Good afternoon and welcome to TD Cowen's Neuroscience Summit. I'm Phil Nadeau, one of the biotech analysts here at TD Cowen, and it's my pleasure to moderate a fireside chat with Neurocrine Biosciences. We have with us today Kyle Gano, the Chief Executive Officer, Sanjay Keswani, the Chief Medical Officer, and Todd Tushla, the VP of Investor Relations. Guys, first, I thought we'd start with maybe a brief overview of Neurocrine's strategy for developing and commercializing drugs for neuropsychiatric conditions. In particular, maybe which indications is Neurocrine most interested in? What types of targets will Neurocrine prioritize?
Thanks, Phil. I'll take that question. Thanks for having us here this morning. Always a pleasure to get together. Your question is actually timely. It's timely for a couple of reasons. One is, as you probably heard from us over the course of the past year, our intentions and being deliberate in really changing our view on R&D strategy. That's point number one. Number two, we have an R&D day coming up in December on the 16th, and you'll hear more about how that evolution is playing out one year in. Coming back specifically to your question, the aspects that are new to our strategy are several fold. One of them is, if you look at the pipeline today, it's largely been in-license through business development. That's something that will still be a big part of Neurocrine moving forward.
Part of the intentionality aspect of it is to bring in the capabilities so we can actually discover programs internally more so than we have over the course of the past decade. That's playing out now. We've had some success there this year, and that will be a bigger part of our story as we move on. The other one that goes with that internal discovery progress is moving away from looking at small molecules exclusively to other modalities, proteins, monoclonal antibodies, and peptides. We just put our first peptide antagonist into the clinic earlier this year as a follow-on to CRENESSITY. As we move on in the strategy, it's also having a focus on validated targets. You touched on targets there. That means for us clinically or genetically validated.
If you put these pieces together, it does afford us to move away from, or at least add to, what we could bring to patients, not only treatments for symptoms, but also disease modification. We're quite excited about these new areas that we're exploring. Ultimately, we can start talking about matching the right modality to the right target. These are two things that, if done right and appropriately, increase R&D productivity, which we've seen as well. At steady state at Neurocrine we're hoping to be able to share and show the investment community that we can put or start at least four new phase I programs each year, two new phase II programs, and have, again, at steady state, at least three programs in phase III development. At the end of the day, this could afford, we believe, one new medicine on average every other year.
A lot of good things are going on here at Neurocrine . We're in the early innings, but making good progress there. I do want to address your question in terms of the areas that we're interested in. You know, we do focus in neurology, psychiatry, and endocrinology. These have been legacy therapeutic areas for us. I also would add to that immunology. If you go back to my time here at Neurocrine dating back 25 years, we worked in all these areas. One of the common threads to them all are targets within the CNS and brain. It's amazing the different diseases that you can work on given that focus.
In terms of diseases, it's probably easier to talk about the things that are difficult for us, things like stroke or traumatic brain injury, where you're working with an endpoint that's not well validated, and you don't know if you've got a medicine until you've completed your first phase III study. Also, longer to develop neurodegenerative diseases are also quite challenging. If you look at what we're doing now, it's more diseases that give us confidence early in development, like phase I biomarker opportunities or in smaller signal-seeking phase II trials. These are the diseases that are of interest to us across the four therapeutic areas I mentioned. Really exciting time here at Neurocrine . Really am pleased with the progress that we've made. I think we'll have a lot more to share about this progress, not only in December, but in the coming years.
On your latter point, CNS disorders are inherently higher risk and perhaps higher reward. How do you balance risk versus reward across the portfolio? What's an optimal balance in Neurocrine ' opinion?
