INmune Bio Inc. (INMB) Earnings Call Transcript & Summary

Greetings, and welcome to the INmune Bio's MINDFuL Phase II Topline Data Conference Call. [Operator Instructions] As a reminder, this conference is being recorded, and a transcript will follow within 24 hours. [Operator Instructions] At this time, it is my pleasure to introduce Mr. David Moss, CFO of INmune Bio. David?

Thank you, Chelsea, and good morning, everyone. We thank you for joining us to learn about the topline results of INmune Bio's MINDFuL Phase II trial in early Alzheimer's disease. With me on the call from INmune Bio are Dr. RJ Tesi, CEO; and Dr. CJ Burnham, Head of Neuroscience. Also joining us is Dr. Judith Jaeger, external consultant to INmune Bio. Before we begin, I remind everyone that except for statements of historical facts, the statements made by management and responses to questions on this call are forward-looking statements under the safe harbor provisions of the Private Securities Litigation Reform Act of 1995, including statements regarding the results of the Phase II MINDFuL trial. Potential regulatory pathways and future development opportunities. These statements involve risks and uncertainties that can cause actual results to differ materially from those such as forward-looking statements. Please see the forward-looking statements disclaimer on the company's press release as well as the risk factors in the company's SEC filings, including our most recent quarterly filings with the SEC. There is no assurance of any specific outcome. Undue reliance should not be placed on forward-looking statements, which speak only as of the date they are made as the facts and circumstances underlying these forward-looking statements may change. Except as required by law, INmune Bio disclaims any obligation to update these forward-looking statements to reflect future information, events or circumstances. Now it's my pleasure to turn the call over to Dr. RJ Tesi. RJ.

Thank you, David, and good morning. Today, we are proud to share the results of the MINDFuL trial. It is a double-blind, placebo-controlled Phase II trial that evaluated XPro as a treatment for early Alzheimer's disease in patients with biomarkers of inflammation. The results show that XPro is safe in early Alzheimer's disease patients and prevented cognitive decline in a predefined population of early AD patients with 2 or more biomarkers of inflammation. To be clear, analysis of the modified intent-to-treat population did not show significant effects on any of the primary or secondary endpoints in this 6-month Phase II trial. Although this is disappointing, we can draw 3 conclusions from the trial. XPro is safe in patients with early Alzheimer's disease, XPro appears to be effective in an easily defined in commercially relevant patient population with early Alzheimer's disease, and we see a path forward. We are also realists. We recognize there is work to be done and that work will require financial resources. We plan to explore partnering opportunities in parallel with our preparation for AAIC in July and the end of Phase II meetings with the FDA that will occur before the end of the year. The modified intent-to-treat sample of 200 patients is smaller than the intent-to-treat population. Modified intent-to-treat does not include patients who signed consent but never received a dose of study drug, either placebo or EXPAREL. The modified intent-to-treat population includes patients who were classified as having Alzheimer's disease based on expert clinical judgment, some of those without the benefit of a biomarker confirming the presence of amyloid. This was the standard of care at the time the trial began. And as you know, Alzheimer analytics -- an Alzheimer's disease analytics is a very active area of academic and business activity. Analysis of patient information after database locked revealed cases of clinically defined Alzheimer's disease that were amyloid negative. Because the modern or the current diagnosis of Alzheimer's requires clinical symptoms and verified amyloid positivity, the amyloid patients do not have Alzheimer's -- the amyloid negative patients do not have Alzheimer's disease, and they were excluded from further analysis. The 150 patient group with early Alzheimer's disease are biologically defined with amyloid and all have at least one biomarker of inflammation. To be clear, this is smaller than the modified intent-to-treat population. In the patient population of interest, we can accurately ask the question, does XPro affect cognition in early Alzheimer's disease patients with biomarkers of inflammation. Although XPro did not have a meaningful effect on the larger modified intent-to-treat group, when we focused our analysis on these 150 biologically defined Alzheimer's disease patients, we see a consistent benefit of the treatment with XPro on the primary endpoint of cognition measured by EMACC based on the number of inflammation biomarkers. The effect became clinically important in patients with 2 or more biomarkers. Positive effects were seen in the key secondary endpoints of behavior, the NPI-12 and on disease-relevant biomarkers in the blood, including pTau-217 and GFAP. To rephrase, there is a dose of inflammation effect. When we analyze the effects of XPro on the population that is biologically confirmed as Alzheimer's disease patients with a high probability of having neuroinflammation as defined by tumor or biomarker and treat the patients for 6 months with XPro, XPro prevents disease progression by preventing cognitive decline compared to placebo. XPro does this without any safety problems. It is a safe drug. There is no area. There are no other safety -- important safety signals other than injection site reactions that are frequent but manageable and will be discussed more later. The modern view of Alzheimer's disease recognizes that neuroinflammation contributes to the cognitive decline. We believe this trial provides concrete evidence that neuroinflammation plays a key role in the disease and targeting patients with neuroinflammation may slow its ravages. Because of this trial, we understand there is an interaction between pathologic burden and time. That is with high levels of neuroinflammation defined by 2 or more biomarkers of inflammation, a 6-month trial appears to be long enough to demonstrate the benefits of XPro using EMACC as a primary cognitive endpoint. With more subtle neuroinflammation defined as one biomarker, a longer or larger trial may be needed. We believe we have the clinical data needed to validate EMACC as the primary cognitive endpoint for an early -- for use of EMACC as in an early trial -- in a trial in early AD. We plan to confirm this with the FDA during our end of Phase II meeting before the end of the year. Overall, we see a path forward for the use of XPro in patients with early Alzheimer's disease and biomarkers of inflammation. And we understand what that trial needs to be and remain confident that XPro has the potential to be a best-in-class Alzheimer's therapy. I will let CJ explain the results of the trial. CJ?

