▎药明康德

Howard Fillit博士是国际知名的老年医学专家、神经科学家及阿尔茨海默病研究领域的资深学者。作为ADDF的联合创始人兼首席科学官，他长期致力于推动神经退行性疾病的创新疗法研发，在老年医学与认知障碍领域具有深远影响力。Fillit博士拥有逾40年的学术与临床经验，发表了300余篇学术论著，其学术贡献获得业界广泛认可。

您好，感谢您接受我们的访谈。您亲历了阿尔茨海默病研究领域从数十年踯躅不前到如今重焕生机的完整历程。在您看来，当前我们应对该疾病的策略发生了哪些根本性转变？

Howard Fillit博士：我们现在所看到的正是药物研发的标准路径——始于基础科学发现，最终转化为临床效益。就阿尔茨海默病而言，这个过程始于一个多世纪前：1906年Alois Alzheimer医生在对早老性痴呆患者进行尸检时首次描述了斑块和神经原纤维缠结。但直到1970年代，阿尔茨海默病才被认为是老年痴呆症最常见的病因，而真正有意义的研究直到1980年代才起步，可以说我们完全是从零开始。这正是早期研究举步维艰的主要原因之一。

直至1984年，β-淀粉样蛋白才被确认为脑内斑块的核心成分之一，这一发现不仅为我们提供了首个AD分子靶点，更掀起了全球范围内的研究热潮。自1985年起，整个研究领域几乎将全部精力都聚焦于淀粉样蛋白，斥资数十亿美元用于阐明其致病机制、开发检测方法并最终实现靶向干预。

近二十年来，我们首次迎来FDA批准的疾病修饰疗法——Leqembi和Kisunla，这标志着AD治疗领域的历史性突破。这些进展得益于生物标志物和诊断工具开发领域的显著进步，堪称一场真正的医学革命。那么，究竟是什么发生了根本性改变？我们终于将人类病理学发现、靶点验证、诊断技术和临床干预连成了一个完整的闭环。

ADDF采取了哪些独特策略来降低研发风险和催化重大突破？其他生物科技公司或医药企业可以从贵机构的模式中获得哪些启示？

Howard Fillit博士：我们建立ADDF的使命只有一个：加速阿尔茨海默病药物研发。这包含三大支柱——加速生物标志物开发、投资并开展早期治疗手段研发、推动预防医学研究。作为一家公益创投机构，我们不资助基础研究，也不参与倡导工作，而是将所有资源集中于转化医学研究，这正是我们区别于多数基金会的核心特征。

我们通过可转换债券、股权认证及风险投资式优先股权等方式，已完成对150余家早期生物科技企业的投资。关键在于，我们的资金严格绑定具体科研项目而非企业日常运营开支。企业必须向我们提交清晰的工作计划、详细预算及明确的里程碑节点。

此外，我们还提供真正的专业支持。ADDF拥有十余位专职神经科学家——其中多数具备深厚的产业界经验。当我们投资一家企业时，我们不仅会评估他们的提案，还会帮助他们改进提案。许多阿尔茨海默病初创团队都来自学术界，他们往往缺乏中枢神经系统（CNS）药物开发的实战经验，无论是在临床还是临床前阶段。例如，他们未必总是清楚特定研发项目中如何选择最适用的动物模型。我们的专家团队会全程指导——从选择最佳动物模型到考虑监管策略——确保项目以最优配置向前推进。

另一个关键要素在于，作为公益创投机构，我们专注于长期投入。一般风投基金的投资期限可能只有10年，而我们能陪伴项目走得更远。有一家生物科技企业与我们持续合作近二十年，在我们的帮助下，如今他们正努力为一个3期临床项目筹集资金——这家企业的发展历程尤其令我们振奋。

我们此前讨论过CNS药物研发史上居高不下的失败率。根据您的经验，这个领域至今仍存在哪些重大认知误区？我们该如何破除这些误区以提振行业信心与投资热情？

Howard Fillit博士：其中一个重大问题在于历史积淀不足——我们缺乏像肿瘤学领域数十年积累的临床试验体系。从1980年代到21世纪初，我们几乎是从零开始学习有关阿尔茨海默病的一切。