A couple of years ago, if you would have asked us this question, I would say that, you know, really trying to weight the portfolio to being more in neurology, maybe endocrinology, and immunology, where there are biomarker possibilities. Science has a funny way of playing out. What happened is our higher probability of success programs, at least that we thought, didn't work out as planned, and the psychiatry programs did. Today we have a more weighted toward psychiatry portfolio, which is something that we're going to play out and lead into because we think we've got some good assets and good data there. I think moving forward, what you expect or would expect is probably reweighting of the portfolio over time, maybe 40% - 50% neurology at steady state, 20% - 30% psychiatry, and 10% - 20% endo, and 10% - 20% immunology.
This is a few years down the road. You can see us shifting towards a higher probability therapeutic area. That's probably where I would point you to. Just know that those are ranges. Sometimes science takes you into different areas.
That's perfect. Moving to some of the specific programs, we expect phase II data by the end of the year from a NBI-770 NMDA candidate. Can you maybe start with discussing what the rationale is for using an NMDA negative allosteric modulator in major depressive disorder? How could NBI-770 be differentiated from other therapies for major depressive disorder?
Sanjay, you want to take this one?
Yeah, sure. Hi, Phil. NBI-770 is an NR2B NAM or negative allosteric modulator. It really builds upon the hypothesis established by ketamine, and specifically esketamine. We're really happy that Spravato is doing quite well in the area of treatment-resistant depression. Here, what we're trying to do is essentially have efficacy at least similar to Spravato, i.e. ketamine, but maybe avoid some of the dissociative and cognitive side effects that one sees with ketamine. The hypothesis rests on NR2B being extra-synaptic versus NR2A, which is synaptic. That may confer a differential safety profile while preserving efficacy. That's essentially kind of the rationale behind NBI-770.
In terms of the data that we'll see at the end of this year or towards the end of this year, can you remind us of the design of the phase II study and what results would Neurocrine consider proof of concept that would justify further development of NBI-NBI-770 and suggest to you that you're on the right path?
I think the first thing is it's a relatively small signal-seeking study. I just want to make sure expectations are appropriate. It's a 72-patient study total. It's divided into three active arms plus placebo. We do have one-to-one weighting for active versus placebo because we have learned that having that equal randomization actually preserves a placebo response and minimizes a placebo response. We're looking at a change in MADRS score at day five in this study on the assumption that our efficacy will be really rapid in terms of onset. As you mentioned, we'd be expecting these results later this year, in fact, in the October-November timeframe.
What magnitude of change in MADRS is possible by day five? That seems like an incredibly short period of time.
Spravato demonstrated about a three-to-four-point change as placebo adjusted. We're looking for that, and we think it's very possible.
What would be the next steps for NBI-770 should the phase II data be positive?
A lot of it relates to obviously the degree of efficacy and the safety profile we see. We have a couple of options, I think. One is to go straight into registrational programs if we're confident that we've got the right dose and dosing frequency. As I said, we have three different doses in our phase II study. If we need to establish more confidence in that dosing paradigm, we would move to maybe a larger phase II. We'll make that decision after we see the data later this year.
Great. Moving on to osavampator. In April of last year, Neurocrine shared positive phase II data from the phase II study. We'll get more data this weekend. I guess maybe to set a background, can you give us a brief overview of the data that were disclosed in April of last year and maybe discuss what we expect to learn from this weekend's medical meeting presentation?
Yeah, with respect to the historical data, maybe Kyle should address that. Maybe I'd start by just, again, framing this compound. We just talked about ketamine and NMDA. Osavampator is an AMPA potentiator, actually the kind of the same mechanism of action, really, but downstream. We think that will confer a differential safety efficacy profile. I think we're really happy with the safety tolerability data we've seen. This is a target, AMPA specifically, that industry has been trying to drug for a long time. One of the issues has been the therapeutic window. We're actually really happy with the molecule we have, which has low intrinsic agonistic activity, but it's a real potentiator. We haven't seen some of the hurdles that prior compounds which have failed have seen, specifically seizures. Fingers crossed, we haven't seen that. We hope not to see that in larger studies.
