Annexon, Inc. (ANNX) Earnings Call Transcript & Summary

Hello, and welcome to Annexon's conference call. [Operator Instructions] Please be advised that this call is being recorded at Annexon's request. I would now like to turn the call over to Jen Lew, Chief Financial Officer. You may begin.

Thank you, and good morning. Earlier this morning, we issued a press release outlining the clinical trial results from our Phase II study of ANX005 in patients with Huntington's disease. The press release and the slides to which we'll refer are available on the investors center of the company's website at annexonbio.com. On the call with me today is our CEO, Doug Love. Dr. Ted Yednock, our Chief Innovation Officer, will join us for the Q&A session. Today's discussion will include forward-looking statements related to Annexon's current plans and expectations, which are subject to certain risks and uncertainties. Actual results may differ materially due to various important factors, including those described in the Risk Factors section of our most recent Form 10-Q and other SEC filings. These forward-looking statements represent our views as of this call and should not be relied upon as representing our views as of any subsequent date. We undertake no obligation to publicly update these statements. Now let me turn the call over to Doug.

Thank you, Jen, and I thank all of you for joining us today to review the encouraging final results from our Huntington's disease Phase II trial with our lead drug candidate ANX005. Targeting the complement pathway has been well validated, both clinically and commercially in a range of diseases. We at Annexon are building on those learnings with a differentiated approach that centers on blocking classical complement activity where it starts with the goal of providing more complete protection against harmful inflammation and tissue damage in a host of diseases. To that end, we're advancing 5 clinical-stage drug candidates to effectively block classical complement activity in specific compartments of the body, the brain and the eye. We're encouraged by the results achieved thus far in our pipeline, having demonstrated strong proof of concept with ANX005 in both autoimmune and neurodegenerative diseases. And this is only the beginning. With 7 clinical trials reading out across our pipeline over the next 2 years, we are exceedingly well positioned to provide significant benefit for patients suffering from complement-mediated diseases. Our late Co-Founder, Dr. Ben Barres, discovered that blocking the classical complement pathway preserves functioning synapses and brain health to prevent neurodegeneration in several preclinical models. This Phase II study in Huntington's disease is the first clinical evaluation of that discovery in patients, and we're quite encouraged with the results. The final data set provides 4 key learnings. First, ANX005 is proving to be a dominant drug candidate that powerfully inhibits both up and downstream components of the classical pathway. In this trial, it demonstrated complete and durable C1q target inhibition in the body and in the brain throughout the treatment period and well into the follow-up period. Second, with complete and durable target engagement, ANX005 stabilized the progression of disease in the full HD patient cohort for the entire 9-month treatment period as well as the follow-up period. Third, ANX005 provided rapid clinical improvement in HD patients with higher baseline complement activity again for the entire 9-month treatment and follow-up period. And finally, ANX005 was generally well tolerated over the course of the study with a favorable benefit-risk profile. Based on the totality of the data generated in this trial, we're excited by the promise of ANX005 as a treatment for patients with Huntington's disease, and we look forward to continuing this development. By way of background, Huntington's disease is a devastating fatal neurodegenerative disease that affects approximately 80,000 people globally with about 300,000 people who are at risk of inheriting the disease-causing gene. Huntington's disease patients unfortunately experience progressive decline in motor, psychiatric and cognitive function and have a life expectancy after diagnosis of approximately only 15 to 20 years. Today, there are no approved therapies that slow or stop the progression of Huntington's disease. And so there is high unmet need for targeted and effective therapies. ANX005 is designed to protect against the loss of functioning synapses to treat Huntington's disease by blocking C1q and the entire classical complement pathway. Synapses are a key component of normal healthy brain function, providing critical connections between neurons that control our motor and cognitive activity. When synapses are damaged or lost, the brain's overall wiring and connectivity is compromised, leading to neurodegeneration. In its normal function, C1q directly recognizes and tags excess synapses in development and initiates the classical pathway for their pruning. However, as part of the neurodegenerative disease process, C1q inappropriately recognizes and tags functioning synapses, activating and amplifying classical complement components that cause neuroinflammation, synapse damage, and ultimately, synapse loss. The results of this aberrant complement-driven disease process is the progressive decline in Huntington's disease patients, which we seek to guard against by blocking C1q in a classical pathway right where it starts. Our Phase II trial in Huntington disease evaluated ANX005 administered intravenously over a 6-month treatment period in patients with or at risk for early manifest disease, followed by a 3-month off treatment period. The study enrolled a total of 28 patients of whom 23 patients completed both the treatment and subsequent follow-up periods. The primary endpoint of the study included safety and tolerability of ANX005, pharmacokinetics of ANX005 in serum and in the CSF and pharmacodynamics as measured by C1q, C4 and NfL serum and CSF concentrations. Exploratory endpoints for the study included the composite UHDRS and total functional capacity or TFC, both clinical