful. I guess, is -- do you have any market research that tells you kind of when the typical range is when people are diagnosed in terms of duration? Is there an understanding of that? Or I guess there's not, you don't really know when you would have -- not...

Yes. You don't really know when you've been infected. It's one of those things.

Okay. All right. I guess as a -- viral titers, is that one way to look at it?

Yes. But once they come up, they kind of -- they're up there for a while. I mean, if you do -- obviously, if you have a substantial viral load, you've been infected for a few days, anyway. And again, we're trying to catch up. We know we're not going to get there before -- treat people before they're symptomatic. But on the other hand, as soon as they become symptomatic, I think there are several days with a non-fusion inhibitor anyway to work with before you get sort of laid into that inflammatory piece. So that's the window.

Okay. And then, just for the hepatitis B core inhibitor, can you maybe compare and contrast your core inhibitor to some of the others that are out there? I think of obviously [Assembly], I think J&J, also Arbutus, maybe you can compare and contrast in terms of the important metrics like potency and anything else that's notable. And also potency across genotypes.

Yes. I -- --you're welcome. We'll have a lot of -- a tremendous amount of data. I mean, you saw how much data we built up on 367. And your note this morning, I think, summarized that very well. What you can expect is equally robust type of data set, but with a molecule that looks even better. And we did quite a few cross comparisons, probably as many or more than a lot of other people have reported, to make it easy to -- for people to see that, even with 367. So it's a very strong profile. I mean, I hope that we can someday show that it's a best-in-class, or if not, it's one of the very, very, very strong contenders and becomes -- it's a base that we basically want to cover. And we want to have a good core inhibitor onto which we can add other therapies such as nucs and then yet other mechanisms as required to get yourself to a functional cure. So we think 514 has the profile to definitely sort of be a ticket to the show, so to speak. And we'll have a lot of data next year, we're just not ready to put it out today. And so we're -- but we will have a big data load next in 2019. The usual places.

(Operator Instructions) Your next question comes from Eric Joseph with JPMorgan.

Just a follow up on RSV and the infant and pediatric opportunity. I'm just wondering how much additional dosing work would be needed ahead of a potential trial there? And when you might look to do that? Is this something that could be done in parallel to the healthy challenge study or would it take place after or after an active infection study in the adult population?

Yes. So we'll -- as I mentioned earlier, we'll go from the challenge study, we'll go into adults after that and then with time, we'll bring on the children and regarding that, we'll follow the recent FDA guidance for now. So I think there will be other studies that we'll use to look at pharmacokinetics, et cetera, but it's -- I think it's actually reasonably straightforward.

Okay. And I'm just looking for a little more color from your opening remarks around PBC enrollment in INTREPID, you expect to provide an update in the coming months. But I'm just wondering if there's -- should be reading that as any flexibility to your anticipated data timelines and whether you're seeing any -- I guess to what extent you're seeing any impact from commercial kind of a -- Ocaliva, whether that's been a headwind to enrollment?

Yes. It's hard to know all the different pieces. I mean, PBC is definitely slower than NASH. I think given the recruiting dynamics that we see, it's difficult for us to give sort of a specific timing at the moment. But what I would say is based on sort of current trends, unlike NASH where we're targeting for data starting in mid-'19, I think it's probably less likely that we're going to have data in PBC in '19. One another thing, we're watching the PBC market opportunity very closely as we're doing this. And I think it's clear that PBC today is, for the market, is aimed at a second line therapy in an orphan indication for which there is a recently approved drug, Ocaliva, and for which generic fibrate competition may be a viable competitor sooner rather than later. So PBC is clearly a very, very small market compared to NASH, which -- and NASH is our principle focus. So we're just -- we'll keep an eye on the whole thing. But it has -- it's definitely been a challenge.

Your next question comes from Jay Olson with Oppenheimer.

Maybe if I could just ask a follow-up question about EDP-305 in NASH. Can you just talk about whether or not you think pruritus and LDL-cholesterol increases are potentially on-target side effects for an FXR agonist? And how likely you think it would be to sort of design an FXR agonist that doesn't have those side effects?

Yes. Really interesting question, Jay. So as you know, we've looked at this a lot. We've thought about it a tremendous amount. We've interrogated this in Phase I in some detail. As we dosed up, so obviously -- let me back up. So obviously, Ocaliva has seen pruritus. They have seen LDL increases. We went in with that knowledge going into our Phase I as did everyone else. We found that at the very high doses where we had sort of maxed out exposures, in a nonlinear way, we did see pruritus. But we didn't see increases in LDL. And we had done a lot of mechanistic work trying to understand why those LDL increases perhaps weren't seen. And we -- as you know, we looked at LDL receptor levels and the regulation of LDL receptor and found some differential effects there that potentially explain what we saw with EDP-305, which was no LDL increase versus what Intercept saw with their increases.

So I thought I had some understanding of this going into the liver meeting. I think neither Novartis nor Gilead had described pruritus changes in their Phase I work, but they did see it to varying degrees in their Phase II studies that they showed just a couple of weeks ago. So I'm not sure -- I don't think anybody fully understands the pruritus question yet, exactly what is causing that. So we're thinking about that and other properties as we work on our follow-on FXRs. We continue to dig around in that area and we've constructed a few theories and we're doing some further tests in-house. But right now, I came away from the liver meeting having looked at some of the other people's datasets as being a little bit confused.

I think some of the -- it appears that Novartis and maybe Gilead are changing their doses around now. So Novartis seems to be dosing higher in the future. And at least in the combination studies, I believe, with Gilead, they seem to be dosing down from their dose that they saw some of their data at, at the liver meeting. So time will tell. Still relatively early days, but we're getting more and more data sets to look at and to scratch our heads on. But again, so far, so good with 305. And again, we're pointing toward mid-next year for our data set to start reading out.

Okay, great. That's very helpful. Any thoughts on a target product profile for EDP-305?

Well, again, we think 3 -- so one thing we did come away from the liver meeting is that FXR and some of the data on FGF19, which is part of the path that FXR exerts, are still probably 2 of the strongest mechanisms when it comes to looking at fibrosis in a NASH patient. So from that vantage point, knowing what we need to try to accomplish in NASH therapy, I view that an FXR mechanism could still be a very critical one in the overall makeup of a combination. So what I want to come up with is a very strong FXR. We're very committed to this, even with our ongoing research, and thinking about how to think about generations of these things so that we, at the end of the day, come up with a molecule that can demonstrate some of the solid anti-fibrotic activity that has been seen with Ocaliva, but with a bit of a better tolerated package, and one that's readily combinable with other agents. So high level, that's the profile, right?

(Operator Instructions) And we have no further questions at this time.

Thank you, everyone, for joining us today. If you have any additional questions, feel free to give us a call in the office.

This does conclude today's conference call. Thank you for your participation. You may now disconnect.