Thank you, RJ. As RJ mentioned, I will be providing top line results with full results coming later in July at AAIC and a subsequent publication. While we haven't had a lot of time with these data, there are some things of which we are confident. First, the data show a reliable signal in the study population for which this drug is intended and hypothesized to work best. Second, the primary endpoint was not met because the placebo group in the overall study population did not decline as expected in the 6-month period. As RJ mentioned, this was driven by a subset of patients who are not amyloid positive and who had the lowest inflammatory burden. Third, the drug was safe. The most common adverse event is injection site reactions. There were no deaths and importantly, no cases of ARIA. Before I get into the data, I would like to describe our approach to evaluating a subgroup after missing the primary endpoint. Due to the smaller sample size of 100 patients, potential benefit was defined using effect sizes. The formal statistical measure is Cohen's d. Effect size is the appropriate metric with small sample sizes and when comparing across different measurements, for example, a cognitive test and a blood biomarker. Moreover, effect size is a measure of clinical meaningfulness, whereas e-value reflects how likely the effect you observed is due to chance. An effect size of 0.2 defines a small but potentially meaningful effect. More importantly, effect sizes are commonly used as an objective measure of signal detection in Phase II studies. We define success as a minimum effect size of 0.2 where XPro consistently outperformed placebo on multiple endpoints and only where those results are consistent with our hypothesis. In other words, if a benefit is observed on a measure that is not linked to our understanding of how XPro works, it's considered evidence against the effectiveness of XPro. On the primary endpoint of the trial, EMACC, which is a sensitive performance-based measure of cognitive function, we observed a benefit of XPro over placebo with an effect size of 0.27. We interpret this result as meaningful in the context of signals on the neuropsychiatric inventory, a measure of psychiatric symptoms where a directionally consistent effect size of 0.23 was observed. Finally, a benefit was also observed on the blood biomarker of AD pathology, pTau-217 with an effect size of 0.2. XPro and placebo-treated patients were not different on our key secondary endpoint, the CDR sum of boxes, and there was little change in CDR scores over the 24-week study in both placebo and treatment groups. We ascribe this to higher measurement noise in CDR, which relies on caregiver reports compared with the EMACC, which consists of objective cognitive performance tests. Despite not meeting the primary endpoint on the mITT population, we strongly believe this was a successful study. We observed a clear signal that was favorable and appropriate in the study population and study design. The results were both consistent across multiple measures and perfectly aligned with our hypothesis. Most importantly, these results provide the necessary information required to advance XPro into the next stage of development. I will now turn it back over to RJ to share the safety profile of XPro.

