Cognition Therapeutics (CGTX) 2026年第三季度业绩电话会：DLB III期计划推进

核心要点

• Cognition Therapeutics 表示，已就 zervimesine（代号 CT1812）用于治疗轻中度路易体痴呆（DLB）拟定开展的 III 期注册临床试验设计与美国食品药品监督管理局（FDA）达成一致。
• 该拟定试验为期九个月，将比较每日口服一次 100 毫克 zervimesine 与安慰剂的效果。在完成临床方案制定并获得充足资金的前提下，管理层计划于 2027 年年中启动该研究。
• 在与 FDA 进行沟通后，该公司预计注册临床试验所需受试者人数将不到此前设计中约 750 人的一半。原先的设计预计耗资约 1.5 亿美元。
• 截至 2026 年 9 月 30 日，现金、现金等价物及受限现金等价物总计约 3150 万美元。管理层预计，现有资源可支持运营及资本支出至 2027 年第四季度。
• 第三季度净亏损由上年同期的 490 万美元（即每股基本及摊薄亏损 0.06 美元）收窄至 410 万美元（即每股基本及摊薄亏损 0.04 美元）。
• 共有 545 名轻度认知障碍和早期阿尔茨海默病患者参与的 II 期 START 研究已完成全部入组，且获得了美国国家老龄化研究所（NIA）的非稀释性资金支持。管理层预计将在 2027 年下半年公布主要（top-line）结果。

关键财务数据

业务与运营表现

Zervimesine 是一种小分子药物，旨在穿过血脑屏障并与 sigma-2 受体结合，以保护神经元免受毒性蛋白侵害。管理层表示，该药物不与斑块结合，也不清除斑块，因此预计不会带来淀粉样蛋白相关成像异常（ARIA）风险。

拟定的 DLB III 期试验将招募轻中度患者，并在基线时重点富集有幻觉和妄想症状的患者。主要终点将侧重于神经精神科问卷（NPI-12）中的幻觉和妄想部分。阿尔茨海默病合作研究日常生活能力量表（ADCS-ADL）计划作为关键次要终点。

该设计基于 II 期 SHIMMER 研究，该研究达到了主要的安全性和耐受性目标。管理层表示，zervimesine 在运动症状、认知波动和神经精神症状方面均显示出积极的改善趋势。该公司称，在 SHIMMER 研究中，精神病性症状的进展减缓了 89%。

Cognition Therapeutics 正将该项目设计为单一关键性试验。FDA 已同意考虑这一方案，但任何最终决定都将取决于其对试验完成后的临床数据审查。

该公司还在推进包含 545 名轻度认知障碍和早期阿尔茨海默病患者的 II 期 START 研究。该研究已完成全部入组，并得到 NIA 的资金支持。最后一名受试者已于 2025 年 11 月完成入组。

管理层还指出，最近获得授权的一项涵盖 zervimesine 新型结晶制剂的美国专利将提供保护直至 2045 年。

管理层展望

管理层计划敲定 DLB III 期试验方案并提交 FDA 审查。与此同时，公司正在评估合同研究组织（CRO）、首批临床试验中心，以及涵盖投资者、银行家和潜在战略来源的融资方案。

公司的目标是在 2027 年年中启动 DLB III 期研究。根据目前的支出水平，管理层预计现有现金可支持运营及资本支出至 2027 年第四季度。

管理层表示，针对阿尔茨海默病的 START 研究的主要（top-line）结果预计将于 2027 年下半年公布。

风险与关注要点

• 计划中的 DLB III 期研究需要额外资金，公司目前仍在评估融资与战略方案。
• 尽管 FDA 会考虑单一关键性试验的方案，但能否被认可仍取决于对完成后的研究数据的审查。
• DLB 缺乏用于衡量疾病阶段和治疗效果的成熟影像学与血液生物标志物。因此，III 期试验设计采用精神病性症状的变化作为潜在疾病进展的替代指标。
• DLB 的认知测试可能会因患者在短时间内及日复一日间注意力和认知的波动而变得复杂。
• 试验的启动仍取决于临床方案的敲定、FDA 的审查、供应商的选择、试验中心的准备以及融资情况。