measures of clinical function. In terms of patient demographics on Slide 8, the patients enrolled in our study were largely consistent with prior Huntington's disease natural history study cohorts, including in TRACK-HD, a large biomarker study and participants with pre-manifest in early Huntington's disease. With that, let's now turn to the data. Looking first at measures of target engagement on Slide 9, ANX005 demonstrated rapid and complete inhibition of C1q in the classical pathway in both the serum and CSF, which was demonstrated throughout the 6-month treatment period and sustained well into the follow-up period. As shown in the graph on the left, ANX005 drug levels remained present in the serum for 4 to 10 weeks following the patients' last dose. Similarly, in the graph on the right, ANX005 also demonstrated full inhibition of C1q across the blood-brain barrier as measured in the CSF through the treatment period and into the follow-up period. Turning to Slide 10. Plasma and CSF NfL levels remained generally consistent through the 9-month study and were comparable to NfL levels described in published natural history data for HD patients, which is shown in the green bands in the 2 charts. Published data continue to suggest in slowly progressive neurodegenerative diseases like Huntington's disease, synapse loss precedes the loss of neurons and NfL changes consistent with the findings from our study. Importantly, there was no meaningful rise in NfL that could be associated with patient worsening in our study. Focusing on the clinical measures, we were quite encouraged with the durable stabilization of clinical disease progression in the overall patient population through both the treatment and follow-up period out to 9 months. Such stabilization was compared to the expected decline on large natural history Huntington's disease cohorts of this progressive disease and was consistent when measured both with the composite UHDRS and TFC, the 2 approvable clinical measure scales for Huntington's disease. In addition to the overall study population, we set out to explore an additional predetermined hypothesis in this trial. That is, patients with higher baseline complement activity might experience a rapid and more robust clinical benefit from treatment with ANX005 by fully blocking the complement cascade before it started. And natural history data produced prior to our clinical trial, the graph on the left shows that as disease progresses through pre-manifest and manifest stages, complement activity increases as measured by the activated complement component C4a in the CSF. C4a is an objective measurement of excess complement activity in the brain. As shown in the table on the right, C4a levels correlate with disease progression in multiple clinical measures in Huntington's disease, including both motor and cognitive function. To evaluate clinical outcomes based on complement activities, patients were split into 2 groups based on their C4a levels at baseline. Those with levels above the median were defined as high complement and those below the median were defined at low complement. In the graphs on Slide 13, you can see that those patients in the high complement group, represented by the green line, experienced a clear and rapid benefit over patients with low C4a, consistent with our hypothesis. This distinct and sustained separation in benefit was seen by week 6 and was maintained throughout the entire 9-month study. We also assessed clinical outcomes at the evaluable 6- and 9-month time points in the study on a per patient basis. Consistent with what was demonstrated over the course of the trial, 75% of patients with high C4a levels at baseline had an improved clinical function at both the 6-month treatment period and the 9-month follow-up period. It is quite encouraging that twice as many patients with elevated C4a levels at baseline improved as compared to patients with lower C4a levels, consistent with our hypothesis and anti-C1q mechanism of action. Turning to Slide 15. We are equally encouraged by the safety profile of ANX005 as a chronically dosed treatment. ANX005 was generally well tolerated throughout the entire Phase II trial with no change from the interim findings. Among all 28 patients enrolled, the most common treatment emergent adverse event reported was transient first dose infusion-related reactions. There were no deaths and no serious infections observed in the trial. A total of 5 patients discontinued ANX005 treatment, 2 of which were unrelated to drug and 3 were determined to be potentially related to drug. The 3 cases of discontinuation all resolved or improved after treatment discontinuation. Importantly, all 3 of these patients' cases had elevated antinuclear antibody or ANA titers at baseline. No patients who had normal ANA titers at baseline developed SAEs or experienced autoimmune-associated complications. As a result, for other ongoing studies using ANX005, we have incorporated enhanced screening of ANA autoantibody levels at baseline and additional monitoring during treatment. ANX005 has now been evaluated in more than 170 patients across multiple indications and have consistently demonstrated a favorable benefit-risk profile. These positive Phase II trial results with ANX005 are a win for Annexon, for physicians and for patients. Based on these findings, we're planning to engage both U.S. and EU regulatory authorities to assess the opportunity to conduct a well-controlled confirmatory study in Huntington's disease, leveraging optionality for a compelling precision medicine approach. We are well on our way at Annexon to achieving our mission of delivering game-changing treatments to patients living with complement-mediated diseases. This is only made possible by the passion and dedication of a lot of people. To our employees, collaborators and advisers, physicians and their medical teams and most importantly, the patients, families and caregivers who participated in our trial, we sincerely thank you. I'm encouraged and energized by what we have accomplished together thus far and by what the future holds for Annexon. With that, we're happy to take any questions. Operator?