Thank you, CJ. I had the privilege of being the company's Drug Safety Officer for this clinical trial. The overall conclusion, XPro is safe. There were no deaths in the trial. The AEs we saw were typical for this age group. The median age was -- or the mean age was 74. And this was -- this fact is particularly important as early Alzheimer's disease patients are often elderly, have multiple comorbidities and are on multiple medications. Any treatment for Alzheimer's disease must deal with this reality and make the patient better. Drug-related side effects are not acceptable. XPro did not cause ARIA, any type of ARIA or other neurologic complications that required an urgent evaluation. There were no emergency MRIs, no emergency evaluations in either the clinic or the ER that required a neurologic assessment. As a reminder, in all trials using immunotherapy to treat Alzheimer's disease, ARIA has been seen and is most frequently seen or majority of which occurs in the first 6 months. Our trial, therefore, was long enough to observe any ARIA if it was going to occur. There was no organ toxicity, lung, liver, kidneys, et cetera. And I think the safety of the therapy is showcased by the way the clinical sites manage the patients. If the patients developed a urinary tract infection or some viral syndrome, the drug was not stopped and the patients were treated and recovered while on drug, placebo or XPro and were allowed to finish the trial. The most common drug-related adverse event was injection site reaction that occurred in 80% of patients receiving XPro compared to only 10% of patients receiving placebo. The injection site reactions were small areas of redness or pain at the needle puncture site in 2/3 of the cases. In general, these occurred early during the course of the therapy. 20 patients were put into a kind of a dose modification protocol that lasts 4 doses to manage their injection site reaction successfully. But ISRs or injection site reactions were the most common reason for stopping participation in the trial. This occurred in 10 patients. And during the trial, we learned how to manage ISRs better. And most of the patients who dropped out of the trial dropped out quite early in the patient enrollment cycle. So in other words, they -- in the first 1/3 of the period where the time of the trial was open, and we got much better at explaining and treating injection site reactions. So in summary, XPro appears to be safe in the target population and could be used to treat early Alzheimer's disease regardless of existing medical conditions or APOE4 status. In the trial, almost 70% of the patients were APOE4 carriers, including 25 that were APOE4 homozygotes, 17 of which received XPro. We have not identified any concomitant medications that need to be avoided, including anticoagulation or therapies for Alzheimer's disease, including anti-amyloid immunotherapy. We believe the cognitive benefits demonstrated in our clinical trial results -- in our clinical trial mark an important step forward in the fight against Alzheimer's disease. To our knowledge, this is the first time that neuroinflammation has been effectively targeted with beneficial effects in early Alzheimer's disease. We recognize the importance of this for those living with Alzheimer's disease and we look forward to moving forward. In summary, we are quite pleased with the results, and now I will turn the call back to the operator to poll for questions.

[Operator Instructions] And our first question will come from George Farmer with Scotiabank.

I was wondering if you could comment on any impact that the injection site reactions may have had on bias related to cognitive assessment population. I guess if you had an injection site reaction, it would have meant that you are on drug versus not. Could that have clouded interpretation in any way?

CJ?

Yes. So this is a great question. It was actually the first question we asked after we looked at the data, right? So what you would expect is if you have unblinding there and you control for that, the effect size would go away. So it turns out that when we controlled for that, and we did that a couple of different ways. We did a sensitivity analysis, and we also controlled for as a covariate in our model. In both ways, the effect was still there. In fact, the effect was somewhat numerically better. And I think that's an important point to this is that you have to make sure that you understand that those effect site or those injection site reactions are not unblinding.

Okay. Great. And then could you just give us some things to look forward to regarding your other 2 pipeline products, CORDStrom and INKmune, what can we expect over the rest of the year?

Yes. So I'll jump in and David may want to jump in also. Obviously, CORDStrom -- the cell therapy programs, such as CORDStrom and INKmune are -- remain a very high priority for active clinical development in the program. We're going to be pivoting towards a partnering profile or a position for XPro for Alzheimer's disease because of the resource needs. But we have the resources, the financial wherewithal to move -- continue to move forward with CORDStrom and INKmune, and we will plan to meet the time lines that we have promised on previous calls, which includes making regulatory filings on CORDStrom by the first half of next year and having additional data on INKmune before the end of this year.

Our next question will come from Gary Nachman with Raymond James.