分析师问答精选

管理层表示，早先的 DLB 试验设想涉及约 750 名受试者，耗资将达约 1.5 亿美元。在与 FDA 进行讨论并进一步聚焦于精神病性症状之后，公司认为受试者人数不到原先一半便能开展一项统计效力充足的注册临床研究。

对于 START 研究，管理层表示尚未发布或分析用于与此前 SHINE 研究进行对比的基线数据。公司计划在即将召开的阿尔茨海默病临床试验（CTAD）会议上讨论招募方法，包括生物标志物的使用。

谈及精神病性症状的临床相关性时，管理层表示，近 80% 的 DLB 患者会出现幻觉和妄想，且通常随疾病进展而恶化。所选评估工具不局限于特定类型的幻觉或妄想，可普遍适用于受试者。

管理层还将安全性描述为区别于目前用于控制 DLB 症状的抗精神病药物的潜在差异化优势。公司强调，zervimesine 并非抗精神病药，其研发目标是延缓潜在疾病进展，而不仅仅是减轻急性症状。

业绩电话会完整纪要

完整财报电话会议逐字稿

管理层陈述

Operator

My name is Mark, and I will be your conference operator today. This call is being recorded. I would like to turn the presentation over to your host for today's call, Mike Mauger, Financing Director for LifeSciAdvisors. proceed, Mr. Moyer.

Mike Moyer

Thank you, operator, and good morning, everyone. Welcome to Cognition Therapeutics' Third Quarter 2026 Conference Call. With me today are Lisa Ricciardi, President and Chief Executive Officer; John Doyle, Chief Financial Officer; and Dr. Tony Caggiano, Chief Medical Officer. This morning, the company issued a press release detailing its financial results for the quarter.

We encourage everyone to read this morning's press release as well as cognitions Form 10-Q, which is now filed with the SEC and available on our website. In addition, this conference call is being webcast through the company's website and movie archived for 30 days. Please note that certain information discussed on the call today is covered under the safe harbor provisions of the Private Securities Litigation Reform Act.

We caution listeners that during this call, management will be making forward-looking statements. Actual results could differ materially from those stated or implied by these forward-looking statements due to the risks and uncertainties associated with the company's business. These forward-looking statements are qualified by the cautionary statements contained in cognitions press releases and SEC filings. This conference call contains time-sensitive information that is accurate only as of the date of this live broadcast.

Cognition undertakes no obligation to revise or update any forward-looking statements to reflect events or circumstances after the date of this conference call. With that, I would now like to turn the call over to Lisa Ricciardi.

Lisa Ricciardi

Mike, thank you, and good morning, everyone. As you can see on the agenda, first topic we'd like to cover on this call is the registrational plan for zervimesine , or CT1812 and the treatment of people with dementia with Lewy bodies, which we'll refer to in this call as DLB. Two weeks ago, we announced that we reached alignment with the FDA on a planned registrational trial for zervimesine and people living with DLB Investors who have been following Cognition are aware that this topic has been the subject of several meetings with the FDA.

We conducted meetings in January, May and October. And all along, we've encountered FDA staff at every step who are engaged and constructive. And as a result of our interactions, Cognition Therapeutics is the first company to align on a Phase III trial with the FDA for this class of drugs to treat DLB, to treat the condition of DLB. Other companies have designed Phase III trials for symptomatic treatments such as antipsychotics or medications for agitation in this population. Zervimesine is unique in that it is designed to slow the course of the disease, not just treat the symptoms. I am proud that we have tackled an unchartered development path together with the FDA.

Now we're working to complete a protocol for zervimesine Phase III program. A complicating factor with DLB is the lack of biomarkers available to measure the progression of disease. Because of this, we and the FDA agreed on clinical endpoints that are aimed at showing changes in disease progression. And with that, let me introduce our Chief Medical Officer, Tony Caggiano, who can discuss these endpoints and the details of our trial. Tony?

