Alumis Inc. (ALMS) Earnings Call Transcript

Good morning, everyone, and welcome to the Alumis conference call on the LUMUS Phase IIb top line results. [Operator Instructions] As a reminder, this call is being recorded, and a replay will be made available on the Alumis website following the conclusion of the event. I'd now like to turn the call over to John Schroer, Chief Financial Officer at Alumis. Please go ahead, John.

Thank you. Good morning, and thank you for joining us today. Before we begin our formal comments, let me remind you that during today's webcast, we will be making forward-looking statements that represent the company's intentions, expectations or beliefs concerning future events. These statements represent our views as of this date are subject to risks and uncertainties and should not be relied upon as representing our views as of any subsequent date in the future. With me on the call today are Martin Babler, our President and Chief Executive Officer; and Dr. Jorn Drappa, our Chief Medical Officer. Martin will open the call and then hand off to John to walk through some of the data, including the prespecified subgroup analyses. Martin will then review next steps before wrapping up the prepared remarks, then we'll take questions. I would like now to turn the call over to Martin.

Thank you, John, and thank you all for joining us. Let me start by acknowledging that the LUMUS Phase II study this trial did not meet the primary and secondary endpoints. This was clearly not the outcome we expected. However, this Phase II trial has provided important insights and a clear path forward for envudeucitinib in type 1 interferon-driven diseases, including SLE. The LUMUS data confirm the favorable profile that we have seen to date. In LUMUS, envudeucitinib was well tolerated with no unexpected findings and was comparable to placebo. These data strengthen the overall safety data set we have for envudeucitinib, especially as we continue to expect our NDA submission for moderate to severe psoriasis in the fourth quarter of this year. Robust clinical responses were observed in the prespecified subgroup of patients with high interferon gene signature, which represent approximately 60% of the patients enrolled in LUMUS. This analysis shows meaningful treatment effects across primary and key secondary efficacy endpoints. These results are consistent with the type 1 interferon targeted mechanism and confirm that envudeucitinib is working as designed. The subgroup of the population is an established and readily identifiable patient group and represents a large opportunity of moderate to severe SLE patients. Jorn will outline the factors we believe contributed to the overall outcome as well as the meaningful clinical benefit that we saw in the prespecified subgroup of patients with high interferon gene signature. On this call, to simplify the lexicon, we will be abbreviating high interferon gene signature to IGS High and, similarly, low interferon gene signature will be IGS Low. We believe that we have a clear path forward for regulatory engagement on Phase III development for envudeucitinib and SLE, and we do not foresee significant changes to our overall development time line. I'll now turn the call over to Jorn to go into more detail on the LUMUS results.