So maybe just give a little bit more on how clinically meaningful an effect size for EMACC of the 0.27 is in the 2-plus biomarker group? And why do you think the CDR didn't show any effect size even in that subgroup? And maybe just as part of that, how much of an issue was the shorter duration of the study just to 24 weeks to see a benefit? Did you see the curve separate more as you got closer to the 24 weeks? So you would expect maybe a bigger separation if you go out farther like you indicated?

Yes. So thank you for the question. Let me answer part of it, and then I'd like to ask Judy to speak to this as well. I think that clearly, the 24 weeks was on the shorter side. We did start to see the curve separate towards the end. And I think that's an important point. And really the reason why the EMACC was chosen because of the potential to see that at that time point. Judy, do you want to comment on the other aspects -- other questions?

Yes. So meaningfulness is always a difficult question, but we do have some other drugs to compare this to now. And the anti-amyloid monoclonal antibodies at 18 months have comparable effect sizes to that we are seeing at 6 months. So it's -- in terms of the size of the effect, it's right there in the window that we expect and are seeing with other drugs.

I don't know if you want to add anything else, CJ. But -- so -- and then also just how are you confident that the number of biomarkers actually correlates with the level of neuroinflammation that each biomarker is incremental? And is it regardless of the type of biomarker. So it doesn't matter which 2 it is. And is positive amyloid actually an even better indicator of neuroinflammation?

So this is a great question. The answer is there's -- we don't know if those -- if adding those together is cumulative. It's not clear. What we think is happening is that these are multiple indicators of an immune system that isn't functioning properly. And the more indicators you have, the more likely that your immune system isn't functioning properly. So that is the way that I would describe it. But to your point, we don't know exactly if that's the case. What we do know is that the hypothesis is that -- or what we know from the literature is the more inflammation that you have, the faster your disease progresses, the worse your outcomes are. And by using the scale of 1 to 2 to 3 to 4 different biomarkers, we're seeing the same effect, right? So the more biomarkers you have, the more it's declining. So I think it's safe to say that there are some parallels there, although it might not be perfectly associated with -- or perfectly defined by that. And what was the second part of your question?

It was just on positive amyloid, just the importance of that as an indicator of neuroinflammation.

Yes.

So if you have more amyloid levels, would that lead to higher neuroinflammation and you could think of that correlation when you look at patients?

Yes. So another -- that is another great question that we think a lot about. I think the answer is probably yes. There's some evidence in the literature that amyloid is an immune molecule and may actually just be one additional biomarker. I don't know if that's true. I think we're getting closer to figuring that out. I think more importantly, though, what amyloid does, at least in this trial, is it provides a more homogeneous group. And in a smaller trial, that's going to really help you identify signals that are smaller or that require a larger sample size to see. So I think all those things are possibly true, but we don't have data to definitively say that.

Yes, Gary, this is RJ. Just to jump in here. I just want to remind everyone that the Alzheimer's community has really begun to focus very much on pTau-217 in the blood because it has both -- it correlates with severity and it gives some prognostic value. So it appears to be a very potent, shall we say, biomarker for the disease. And in this trial, in the -- in our target population, which are amyloid-positive patients with 2 or more biomarkers, those that received XPro had a favorable impact on pTau-217. So not only were we affecting cognition and behavior, but we believe we were affecting the basic biology associated with progression of the disease. So they all lined up well, and that's a very positive result from this trial.

Okay. Great. And then just the last question. If you'll meet with FDA if they sign off on the EMACC as a primary endpoint, how confident are you that you can move directly into a Phase III rather than conduct another Phase II to ensure you have the right patient population? And is there anything different you can do with the dosing or the administration of XPro given how safe it is and maybe to minimize the ISRs?

Yes. So let me start there and CJ will jump in. I want to emphasize on the safety part first that we learned a lot about the ISRs. All PEGylated drugs that are used in a subcutaneous manner like XPro have injection site reactions. And we learned a little bit about premedication too and et cetera, and dose escalation to prevent it -- prevent the problem. So we believe that moving forward that we will be able to manage this and decrease the issues associated with really the nuisance value, maybe nuisance is not an appropriate term, but the clinical implications of these injection site reactions. So we don't believe that's a problem. I don't think we want to say that we're going to -- we know if we can jump into a Phase III without talking with the FDA. We have the -- first things first, we need to make sure they're comfortable with EMACC and they give us the green light to use it as a primary endpoint in a registration trial. Given that, we can then use the data we have here to model exactly what that trial looks like. We've done the back of the envelopes, but they all depend on the FDA agreeing that EMACC is the best cognitive metric for use in patients with early AD. We are confident, but we need to hear it from them.