2010年前后有一个典型案例：多家企业都在测试抗β-淀粉样蛋白单抗药物。某大型3期试验中，一家公司使用正电子发射断层扫描（PET）示踪剂检测大脑中的β-淀粉样蛋白并进行了亚组分析，结果令人震惊——约30%受试者的淀粉样蛋白扫描呈阴性。这意味着他们脑内根本没有该药物的靶点蛋白。更有可能的是，这些被领域专家纳入试验的痴呆患者很可能根本未患上阿尔茨海默病。

这给我们敲响了警钟。它表明主观诊断不够准确，也指明了试验失败的一个关键原因。解决方案是采用基于生物标志物入组标准，使临床试验筛选变得更加客观。ADDF支持了淀粉样蛋白PET成像技术的早期开发，该技术应用后，临床试验的严谨性和效率显著提升。

如今，血液生物标志物正推动新一轮变革，这将使试验设计更加精准和可扩展。结合认知评估与tau蛋白成像技术的进步，我们已掌握开展严谨、可靠且高效的阿尔茨海默病临床试验的方法论。无论药物最终成败，至少我们能确保试验本身科学可靠——这与十年前相比已是天壤之别。

图片来源：123RF

随着基于血液的生物标志物和神经影像学的进步，我们离实现阿尔茨海默病的实时精准医疗还有多远？

Howard Fillit博士：我认为我们已处于一个重要的拐点。阿尔茨海默病的血液检测首次成为现实，而且已有多种检测方法进入市场。它们采用不同的生物标志物和技术。最具前景的靶点是p-tau217，其与β-淀粉样蛋白斑块密切相关；其他方法还包括Aβ42/40比值检测、磷酸化tau/非磷酸化蛋白比例分析等。现有的检测方法五花八门，新技术仍在不断涌现。

阿尔茨海默病的诊疗进展始于创新，成于协作。通过"诊断加速器"计划——一项由Leonard Lauder、Bill Gates、Jeff Bezos等慈善家共同出资1亿美元发起的全球生物标志物倡议——我们已在全球支持70余个生物标志物项目。这些项目不仅涵盖体液标志物，还包括用于改善和简化认知评估的数字生物标记物，它们为AD的早期干预、更具包容性的临床试验以及未来个体化的预防性AD护理铺平了道路。我们正迈入阿尔茨海默病诊断的变革时代，这着实是一个激动人心的里程碑。

预防医学研究是ADDF的三大支柱之一。在预防AD方面，您认为哪个方向最有前景？

Howard Fillit博士：最令人振奋的工作之一当属里程碑式的FINGER研究——这是首个评估生活方式干预能否预防阿尔茨海默病的随机临床试验。其研究结果令人信服：约40%的AD病例或可预防。该研究采用五大干预方案：积极的社会互动、持续的职业活动、严格管控糖尿病与高血压、规律运动及地中海饮食。简言之，它将行之有效的心血管疾病预防策略应用于脑健康领域，同时强调认知刺激——本质上，这是一种"用进废退"的脑力维持模式。

现在我们正推进ADDF资助的FINGER 2.0研究。正如心脏病学从单纯生活方式干预发展到联合他汀类药物，我们正在干预方案中增加药物成分。这项研究选用二甲双胍，一种具有多重神经保护潜力的抗衰老候选药物。目标是通过生活方式、共病管理与药物三联方案，验证其能否更有效延缓早期阿尔茨海默病（包括轻度认知障碍）的进展。这标志着预防医学的新疆界，我们对研究前景充满期待。

图片来源：123RF

感谢您的真知灼见！最后，您认为阿尔茨海默病领域研发的未来会如何？

Howard Fillit博士：我相信在不久的将来，我们将拥有血液检测等实用工具，来识别处于疾病极早期阶段的患者。在此基础上，我们将向精准医疗模式迈进——利用生物标记物，以衰老生物学为基础，根据患者独特的神经退行性病变通路对其进行表型分析。这将催生个体化、多模式的治疗方案：针对每位患者的生物学特征定制联合疗法。