Thank you, Martin. LUMUS was a Phase II randomized, double-blind, placebo-controlled study. that evaluated multiple doses of envudeucitinib in adults with moderately to severely active autoantibody-positive SLE. The trial enrolled 408 patients who received 1 of 3 envudeucitinib doses or placebo of 48 weeks in Part A. The primary endpoint was the assessment of improvements in overall disease activity using the composite endpoint BICLA at week 48. Key secondary endpoints are listed on this slide. After week 48, patients had the option to roll over into the long-term open-label extension portion of LUMUS, which is Part B. Let's now review the efficacy data. As Martin said, envudeucitinib did not achieve statistical significance with the primary endpoint in the overall population at week 48. The same was true for SRI-4 and other key secondary end points. With respect to safety, we were pleased to see that in LUMUS, envudeucitinib treatment was well tolerated through week 48 with no new safety signals. Overall, incidence rates were lower on active treatment compared with placebo for treatment-emergent adverse events, serious adverse events and adverse events of clinical interest. Notably, there were no reports of MACE, extended MACE or malignancies in any treatment arms. We were encouraged to see that the pharmacodynamic data from LUMUS confirms that envudeucitinib worked as intended. As a TYK2 inhibitor, it blocked downstream type 1 interferon signaling. And here, we see that in a dose-dependent manner. On the left panel, you see expression levels of [ SICLAT 1], a biomarker that is exclusively sensitive to TYK2 inhibition. On the right panel, you see a 4 gene panel that measures the expression of interferon regulated genes. By both of these measures, the top dose of 40 milligrams BID achieved strong target engagement with maximal down-regulation of the type 1 interferon pathway. Before we go further into the LUMUS analysis, I would like to provide a little bit more context about the interferon gene signature characteristics in lupus patients, which play a key role in treatment response to type 1 interferon targeted treatments. The distribution of interferon gene signature is bimodal with a natural separation into 2 distinct groups: High and Low IGS groups. IGS High is an established patient group that represents approximately 70% of moderate to severe SLE patients, as reported in the [ SPOX ] cohort study. And as Martin mentioned, this subgroup is readily identifiable using commercially available assays. IGS High patients tend to have greater disease activity and typically respond more favorably to interferon pathway targeted therapies with lower placebo [ risk of ] rates. Conversely, IGS Low patients respond less favorably to interferon targeted therapies and show higher placebo response rates. Now let's look at the key factors that affected the LUMUS study outcome. The baseline demographics were generally well balanced across groups. There were 2 key factors that impacted the overall study outcome as follows. The first is the proportion of patients with the interferon signature. The second is the disease activity as measured by BILAG. BILAG stands for British Isles Lupus Assessment Group, which measures how active a patient's lupus is across 9 organ systems. Symptoms are graded through a letter scale where A denotes severe disease activity, B denotes moderate disease activity, and C denotes mild disease activity. First, let's talk about the interferon signature. LUMUS enrolled a lower-than-expected proportion of IGS High patients and, therefore, also a lower-than-expected proportion of patients with severe disease. For context, other Phase II and Phase III trials achieved approximately an 80-20 split between IGS High and IGS Low patients. The split in our study was closer to 60-40. As to why this occurred, one reason we have [ often cited ] is that having more approved treatments available for SLE has reduced the pool of patients with severe disease activity that are available for clinical trials. In LUMUS, the interferon gene signature status was a stratification factor, but was not a requirement for inclusion into LUMUS. The result was more IGS Low patients being enrolled than we had anticipated. Second, with regard to disease activity, the proportion of patients with severe disease activity, as indicated by BILAG-A, was slightly lower than the proportion of patients with moderate disease activity as indicated by the presence of at least 2 BILAG-Bs. Patients with IGS Low respond less favorably to interferon pathway targeted therapies and also show higher placebo response rates, and both of these turned out to be true in our study. When we analyzed the LUMUS data for the prespecified subgroup of IGS High patients, we observed clinically meaningful benefit across primary and secondary treatment measures in these patients. I will now walk you through these compelling results. On the left side of the table, you'll see the IGS High subgroup, and on the right side of the table, you see the IGS Low subgroup. In the green rectangle, you'll observe that for the IGS High patients, on all endpoints, there are robust responses with clear separation from placebo. I'll also point to impressive results, not only on BICLA but as well on [ CLASI, SRI-4 and LL ] death. To achieve LL death, several criteria that incorporates both disease control and limitation of steroid exposure have to be satisfied. LL death is important because it answers the question of whether patients achieve a remission-like state of low disease activity, and it has a strong connection to long-term outcomes. Conversely, in the red rectangle, you'll see that there was no apparent clinical benefit from envudeucitinib in the IGS Low subgroup. And these results are consistent with the TYK2 mechanism of action, which primarily acts by demodulating downstream type 1 interferon. On this slide, in the top panel, in the next slide, we show the response over time by dose in the IGS High subgroup, with a clear separation from [ CBO ] shown in the yellow line as early as 24 weeks. In the bottom panel, you'll see the lack of response in the IGS Low subgroup. I'll now walk you through several slides comparing our data to published data on deucravacitinib and anifrolumab in IGS High and IGS Low subgroups. The usual caveats around cross-trial comparisons, of course, apply. Starting with Slide 13 for the IGS High subgroup, we see overall strong response rates for BICLA for envudeucitinib compared to placebo and also compared to deucravacitinib and anifrolumab as well. In contrast, this slide shows that it's much more difficult to characterize responses in the IGS Low subgroup. Response rates are a lot more variable with no clear patterns. The same holds true if we look at the SRI 4 endpoint. There was a clear separation from placebo in the IGS High subgroup. And again, in contrast, this slide shows that for SRI 4, it's more difficult to characterize responses in the IGS Low subgroup with variable response rates with no clear pattern. Finally, on Slide 17, we captured BICLA plus 4 key secondary endpoints, including CLASI 50 and SRI 4 response, active joint count reduction and glucocorticosteroid taper for envudeucitinib compared to anifrolumab. Again, there was clear separation from placebo on primary and these key secondary endpoints. To conclude, although LUMUS did not achieve its key objectives in the overall study population, the trial did provide us with deep insights that we can build upon moving forward and incorporate our filings into Phase III development. Before I turn the call back to Martin, I would like to express our gratitude to the patients, families and investigators whose participation made the LUMUS study possible. Martin?

Thank you, Jorn. The data in the IGS High patients show a clear treatment effect, which we believe defines a patient population for a potential Phase III design and sizing. We are preparing for an end of Phase II meeting with regulatory authorities and do not foresee significant changes to our original development time line. We have some additional work to further characterize the responder population and prepare the LUMUS results for a future medical meeting. We believe that this represents a clear path forward for SLE. We have accomplished a lot this year, and we have some additional milestones that will be important for Alumis going forward as well. We have now demonstrated that envudeucitinib has opportunity to help patients with both IL-23 and type 1 interferon driven diseases. We recognize that the key value drivers are psoriasis and pursuing interferon-driven diseases, such as SLE. More broadly, we will continue to evaluate opportunities to maximize the value of our TYK2 portfolio for both patients and shareholders. With that, we'll take your questions. Operator?

Thank you, Mr. Babler. We would now like to begin the Q&A portion of the call. So please hold for a brief moment while we poll for questions. Our first question comes from Eric Schmidt at Cantor Fitzgerald.

Sorry about the outcome here. Yes. I guess, how do you think about a future Phase III trial controlling for interferon gene signature high such that this doesn't happen again? Is there a way of instituting an assay that requires some percent of patients to be GSI -- interferon gene signature high or some other metric that you're going to look at? And do you need to develop an assay if you're going to do that?