Our next question will come from Tom Shrader with BTIG.

Can you remind us -- I think you said it, but what's the number of patients that weren't plaque positive? Did you give that exact number?

Yes. Well, it turns out that we went from 200 to 150. So 50 patients were excluded because they weren't confirmed to have Alzheimer's disease.

Got it. And then if I remember, APOE4 was a sole criteria to get into the trial. In retrospect now, is that a mistake? Was that confounding or...?

No, no. no. Remember, we had -- you had to have 1 of 4 biomarkers. And 1 -- and the 4 biomarkers were either APOE4 carrier or homozygote carrier, an elevated CRP, C-reactive protein, an elevated hemoglobin A1c obesity, metabolic syndrome, diabetes or an elevated ESR, erythrocyte sedimentation rate of greater than 10 seconds. And the CRP was greater than 1.5 milligrams per liter. So you only had that one of those. 40% of the patients had 1 and the rest had more than one. So that's the way it broke out. And all that data was presented in detail at AD/PD back in March, the initial demographic.

I guess my question is of the APOE people, were they a confounding group that they have a high tendency of not having anything else? Or was that not a problem? But that's your biggest group to get in, I assume.

So interestingly, APOE4 -- go ahead, CJ. I'm stepping on here.

No, no, that's okay. I think the answer is I don't think they were confounding. In fact, most APOE4 patients -- most of the patients in this group, as RJ said, were APOE4. I think they were more likely to have an additional biomarker. And I realize I didn't fully answer an earlier question, and I'll answer here. As it relates to the biomarkers, it does not appear that one biomarker was more important than the other. This is sort of early analyses, but it looks like none of them were correlated with each other, and we don't see any signal where one biomarker does better. It really does look to be that the combination of biomarkers is what's driving the effect.

And then -- go ahead, Tom.

Just quickly, do you have p-values for either GFAP or pTau-217? GFAP has been pretty well behaved in a lot of trials that worked. I'm wondering if you have a p-value.

So you mean in the smaller sample population?

Yes, that's right.

So we don't -- well, we do have them, but they haven't been corrected. So the p-values that I would give you wouldn't be correct. But what I can tell you is that without the correction, we do have a nominally significant change in GFAP. But GFAP, the effect size in GFAP was 0.17 and it didn't meet our threshold of 0.2.

But it was moving along the lines of pTau-217 and the correlation of the 2 was quite high. So we'll provide more of that information at AAIC because like you, Tom, we believe GFAP probably is a good biomarker for this drug in this disease.

Okay, good.

This is Judy. I would just add, there was a high correlation with EMACC for both of these biomarkers, high and highly statistically significant.

Tom, I want to make a correction. I apologize. It was not GFAP that was nominally significant. It was pTau that was nominally significant.

Got it.

Our next question will come from James Molloy with Alliance Global Partners.

It's Matt on for Jim. Firstly, I wanted to talk about the 25% of the sample that you lost because they didn't have biologically confirmed AD as measured by amyloid. Can you just help us understand how that occurred and going forward in the potential Phase III, how that would be avoided?

CJ?

Yes. So this was really a function of clinical trials and the constraints and limitations that occurred. So the protocol originally stated that they had to be amyloid positive. At the time, we were using blood biomarkers, right? So we're measuring amyloid with blood. This is very early days. There was only one company doing it. And it just turned out that we lost too many patients to screen failures, and they couldn't quite scale up. And so we had to remove that criterion for a time until they could get back up to speed, in later protocols we were able to do that. But in the meantime, we just lost some patients. It's just a function of clinical trial logistics. That's not the case in the future. The blood biomarker tests are very good. There's other companies in the space, and they're easier to get back in time. So I think it's a function of being early and adopting new technologies. But the alternative was to do PET or CSF, neither of which were within -- were practical from a budget or a patient interest perspective.