[Operator Instructions] And our first question is coming from the line of Pete Stavropoulos from Cantor Fitzgerald.

Doug and team, congratulations on the data. So -- can you hear me?

Yes.

Yes. So HD patients with higher C4a levels in the CSF demonstrated a rapid clinical benefit as assessed by the UHDRS. So when they entered the 3-month off-treatment period, was there any trends or anything observable that suggests like an increase in complement activation? And can you speculate what happens as you go 6 months post treatment for these patients or a year post-treatment?

So if I understand your question correctly, looking downstream at complement activation, we measured C3a, C3, and they went down with treatment and actually stayed down during the off-treatment period, consistent with the enhanced pharmacokinetics going into the off-treatment period.

Okay.

So we were pleased with the C3 because C3 is expressed by activated toxic A1 astrocytes and the decreasing level suggested to us that, that would be consistent with decreased neuroinflammation.

Okay. And the clinical benefit that was maintained for 3 months off drug for these high complement-activated patients, how does this sort of form a dosing regimen. Can you see a treatment paradigm based on complement activation levels for C4a levels?

Yes. Going forward, we think that this indicates that we can be looking at lower dosing of frequency, probably, or maybe also low dosing levels, but that has yet to be determined how we'll go forward with that. I think the most important thing that we want to do is to have a confirmatory study and then we can explore different doses around that.

Okay. And I know this is initial data, but maybe just your early thoughts on next steps and how to proceed here with additional studies, would you be taking a look and focusing in on patients with high versus low complement activity going forward or stratify accordingly?

Yes, that's right, Pete. So next steps, we'll have discussions with the regulators. We certainly are enthused by the differentiated response we're seeing with the high and low C4 patients. And we'll utilize that in the next study. The question will be whether we do an all-comer study and stratify based on complement levels or if we target specifically a study only in the high treatment group. We'll have those discussions with the regulators, and we'll update thereafter.

Our next question coming from the line of Anupam Rama with JPMorgan.

Maybe I missed this, but can you please describe the cutoff for, say, the levels for complement high group versus complement low group? And then were there any other baseline demographics that correlated with benefit given there were some complement low patient that also looked like they had some benefits, I realize it's small numbers.

Yes, high and low was just based on the median of the values. And there were -- so C4a actually was the best predictor of response. There really weren't any other strong demographic indications.

I guess one thing I'll add to that, Anupam, is that we do know that the 2 treatment groups, the C4a high and low groups were very well balanced at baseline on all the other measures that you would think about, whether it was their baseline TFC or TMS scores, their CAG levels, et cetera, age. So...

Our next question coming from the line of Phil Nadeau with Cowen.

Doug, just to follow on Anupam's question in the comment that you just made. I guess one concern you'd have with a rapid increase in function for the high C4a group was that there's some reversion to mean going on for some reason. They're based on levels of UHDRS or TFC were suppressed and you're just seeing kind of a bounce in that first period. It sounds like there's nothing to indicate that what's happening. Can you maybe speak a little bit more about their baseline characteristics? And is there any reason to believe that, that bounce during the first initial treatment period is not a biologic effect.

Yes, there's no evidence of that, that we've seen thus far. As I said, the 2 groups are definitely well balanced. When you look at their CAG repeats, their age, their CSF, NfL levels and the like. But I don't know, Ted, if you want to expound upon that.