In terms of the historic disclosure, maybe I'll toss it to you.
Yeah, not a lot of data came under when we first had the phase II trial, and that was done deliberately. You know, this is a program that came out of our collaboration with Takeda. It was part of a profit share, and both organizations agreed that let's get to the data, and then we'll figure out next steps there. The data was overwhelmingly positive. We wanted to, first and foremost, protect the data and memorialize that in a number of patents that were filed around that. We wanted to keep the data close to the vest there. Also, we know that this target has been high interest to companies over the years, and there's a number of compounds on the shelf that touch on this target. Certainly didn't want to give them any heads up on potential trial designs or data that could inform or help accelerate those efforts.
We've had a long history of doing that here at Neurocrine . The data itself, this was a study of about 185 subjects. We tested two doses of osavampator versus placebo. Both doses were active, and we picked one of those to move forward into the phase III study as we've designed them now. The dose that was the most effective really had some spectacular results at both day 28 and 56 and showed a continuing improvement in the MADRS and efficacy as time went along. At the end of the treatment period, not only were the results statistically significant, but we saw an effect size that ranged from about 0.55 - 0.75 going from day 28 to day 56. That really puts you in quite rare air in terms of efficacy there at that one dose that looked really, really clean.
That's the dose that we selected to move forward in phase III. Safety and tolerability was quite good. The only thing that really separated from placebo was headache, and that really is quite a benign profile versus the other medicine that Sanjay already discussed, ketamine, that works upstream at the NMDA receptor. Very clean profile overall. We're excited with what we have. That's why we started all the phase III trials with osavampator. They're all ongoing now, and the plan is, you know, looking at a high-level data in 2027. We'll be able to fine-tune that as we get further along in recruitment. The data across both doses, greater visibility in the safety and tolerability will all be available at the data set we'll be sharing at site congress here over the weekend.
On the data this weekend, I think the most controversial part of the initial disclosure was that it wasn't disclosed which dose was the one that was statistically significant. We will see data from both doses.
All will be known.
All will be known.
By Saturday evening.
That's great. I think investors will try to figure out how the doses compare and contrast. Anything else that we should be paying attention to as we look through the scientific presentation?
Anything you'd add?
For me, Phil, I would just add, especially on the safety and tolerability of this class, when you look at some more detailed data on safety, there is really no substantial difference between both of the doses versus placebo, which is going to be important as we take this forward into phase III. You will get some exposure response data that you'll see this continuing effect that Kyle talked about. When you layer in the safety and tolerability, you really have a compelling program here that's in phase III studies now.
I would add this approach that we have that really incorporates NBI-770 to osavampator. It's consistent with our strategy working in validated biological pathways. This was really defined by ketamine and years earlier by PCP. They worked through the NMDA receptor and playing different targets along that pathway. What we're seeing is a potentially evolving view that working at the NMDA receptor can give you a fast onset of action or working downstream at the AMPA receptor directly could give you a big magnitude effect of really safe, well-tolerated, and safe medicine as well. We're playing both sides of this pathway. I think that there's ample opportunity here and the different profiles will be able to benefit a wide range of patient types with major depressive disorder.
That's another example of de-risking, right, with attacking a disease with different mechanisms.
On the— Sorry, go ahead.
Yeah, I was just going to add finally, it's really intriguing data. Clearly, we'll be sharing all this later this week. I think what's interesting is that the efficacy seems to increase over time. Our hypothesis here is that we're actually remodeling synapses secondary to BDNF and mTOR signaling. If one looks at the efficacy data at day 56 versus day 28, it seems to get better over time, which is interesting. We have the opportunity to look at maybe longer durations with our current phase III program where we actually have both open-label safety, and we have a randomized withdrawal study where we'll be dosing more chronically than our acute depression studies.
One of the questions we get on the phase III trial design is why Neurocrine is looking at a 56-day MADRS in phase III, whereas phase II had a 28-day primary endpoint. Is that because the efficacy increases over time? You think that day 56 results will be more impressive, more competitive?