Yes. I just want to reemphasize the rapid rate of change that's occurring in this disease. And as CJ said, the kind of the biomarkers that everybody is interested in things like amyloid, APOE4, pTau-217 and for all practical purposes, maybe not GFAP, they're all available within 5 or 6 days after you draw the blood sample today. Back when we started, it would take 60 days to get a blood amyloid back. So we've learned a lot from this trial, and all those learnings have really benefited our thinking on how to move to the next level of development with this trial. And many of the hurdles that we suffered and lived through early in this trial are no longer present. So I think we should be able to move quite efficiently.

Got it. And earlier, you mentioned that there was a similar effect size in this trial as compared to the amyloid in their 18 months or at their 18-month endpoints. Can you comment on which of the endpoints in this trial that was related to? Was it related to EMACC, CDR-SB, the biomarkers? If you could just give a little color on that.

Judy?

So I can handle the -- yes, I can handle the EMACC. So [ donanemab and lecanemab ] for their cognitive measurement used [ CDR ]. And for both of those drugs at 18 months the -- or effect sizes were below 0.2. So now for their CDR-SB was just around 0.2. So for comparison, we believe that the CDR is a noisier tool that requires -- that may require a longer observation period than we had. And remember, that was at 18 months. We're reporting results at 6 months.

Got it. And then lastly, if you could give any color on what a Phase III trial in the future might look like or whether that would be dependent on a partner coming onboard and what an ideal partner might look like for a Phase III trial for you guys?

Yes, and let me jump in. Go ahead, CJ.

Yes. I think the -- well, we still have some modeling to do. I mean these data are pretty early. But what's clear is that we have information that allow us to model it now. So we're going to have a clear understanding of what we need to do in the study population that we know it's most effective in. And then we're going to take that information, we'll bring it to the regulatory agencies. We're certainly going to talk to partners and the next steps will be decided there. So I think that's the best explanation we can have. But I think we have multiple different options, and that's a good place to be.

Yes. And if I can just add a little bit more color. As Judy said, with the effect sizes we're seeing, which are better than the cognitive measures used in the donanemab and lecanemab trials at 6 months, we are confident although it's not assured until we talk to the FDA, that the trials can be shorter and smaller than those trials, which, as you know, were 18 months and 1,600 patients. They were massive and long. We believe that when we get done with our end of Phase II meeting, we're going to have very good news on those 2 issues on size and duration. But it depends on the FDA agreeing that EMACC is an appropriate endpoint, cognitive endpoint for a pivotal trial.

Great. I'll hop back in the queue.

All right. This does conclude our Q&A portion. And I would now like to turn the call back over to RJ for any additional or closing remarks.

Yes. So I'm delighted to be able to make some closing remarks here. We recognize the need for an effective therapy for patients with Alzheimer's and for their loved ones and their caregivers, they need help taking care of these patients. XPro appeared to flatline cognition in the early AD patients with 2 or more biomarkers of inflammation. Admittedly, the full Phase II trial did not meet its primary endpoints in cognition, but we completed 4 important goals. XPro is safe in patients with early Alzheimer's disease. XPro seems to provide benefit in a well-defined, commercially relevant early Alzheimer's disease patient population. EMACC performed well, very well. We believe we have the data needed to convince the FDA that EMACC is a better cognitive endpoint than CDR for early Alzheimer's disease trials. And finally, there is a clear path forward, and we consider the trial results from MINDFuL, the MINDFuL trial, a win, and we'll seek the resources to move XPro forward in early Alzheimer's disease to the benefit of patients and the caregivers. Our next steps are to present the fully analyzed data at AAIC in the end of July, file for breakthrough designation with the FDA, conduct our end of Phase II meeting with the FDA. And during that meeting, we hope, as I said, to get agreement from them that EMACC is an appropriate endpoint for further trials, and we can come to you with exactly what that trial looks like. We will keep shareholders apprised as we make progress on these steps. I want to take a moment to thank the participants of the -- the participants or the patients in the study, their study partners because it's critical that they have a study partner, a loved one, the clinical trial sites and our vendors for making this trial possible. MINDFuL enrolled over 200 individuals and their accompanying study partners in 7 different countries across 3 continents from a total of 29 clinical sites. Without these incredible people and teams, and I will say without our incredible clinical operations team here at INmune Bio, we would not be in the position to learn or advance this development for the benefit of patients and their families with Alzheimer's. So with that, I will close the call and thank all you for your participation today.

Thank you, ladies and gentlemen. This does conclude today's conference call, and we appreciate your participation. You may disconnect at any time.