No. The only other thing I would add is if there was such a bounce, it bounced in the way consistent with the mechanism of action because the high C4a levels patients responded very differently than those with low C4a. And also the fact that the complement levels like C3 and C3a went down with treatment, again, is consistent with a decrease in neuroinflammation.

Perfect. And then in terms of the overall study population, it looks like there's actually an improvement from weeks 18 to 24 in both UHDRS and TFC. Any reason to think that there could be delayed activity in the overall group that would take longer for the C4 low patients to respond?

No, thank you for seeing that. We look at that and want to believe that actually. It looks interesting that there could be a delayed response in the low patients. But these are -- it's a really low end number and a small study. So we just think that as encouragement, but we'll have to do that in the confirmatory study to understand that better. But I think it is one reason to be looking at all comers in future trials.

Yes. I'll just add, mechanistically, it makes sense that there may be a delay in the lower C4 group. They have complement activity ongoing, but maybe perhaps not as profound as the high group. And so over time, it may take longer to see an improvement there. The data suggests that again small lands. And so we don't want to make too much of it. But it's certainly something to further explore.

And then on safety, the press release mentions a case of pneumonitis and a case of hemolytic anemia that improved with the suspension of dosing. Is there a reason to believe that either of those are mechanism related? Or do you think that's more a consequence of the disease?

It's harder to know with those 2 indications. But one thing that we do note -- did note is that all 3 were positive for ANA titers, suggesting some underlying autoimmunity happening and that none of the patients with normal ANA developed these AEs.

Yes, I think that's right. And we've seen that consistently across all of the studies where we're dosing ANX005. So it does appear that ANA titers are playing a role with regard to that -- the SAEs we saw in this particular study, but we'll continue to further monitor this.

Our next question coming from the line of Joseph Stringer with Needham & Company.

Two from us. One, do you -- can you comment on UHDRS versus TFC in terms of how sensitive do you think a potential decline in either of those is to sort of PK levels of the drug, meaning -- it looks like you're seeing a potentially more rapid change in TFC with the caveat that it looks like there's still drug levels around during the off-treatment period. So is TFC generally considered more sensitive once you get less target engagement. And then I suppose in terms of the off-treatment period, do you think that in a confirmatory trial or additional trial, you would want to see -- sort of extend the off-treatment duration, would you expect both the UHDRS and TFC, how long do you think you would need to see a decline in those once drug -- PK levels of the drug sort of dropped to 0, the drug was washed out. Would you want to see just for confirmation to see both UHDRS and TFC decline?

All right. So to your first question, the UHDRS is a composite score with 4 different domains, one of which is the TFC, and these 4 domains were -- they're both cognitive and motor measures, and they were chosen to create the most sensitive composite scale that they could. And so I would say that the UHDRS is more sensitive than TFC overall. TFC is the endpoint used in the U.S. by the FDA and the composite score is used in the European Union, which is why we put both of those out there. As to your second question, yes, I think that once the drug goes away and complement activity returns, I think you would begin to expect to see a decline. Whether or not the separation would remain? It could be, it could be both groups would then decline according to their original rates, and you continue to see a separation. But again, that really is just speculation at this point.

Got it. And then a follow-up question on the ANA screening and potential exclusion of those patients. How much would that -- could that limit you, if at all, in terms of next steps in development. Would that -- do you believe that, that comprises of a significant number of patients going forward? Or is it -- do you think it'd be relatively small number of patients?

Yes. We know it's quite manageable, Joey. It's about 20% of the population. And in fact, we've already modified protocols for other ongoing ANX005 studies. And so we're managing this real time. It's not been a topic much at all. So it's certainly a mechanism we can deploy clinically in further evaluating of the drug. All right. I think that's our last -- go ahead, please, operator.

I'm showing no further questions, sir. I'll turn it back over to you.

Yes. Well, there you go. So that's our last question. We want to thank all of you for joining us today. We're super encouraged and proud of the progress we've made thus far with ANX005, and we'll continue to share updates with you as we advance the Huntington's disease and other programs. We appreciate all of your continued support of Annexon and look forward to speaking with you in the weeks and months ahead. Thank you very much.

Ladies and gentlemen, that does conclude our conference for today. Thank you for your participation. You may now disconnect.