No. So the assays are commercially available. We did actually use this assay at screening and already in the LUMUS study. But since at the time, we did not have evidence that envudeucitinib would have no benefit for the negative population, we decided not to make this a requirement. That is something that now can clearly be adjusted since we have compelling evidence that there was no apparent treatment effect in the negative subgroup while there was quite convincing efficacy in the positive group. So you could -- as a consequence, you can either completely focus future study on the interferon signature high subgroup or at least cap the proportion of negative patients. And so we'll do further analysis to see which one of these possibilities makes more sense. It does seem pretty clear that this drug did not have a beneficial effect for the negative subpopulation. So one, certainly, we would focus the future development on the population where it actually works. And that is quite consistent with our mission a precision immunology company.

Our next question comes from Thomas Smith at Leerink.

A couple, if I could. Just wondering if you could give a little bit more color on your expected timing here for the regulatory engagement in the next update? And is the expectation here, you're thinking this would require 1 additional study, like 1 Phase III study? Or is there potentially a need for 2 Phase III studies? And then secondly, I was wondering if we could just get your thoughts on the dose response. It seems like the 40 mg BID performed quite well in that IGS High subgroup on BICLA, but it seems like the 20 mg QD dose also performed well on some of these other secondary endpoints. So just wanted to get further thoughts on potential dose response in this study.

Yes. So to start with the first question, we will engage with regulators as soon as possible, hopefully, by the end of this year. Although as you all know, the FDA's calendar, tends to be quite busy towards the end of the year. But we will work expeditiously to prepare briefing materials to request an interaction. With respect to dose response, I think we have a little bit more work to do. In this study, we used certain imputation methods where patients who either fail to taper steroids or who had adverse events or clear evidence of no efficacy were imputed as nonresponders. And so that can actually lead to some distortions across those groups. And so we will do some further detailed analysis, including the detailed dose response modeling to get a better handle on dose response. You're right that in the slides, at first sense, it does not appear that there was a dose response for some [ infants ], whereas there was for others. But I think we will reserve judgment until we have actually done this dose response modeling. Certainly, if you look at the PD, there was a very clear dose response where there was progressively deeper inhibition of the type 1 interferon signature with increasing doses. But that does not immediately become apparent in the clinical outcomes.

Our next question comes from Terence Flynn of Morgan Stanley.

Great. I guess 2 for me. The first, I was just wondering how you think about these data in the context of other indications that you might be considering? I know that was something that we're expecting to hear more about later this year in terms of how you guys are thinking beyond the current indication set. And then the second question, was just wondering if you could comment on the placebo behavior in the 2 different subsets. It looks somewhat different. So just any theories for the differences in the placebo behavior across the 2 subsets.

Yes. So to start with the second question. I think that has been a consistent [ validation ] across multiple trials, and it was already seen in the anifrolumab trials that the interferon negative subgroup or the interferon gene signature low subgroup had a much higher placebo response rate. So if you recall, the anifrolumab [indiscernible] to the programs, so it was the active arms actually had relatively comparable response rates in between the signature high and signature low group. But the placebo response rates were much higher. And the precise explanation for this is, I think, not completely clear. It just reflects the fact that these are probably biologically distinct populations. With the ISG High population typically having more severe disease, that is probably a little bit less susceptible to placebo responses than the more moderate or milder disease that the interferon gene signature negative patients have. And I'm sorry, remind me of the first question. It escaped my mind.

Just the other indications. So as you look data like other indications you might pursue, or does this change anything in terms of development plan for those indications?

I think it has strengthened our conviction that this is a rational target in type 1 interferon mediated diseases, at least those patients who have a bona fide interferon signature. So certainly, the results we've seen in the positive subgroup with respect to the [ CLASI ] would strengthen my conviction that this should be working in cutaneous lupus as well, that we have previously communicated that we're considering development and in Sjogren's, and we'll continue to evaluate that.

Our next question comes from Derek Archila at Wells Fargo.

Maybe just to carry off of Terence's question, just in terms of the future indications like Sjogren's and CLE, is this now the strategy to enrich for interferon high patients as well for those studies to kind of enrich the population? And then Jorn or Martin, just curious, like, are there any precedents supporting enriching for interferon gene signature high patients after an overall miss? And any other types of indications or what precedents would you point to in terms of bringing this to the FDA to narrow down the population?

So I think there's plenty of precedent for overall negative trials where convincing responses were identified in a subgroup and then subsequent trials honed in on that subgroup that seems to be more amenable to treatment with a given mechanism of action. I don't think that for this -- for the type 1 interference signature, there's a precedent quite yet. But it is certainly a well-known paradigm in drug development that if you have a study in a broader population that clearly identifies a signal in a subgroup, that you can then subsequently go on to develop in the subgroup. And I would expect that the FDA would look at this at our data and come to the same conclusion as we did that there really did not seem to be any appreciable benefit for the interferon low, and then agree with us that subsequent development in lupus and potentially in other indications as well should be focused on those patients that have the biology meshing the mechanism of action.

Our next question next question comes from Alex Thompson at Stifel.