That's exactly right. Conventionally, week eight or day 56 is the primary endpoint for antidepressants. We were looking at earlier time points because we're wondering if this would be more ketamine-like in terms of onset of efficacy. Essentially, we're happy that at least at the 1 mg dose, we saw significant efficacy at both time points.
Right. Just two more questions before moving on to NBI-568. First, in terms of the phase III data, what do you think you... What does the profile need to be for osavampator to be competitive? What change in MADRS at day 56? What side effect profile do you think would position it to be a standard therapy?
I think there's a lot of room in the MDD space, particularly for individuals who've failed antidepressants. There's a huge unmet need. Estimates in terms of remission rates vary, but we think that less than half of individuals actually achieve true sustainable remission with the currently available antidepressants. Again, huge unmet need. I think a lot of it will relate to how good the safety profile is. So far, it looks great. It could be used adjunctively to existing antidepressants without conferring an initial burden with respect to side effects. Clearly, we've been looking at efficacy. We're really bullish about the efficacy we saw. We saw actually a massive effect size of 0.73. It probably doesn't have to be as good as that, quite frankly, in the phase III program to be a really worthwhile and meaningful antidepressant.
At earlier time points, I think the effect size was also at least 0.5. These are big, meaningful effect sizes, at least with respect to the MADRS scale.
Yeah, I think there's such a low bar out there for antidepressants in general, whether you're looking at SSRIs or SNRIs. Typically, you look at effect sizes of 0.3 and above as being something that's clinically meaningful, and we're far above that in phase II. The bottom line is if you can offer something that's safe and well tolerated with the efficacy, and it's the only mechanism of that type out there, that's going to be something that people are really going to want to try.
That's perfect. Moving on to NBI-568. In August of last year, you disclosed the data from the phase II trial evaluating it in the treatment of schizophrenia. For those less familiar, can you briefly summarize the highlights of those data?
Yeah, so that was a phase II trial, about 210 subjects, consistent with what Sanjay described about NBI-770. This was a smaller phase II signal-seeking study. Each of the dose arms had 30 to 40 subjects, pretty small overall. It was an ambitious study because we really had to catch up with a lot of other players in this space. We had three main objectives. One is to study this molecule, dracoledine, for the first time in patients, see if we're leaving any efficacy on the table above the 20 mg dose, which is our target dose, the one that we saw in phase I, produce a nice biomarker effect as well as perform preclinically, and then pick a dose and dosing regimen for phase III. We were able to check all those boxes.
In our view, across all the doses, we used 20 mg, 40 mg, 60 mg once a day, and then a 30 mg b.i.d. arm. All the doses worked similar to what we saw with osavampator. Even if you pool the doses together, we saw a stat-sig result on PANSS, which is the primary endpoint for the study. The dose that performed the best was the 20 mg once-daily dose. With that, we saw about an 18.2 improvement in the PANSS, and that resulted in a placebo-corrected improvement of about 7.5 points with an effect size of 0.61. Pretty robust results when you stack that up with other approved antipsychotics. No food effect, no titration, no GI issues.
The only thing that separated from placebo on safety and tolerability was a little bit of dizziness, and that's consistent with the mechanism, but nothing there that you would see by agonizing M2 and M3 in terms of those GI side effects. That is a result of the medicine that we're developing being highly selective for M4. Overall, very robust data set. Not only did we hit on PANSS, but all the other scales that we had in the study, whether it was the CGI-S, the Moderate Factors Positive Score, the Moderate Factor Negative Score, all of them were positive at the 20 mg dose and as early as week two. That continued throughout the course of the study. That is the dose at strength and dosing regimen that we took into phase III. Now we have both of our phase III studies up and running.