Anthony Caggiano

Thanks, Lisa. Cognition's Founder, Dr. Susan Catalano, was strategically focused on small molecules that would protect neurons while efficiently crossing blood-brain barrier. As a small molecule, zervimesine has no need for brain-shuttle technology to access the CNS. It binds to the sigma-2 receptor and protects neurons from toxic proteins. It does not bind to or remove plaque so we would not expect ARIA to be at risk with zervimesine.

And finally, we have an extensive patent estate on zervimesine. The U.S. CTO recently issued a patent on a new crystal formulation of zervimesine, which provides protection through 2045. Here, you can see the proposed study schematic for our Phase III trial of zervimesine in mild-to-moderate DLB. Participants will be randomized to receive either 100 milligrams of zervimesine or placebo orally once a day for 9 months. Many of the endpoints will look familiar to you as they were employed in the Phase II SHIMMER study. The SHIMMER study met its primary goal of safety and tolerability. Zervimesine treatment in SHIMMER was shown to have a meaningful impact across the core features of DLB, including motor symptoms, fluctuations of cognition and neuropsychiatric symptoms. In brief, everything was moving in the right direction.

Neuropsychiatric symptoms in SHIMMER were evaluated using the NPI-12, a tool that measures 12 behavioral symptoms, including psychosis, which is defined as hallucinations and delusions. Other behaviors, including agitation, anxiety and apathy are also reported in the NPI-12. When we focus on just the psychosis systems, progression was slowed by 89% with zervimesine treatment. Many of the symptoms we measured at SHIMMER are core features required for diagnosis of DLB. The fact that zervimesine treatment impacted all of these symptom domains confirms that it is not asymptomatic drug. By protecting neurons, we believe zervimesine is slowing progression of the underlying disease or it is a disease-modifying agent. The measurement of disease progression DLB is challenging.

Alzheimer's disease has imaging and blood test designed to determine disease stage and treatment effects. None of these are available yet in DLB. During our meetings with the FDA, we discussed options for measuring changes in DLB disease progression. Both FDA divisions of neurology and psychiatry agreed that changes in psychosis may serve as a proxy for changes in the underlying disease. Today, clinicians rely on changes in key symptoms to monitor the patient's disease. However, not all symptoms are easily evaluated in DLB. For example, measuring cognition is complicated by the fluctuations that people and DLB experience. These fluctuations of attention and cognition can change for a minute to minute and day-to-day. So a cognitive test taken Tuesday at 3 p.m., for example, may have a very different result from the same test taken Thursday at 10 a.m. This makes it difficult to reliably test cognition and clinical trial participants.

Interestingly, psychosis affects up to 80% of people with DLB. It is often present in early stages before the other hallmark symptoms emerge, and it worsens in a predictable fashion as the disease progresses. Psychosis is easily measurable and maybe a proxy for changes in overall disease progression. The FDA agreed with its assessment, and we are designing a Phase III study with this in mind. The 2 end points we discussed with FDA are shown on this slide. Within the NPI-12, 2 of the 12 symptoms are hallucinations and delusions. This tool is called the NCI in measures the frequency and severity of 2 symptoms. The primary outcome measure in the Phase III trial is the NPI. This assessment tool is well known to investigators and has been used extensively in clinical trials. When we discussed the summer data with the FDA, we identified that hallucinations and delusions, the lutein dilution results were some of the strongest from the study. The NPI which analyze these data has some excellent signal-to-noise ratio, suggesting it will be a powerful tool treatment differences in our planned registrational trial.

In addition, enrollment will be enriched with patients who have hallucinations and delusions at baseline. Together, these features give us confidence that the trial we have designed will be adequately powered and sufficiently controlled with a moderate sample size. In addition to the NKI2, we are planning to include the ADCS-ADL for Alzheimer's Disease Cooperative study activities of daily living scale as a key secondary endpoint. By using these 2 together, we will show how zervimesine treatment impact the progression of the disease. Before we move on, I want to make an important distinction. -- even machine is not an antipsychotic. Antipsychotics are effective on reducing acute symptoms of those inflammation or delusions. Antipsychotics are not, however, effective as slowing in the course of the disease. As they're anything as the disease-modifying agents, we had approached the FDA to use a composite endpoint to assess the diverse symptomatology of DLB.