我认为，未来五年内，我们有望见证首批超越现有治疗标准的联合疗法的诞生。届时，认知功能衰退的延缓率可能从当前的30%提升至50%甚至60%。同样重要的是，我们将获得"阿尔茨海默病可预防"的概念验证证据。其影响将极为深远：如果我们能将AD的发病时间或病程进展推迟五年，患者数量便可减半。这一目标不仅切实可行，更令人无比振奋。

Renewed Optimism on Alzheimer’s Disease: A Conversation with Dr. Howard Fillit, Co-Founder and Chief Science Officer, Alzheimer’s Drug Discovery Foundation

Thanks for joining us, Howard. You’ve seen the Alzheimer’s field evolves—from decades of frustration to a moment of renewed optimism. What do you think has fundamentally shifted in our approach to the disease?

Howard Fillit: That’s a great question. What we’re seeing now is the usual course of drug development—starting with foundational science and ultimately translating into clinical impact. In Alzheimer’s, that arc began more than a century ago, when Alois Alzheimer first described plaques and tangles during the autopsy of a patient with presenile dementia in 1906. But it wasn’t until the 1970s that Alzheimer’s was even recognized as the most common cause of dementia in old age, and meaningful research didn’t begin until the 1980s. So, we started from scratch—literally zero. That’s a key reason for the frustration.

It wasn’t until 1984 that beta-amyloid was identified as a core component of the plaques, which gave us our first molecular target and catalyzed a massive research effort. From around 1985 onward, we’ve had an intense, nearly singular focus on amyloid, with billions of dollars invested in understanding it, detecting it, and ultimately targeting it.

For the first time in nearly two decades, we have disease-modifying therapies approved by the FDA— Leqembi and Kisunla — marking a historic breakthrough in Alzheimer’s treatment. These advances are possible because of remarkable progress in biomarkers and diagnostic tools. It's really a revolution. So, what’s fundamentally changed? We’ve finally connected the dots between human pathology, target engagement, diagnostics, and clinical intervention.

At the ADDF, how do you uniquely reduce risk and catalyze breakthroughs? And what lessons can others in biotech or pharma take from your approach?

Howard Fillit: We built the ADDF with a single mission: to accelerate drug development for Alzheimer’s. That included three pillars—acceleration of the development of biomarkers, developing and investing in early-stage treatments, and prevention. As a venture philanthropy, we don’t fund basic research, we don’t do advocacy—we put all our resources into translational science. That's one thing that makes our model different from many other foundations.

We’ve made over 150 investments in early-stage biotech companies, using convertible notes and warrants and we also use VC-style Preferred Equity. The difference is that our funds are tied strictly to a specific scientific project, not general expenses. Companies have to come to us with a clear work plan, a detailed budget, and defined milestones.

We also bring real expertise. We have more than 10 neuroscientists on staff—most with deep industry expertise. When we fund a company, we don’t just evaluate their proposal—we help improve it. A lot of early-stage Alzheimer’s startups come from academia and don’t have experience navigating the nuances of CNS drug development, either clinically or preclinically. For example, they don’t always know which animal models are the best ones to employ in that particular drug development program. Our team helps guide them—whether it’s choosing the right animal model or thinking through regulatory strategy—so the programs are in the best position to move forward.

Another key is that, as a venture philanthropy, we're in it for the long run. Whereas a venture capital fund might be limited to 10 years, we can stay in programs much longer. There is one biotech that we've been working with for almost 20 years. And we've helped them to get to the point where they're trying to raise funding now for a phase III program. And we're very excited about that particular company.

We’ve talked about the historically high failure rate in CNS drug development. From your experience, what are some of the biggest misconceptions that still haunt the field—and how can we overcome them to build more confidence and investment?

Howard Fillit: One big issue has been the lack of history—no decades of trial infrastructure like we’ve had in oncology. In the 1980s through the 2000s, we learned everything from scratch about Alzheimer’s.

An example came around 2010, when companies were testing monoclonal antibodies against beta-amyloid. In one major Phase III trial, a company used a PET tracer for beta-amyloid in the brain and did a sub-study. What they found was striking: about 30% of trial participants had negative amyloid scans. So, they didn't have amyloids in the brain, which was the target of the drug they were testing. And more than likely, they didn't have Alzheimer's disease, even though they had dementia and were enrolled by experts in the field.