Just as a quick reminder, it's one dose versus placebo, a much more simplified phase III trial than our ambitious phase II design in about 280 subjects. Replicate studies are ongoing now, with data earmarked for 2027.
Perhaps the most controversial aspect of the data was the fact that the 20 mg dose looked a bit better than the higher doses, at least in terms of PANSS score. Is there a mechanistic rationale for why patients on a lower dose would do better? Do you think the difference between the doses was real, or is it just within the noise of the measurement?
Yeah, I think it's just part and parcel with the variability of the study, given the small intercross of different doses. The other thing to keep in mind here, it's not something that I could quantify for you or anyone listening, but this was kind of an adaptive trial design in the sense that we started at the 20 mg dose, and only after did we see good safety and tolerability did we go to that next study. We're really interested in preserving safety and tolerability on this mechanism, as we believe this is a key differentiator for us. As you take that approach and you go to the next cohort, time has gone by. It's not a parallel design. All the subjects aren't getting at different doses, aren't getting their doses at the same time, and they're across different sites.
All those things aid into presenting greater variability than if you would have done a traditional one-to-one active to placebo parallel study design. That's probably it more than anything else. We see the same transient increase in heart rate across all the different doses and at a similar effect. Overall, you know, our view of an irregular dose response is, you know, we didn't expect it in the phase II, but it's not surprising. We've seen this in our own other programs in years past. We've seen that on contemporary medicines like Caplyta and Vraylar in this space. The bottom line, at the end of the day, we don't think you could have had the outcome that we have on the 20 mg dose by accident. It's a very robust data result.
Again, similar data across all the sites on the 20 mg dose and all the different efficacy parameters move in the same direction. We're excited about the data that we have. Obviously, you know, science has to work for us in phase III, but I think we have good reason to believe, based on what we saw in phase II, that we'll have a good chance for success in phase III.
Perhaps one of the most surprising developments in the muscarinic program space over the last year was INGREZZA's failure. Did Neurocrine learn anything from that? Is there anything from that failure that informed the development plan for NBI-568?
I guess I would say two things. A reminder that immaculate dean works through a different mechanism than dracoledine. It's a positive allosteric modulator. It requires acetylcholine to drive efficacy. We don't have that obstacle, if you will, with dracoledine because it directly activates M4 itself. It doesn't need anything else to aid in that effort. That's number one. Number two, I think that it speaks to the benefit of this being a very simple, well-controlled, managed study. We all appreciate that dracoledine's first study that got a lot of people excited about just M4 was a very small study. It was almost academic in size. It was about 75 - 80 subjects across five clinical sites. In their phase III trial, they went from five sites to 25 sites. They went from 75 - 80 subjects to 350 subjects. U.S. only to U.S., ex-U.S.
Multiple CROs involved, and they were required. All those things, you can still be successful, but they don't help in having a positive outcome there. In our phase III study, we really want to make sure that we pick up on the benefits of simplicity. Again, just one-to-one active to placebo parallel design for our phase III trials that we're running with dracoledine. That's the plan. Let's continue to manage and oversee the studies as we do at Neurocrine and let's wait for the data in 2027.
In terms of the phase III data, what do you think NBI-568 needs to produce in order to be competitive? What magnitude of PANSS change? Is it possible that NBI-568 could compete simply on safety, tolerability, and convenience, even if the PANSS falls a little bit short of Cobenfy?
The last aspects you mentioned are ones that we know we have out of the gate. Just the selectivity and the physical chemical properties of the molecule being once a day. These are all things that are ones that we think we have good reason to believe the efficacy piece. Obviously, you need to manage the trial. You need the science to work out. The last time we drew a line in the sand and shared where we thought we should be in terms of efficacy, we missed and we paid the price on that. I appreciate the question. I will remind everyone listening in that Cobenfy in their phase III trial saw a total PANSS, placebo-controlled PANSS improvement of about 10 - 12, somewhere in that range. I'm sorry, that was 8 - 10.
8 - 10.