In meeting with FDA, it became clear that there is no validated measure developed to assess its diverse symptomatology. FDA cognition agreed that cytosis could be used as an indicator of the overall disease progression. I'll turn the call back over to Lisa now, who can provide updates on the next steps for both DLB and Alzheimer's program.

Lisa Ricciardi

Thank you, Tony. Our immediate next step is finalization of the protocol, which will be submitted to the FDA for review. Consistent with our discussions with the FDA, our plan is to design a single pivotal study for zervemecine in DLB. We've had several conversations with the FDA, and they have agreed to consider a single pivotal trial. Their decision will be a matter of review upon receipt of our completed clinical trial data. Having just concluded our discussions with the FDA and having just received their minutes, we're now evaluating our options with investors, bankers and strategic to secure sufficient study funding to conduct the study.

And in parallel, we're evaluating vendors for CRO partners as well as the look at initially clinical sites for the trial. Our goal remains to initiate this study in mid-2027. Now in addition to our work on DLB, we're also approaching an important milestone in our START study. The START study an ongoing trial is a Phase II study in mild cognitive impairment and early Alzheimer's disease. This study has 545 patients in it. As a reminder, START is fully enrolled, fully funded with nondilutive capital from the NIA. The last participant in start was enrolled in November of 2025. thus, we're expecting top line results in the second half of 2017. With that, John Doyle will review our financial results and provide a bit more color around our cash position. John?

John Doyle

Thank you, Lisa. Cash, cash equivalents and restricted cash equivalents as of September 30, 2026, were approximately $31.5 million. Total obligated grant funds remaining from the NIA were $18.2 million. R&D expenses were $5.1 million for the third quarter of '26 compared to $3.8 million for the comparable period in '25. The increase in research and development expenses was driven by higher trial activities with contract research organizations and professional fees. General and administrative expenses were $2.3 million for the quarter compared to $2.6 million in 2025. The decrease in general administrative expenses was driven primarily by lower professional fees and compensation.

The company reported a net loss of $4.1 million or $0.04 per basic and diluted share for the third quarter compared to $4.9 million or $0.06 per basic and diluted share for the same period in 2025. Based on our current spend, we estimate that we have sufficient cash to fund operations and capital expenditures into the fourth quarter of 2027. Lisa?

Lisa Ricciardi

Thank you, John. Operator, please open the call to questions.

Operator

[Operator Instructions]

And your first question comes from the line of William Wood with B. Riley Securities.

分析师问答

William Wood

So 2 for us if we can with your current funds submission into first quarter 2027, could you provide -- or fourth quarter, sorry? Could you provide any extra details on the cost of your DLB trial, maybe you're thinking on advancing that forward where the funds might come from? And then also in terms of with the START trial moving quickly towards your readout in the third quarter, could you provide any color on the baseline there, how it may compare to Phase 2 shines load pause population analysis?

Unknown Executive

Yes. Thanks for the questions. Regarding the DLB study, we have been considering trials of about 750 participants when we initially planned the study and this was targeting symptoms that include cognition and fluctuation. Trials of this size would have been about $150 million. Given our discussions recently with the FDA and clear focus on DLB and their agreement for us to use we can come back to a well-powered registrational trial with less than half of this original number of participants. Regarding the START trial, we have not released or even analyzed the baseline data in order to compare it to the SHINE study. There will be a detention at the CTAD meeting coming up in 2 weeks.

Looking at how we recruited individuals in this study, which is slightly different than other studies had to the use of biomarkers and so forth, but [indiscernible] data will not be presented yet. Thank you very much.

Operator

Next question comes from the line of Ram Silverado with H.C.Wainwright.

Raghuram Selvaraju

Congratulations on all the recent progress, particularly on the FDA front. I was wondering if you could give us some additional qualitative color around the nature of hallucinations and delusions suffered typically by DLB patients and how these worsen, along with the kind of inevitable progression of the disease as well as what you anticipate would be considered a clinically meaningful impact on that specific arena of DLB symptomatology.