That was a major wake-up call. It showed that subjective diagnoses weren’t good enough, and it highlighted a key reason why trials were failing. The solution came with biomarker-based enrollment, so the criteria for clinical trials became more objective. At the ADDF, we helped support the early development of amyloid PET imaging, and once that was implemented, the rigor and the efficiency of clinical trials improved dramatically.

Today, we’re seeing the next leap forward with blood-based biomarkers, which will make trials even more precise and scalable. Combined with advances in cognitive assessments and tau imaging, we now know how to run rigorous, robust, and efficient Alzheimer’s trials. Whether a drug succeeds or not, at least we know the trial was done right—and that’s a huge shift from where we were even a decade ago.

图片来源：123RF

With the advances in blood-based biomarkers and neuroimaging, how close are we to achieving real-time precision medicine in Alzheimer’s?

Howard Fillit: That’s a great question. I think we’re already at a major inflection point. For the first time, we now have blood tests for Alzheimer’s disease—and several are on the market. They use different biomarkers and technologies. One of the most promising is p-tau217, which correlates strongly with beta-amyloid plaques. Others include the Aβ42/40 ratio, or the percentage of phosphorylated vs. unphosphorylated tau. So, there are different methods, and new ones are emerging.

Progress in Alzheimer’s starts with innovation and is made possible through collaboration. With the Diagnostics Accelerator (DxA), a $100 million global biomarker initiative—funded by leading philanthropists like Leonard Lauder, Bill Gates, Jeff Bezos and more—the DxA has supported more than 70 different biomarker programs worldwide. These include not only fluid-based markers but also digital biomarkers to improve and simplify cognitive assessment that pave the way for earlier intervention, more inclusive clinical trials, and a future of personalized, preventive Alzheimer’s care. We're entering a transformative era for Alzheimer’s diagnostics, which is an exciting milestone.

Prevention is one of the three pillars of the ADDF. In terms of prevention of Alzheimer’s, which direction do you find most promising?

Howard Fillit: One of the most exciting efforts is the landmark FINGER trial—the first randomized clinical trial to evaluate whether Alzheimer’s can be prevented through lifestyle interventions. The findings are compelling, suggesting that up to 40% of cases may be preventable. It uses a five-pronged approach: social engagement, continued occupational activity, rigorous management of diabetes and hypertension, regular exercise, and a Mediterranean-style diet. In short, it applies proven heart disease prevention strategies to brain health, with the added emphasis on cognitive stimulation—essentially, a “use it or lose it” model for maintaining mental vitality.

Now we’re moving into what we call FINGER 2.0, an ADDF-funded study. Just as cardiology evolved from lifestyle changes alone to combining them with drugs like statins, we’re adding pharmacological interventions to the mix. In this case, we’re testing metformin—one of the leading anti-aging candidates with multiple potential neuroprotective effects. The idea is to combine lifestyle and comorbidity management with a drug to see if we can better slow progression in early-stage Alzheimer’s, including those with mild cognitive impairment. It’s a new frontier in prevention, and we're hopeful about where it’s headed.

图片来源：123RF

Thank you for your insights! To conclude, if you could make one bold prediction, what do you see as the future of Alzheimer’s research and development?

Howard Fillit: I believe we’ll soon have practical tools—like blood tests—to identify patients very early in the disease. From there, we’ll move toward a precision medicine model: using biomarkers to phenotype patients based on their unique neurodegenerative pathways, grounded in the biology of aging. This will enable a personalized, multimodal approach to treatment—combining therapies tailored to each individual’s biology.

Within the next five years, I think we’ll see the first examples of combination therapies that move beyond our current standard. Instead of a 30% slowing in cognitive decline, we could be looking at 50% or even 60%. Just as important, I believe we’ll have proof of concept showing prevention is possible. And here’s the impact: if we can delay the onset or slow the progression of Alzheimer’s by just five years, we could cut the number of cases in half. That’s achievable and incredibly exciting.