If you look at some other antipsychotics in this space, like Caplyta, it was in the 4 - 5 range. There is a lot of variability in terms of placebo-corrected PANSS scores and success on those products. I would say Caplyta has been a great success out there and maybe not the most efficacious, at least on clinical outcomes. I tend to look at for myself more on effect size because that level sets the different clinical data sets in terms of variability. We did see a nice effect size in phase II of about 0.6. I should call out that Caplyta, Vraylar, and Rexulti, three medicines approved for schizophrenia, have effect sizes around 0.3. I would all put those in successful medicine categories out there.
I think ultimately, at the end of the day, we need to see a good effect size that's statistically significant, of course, and show good safety and tolerability, and that once a day medicine is going to pay dividends in this patient population. We know that's been the case because of INGREZZA.
Yeah, and maybe just to add to what Kyle mentioned, safety and tolerability is so important in terms of long-term compliance, particularly in this patient population. Just to add to the once-daily, the no need for titration is a big deal, a much better GI profile. Indeed, we didn't really see any adverse events related to the GI system between active and placebo. Another thing is food effect. If CRENESSITY has been given twice a day or 3x a day and you can't take it an hour before eating and two hours after eating, there are actually not many hours in the day you could take that medicine. Just stuff like that is super important in terms of compliance, particularly for chronic psychiatric diseases. I just want to make sure that, yeah, we just emphasize that.
Perfect. There are a number of early stage candidates we didn't discuss specifically, including other muscarinics, a couple new VMAT2 inhibitors, and the selective Nav1.2/1.6 inhibitor for epilepsy. Which one of those would you like to highlight for investors? Which should we start to dig into?
Obviously, given success of INGREZZA, we would point you to the next-gen VMAT2 inhibitors. Those are the ones that we have great confidence in. We know the clinical pathway. We know what that looks like for programs at this stage and getting them through for tardive dyskinesia. The real excitement comes here, comes from knowing, appreciating that we believe we've dialed in the right properties to these molecules to offer not only those same things that are ones that we like about INGREZZA, but also a long-acting injectable option. We think that will help drive compliance and perhaps greater efficacy by getting optimal exposure in patients over a longer period of time. These are the things that we're looking at building in and having be a part of our next-gen efforts. There's two molecules in phase I right now. That's NBI-890 and NBI-675.
We'll look at taking those forward to a similar development point and picking the best one to move forward on that. That particular program is one that should move through the clinical development space pretty rapidly, given what we know.
Great. We are just about out of time, but one last question. Because this is a neuropsych day, we've focused our questions on your neuropsych programs, but obviously, the commercialization of INGREZZA and CRENESSITY are key topics for investors. Any update or comment on either of those you'd care to share as the third quarter approaches this end?
Obviously, we don't give forward-looking insights into the business, but I will say the company overall is firing on all cylinders. If you look at enterprise-wide, combining the revenue from CRENESSITY with INGREZZA, we had $682 million in revenue in Q2. That's 17% year-to-year growth. On INGREZZA, we saw record back-to-back NRX quarters from Q1 to Q2. We saw a record TRX quarter in Q2 as well. Our market share increased in new to brand as well as TRX. How this is all happening eight years in is pretty remarkable. I think it says a lot for the strength of our commercial team. With CRENESSITY adding $53 million on top of the $624 million for INGREZZA, really going down that path of changing the standard of care for patients.
That's all been aided by the great positive receptivity that we've had out there in the advocacy community as well as physicians. The company's never been stronger from a financial perspective, and I'm really liking what we're seeing from our R&D organization. I think we're well on our way of becoming that leading neuroscience company, but it does start with INGREZZA and CRENESSITY to help us go down that path.
That's perfect. With that, I think we are out of time. Kyle, Sanjay, and Todd, thank you for joining us today. Very interesting discussion.
Thanks, Phil.
Thanks, Kyle. Take care.