Anthony Caggiano

Yes, thanks for those questions. So people at DLB have hostesses and dilution, then the interesting thing is that Nearly 80% of individuals have hallucinations and delusions. Now the type of hallucinations and delusions in this disease is a little different than other conditions. But in reality, that's not particularly important. We certain types of [indiscernible] in that in character are somewhat different than you might expect in schizophrenia or Alzheimer's disease. Now the nice thing with the tool that we're using it's not restricted based on a particular type of hallucination or dilution. It's general and allows it to be really crafted towards each and every participant.

So indeed, people with DLB, their hallucinations and delusions worsen throughout the course of the disease. And the important thing that we did over the last year here is to work with the FDA to basically agree that whether you're looking at times or fluctuations or motor center, in this case, the neuropsychiatric symptoms they all could be a reasonable proxy for how the disease is progressing. And because we look at hallucinations and dilutions over the last months, if that subject to the issues of fluctuations that these other measures are, so it's a very good clinical tool to use. So that's why we're using this as a general measure of disease progression.

Lisa Ricciardi

Ram, I can add that we have done a great deal of work with people who live with DLB. We don't refer to them as patients because we're not their doctors and what you hear from these individuals and hallucinations of several nature. One of the things they talk a lot is about smell. I'm selling bird coast. That seemed to be a common thing, surprisingly other hallucinations our oral AURA, why is the American -- the [indiscernible] banner playing all day long. There are visual hallucinations. I've heard repeatedly from physicians about patient people with DLB commenting on trains running through the back there are children coming to the home for things of this nature. Now they're largely not bothered by these things. but it does take them a while to come forward and talk about it.

On the dilution side, I think it's pretty straight forward what's most common [indiscernible] syndrome, so you're looking across the table at your husband or your wife and you're going, you are not my husband or my wife. Where is that person, you actually see correctly something is there, but the brain interprets it as something else. So at a fundamental level, these are sort of definitional characterizations of hallucinations and delusions from patients as well as from the people who treat them. It's pretty consistent.

Raghuram Selvaraju

And then just with respect to the risk-benefit profile that you expect to target with zervimesine . Lisa, you mentioned during your prepared remarks that atypical antipsychotics has historically been used in managing the symptoms of DLB patients. But of course, that group of drugs is well known to have well-documented chronic side effects that can be potentially very debilitating. These are not side effects that are associated with zervimesine or the class of drug towards urban belongs. So would you deem that to be potentially a future meaningful clinical and competitive edge for zervimisine were it to be approved in DLB?

Lisa Ricciardi

Excellent question. The answer is very much yes. We've done a fair amount of research with prescribing doctors who treat this population. And they say pretty much what you said, Rob, they go, "Look, I know I have a lousy hand of cards here, if you will. -- there are drugs that have warning black box warnings. There are drugs, let's say, there are fatalities in DLB patients. And so when they talk about how they treat these patients, they carefully weigh where they want to take the risk, but they're aware of it. And so we believe that coming forward with a drug that can treat this population and have a safety profile that's been demonstrated in this population as well as what Tony indicated earlier, a very unlikely incidence of any ARIA at all.

We feel that, that will be an important differentiator. There really are not drugs for this patient population. You open the medicine chest and you find things from other conditions and you apply them to the symptoms, and that's about as good as it gets. And the language I've articulated here is not different from the caregivers or the physicians to think about it in the same way. So I do think what you've expressed is a very important potential point of differentiation that we hope to demonstrate in our registrational trial.

Operator

Thank you. There's no further question at this time. And now I will turn the call back to Cognitions's CEO, for closing remarks. Ms. Ricciardi, please go ahead.

Lisa Ricciardi

Thank you. In closing, I want to reiterate how excited we are to be able to advance zervimesine into a Phase III study. We plan to evaluate the potential to slow the progression of this disease. The work we undertook with the FDA's office of Neuro Science concluded with the regulatory path for revenue. But you know what, we believe we're paving the way for future innovations in this disease area.

We want to thank the trial participants and their families as well as our advisers clinical trial investigators and our staff. Thank you for joining us. I'll turn the call back to the operator to conclude our call.

Operator

That concludes today's call. You may now disconnect.