A few weeks ago, we took a look at how artificial intelligence (AI) fared. a transformative impact on cancer research. Today, we take a look at how AI tools are helping to tackle another formidable challenge in modern biotechnology, Alzheimer’s disease.

“Alzheimer’s is an incredibly complex disease with many factors at play. You have accumulation of amyloid plaques, your entanglementsand loss of connections between nerve cells. If you put yourself in the shoes of a drug manufacturer, how do you choose which one to target and how do you determine the sequence of events in a certain way?” said Nadia Harhen, general manager of AI simulation at SandboxAQ.

With only traditional research methods proving insufficient, AI has emerged in Alzheimer’s research, addressing hurdles in drug discovery and diagnosis. AI’s ability to analyze massive data sets and model biological processes opens up opportunities for early detection, understanding disease mechanisms, and even predicting drug efficacy.

The field has already benefited from significant improvements, but can AI bring even more to Alzheimer’s research?

Diagnosing Alzheimer’s earlier and more effectively with AI

“So far, the Alzheimer’s area where AI has had the biggest impact is probably in diagnosis and detection. Artificial intelligence-assisted medical imaging allows us to detect disease much earlier,” Harhen pointed out.

AI has had a significant impact on Alzheimer’s diagnosis, particularly by assisting in medical imaging and biomarker analysis. Advanced machine learning models are now being trained to detect early signs of Alzheimer’s disease in brain imaging scans, such as amyloid plaques, with impressive accuracy.

Indeed, a study from the University of California, San Fransisco (UCSF) has shown that artificial intelligence can analyze patient records to predict Alzheimer’s disease up to seven years before traditional clinical symptoms appear, making it a game changer for early detection and opening doors to earlier interventions.

Beyond imaging, AI has vast potential in analyzing biomarkers – measurable indicators in blood, cerebrospinal fluid and genetic data that could pre-symptomatically predict Alzheimer’s disease. “More and more people are working on predictive tools, but if I had to identify an opportunity, I see very few working on biomarkers,” Harhen confirmed.

Similar to how AI has contributed to oncology research, these tools have rapidly transformed the field into diagnosis and prediction. But another obvious area where AI has a lot to offer Alzheimer’s research is drug discovery.

Alzheimer’s Drug Discovery: Can AI Change the Paradigm?

“Beyond diagnosis, another transformative change is in drug discovery; we can now simulate how hypothetical drugs work at the molecular level and analyze large sets of compounds much faster,” said Harhen.

But how does it work? Some of SandboxAQ’s work in the neurodegenerative space was with Nobel laureate Dr. Stanley Prusiner, Harhen told Labiotech. She explained how the process went and how AI helped.

“Dr. Prusiner had several hypotheses about how Alzheimer’s develops. What we did with computational AI was not only replicate what they were seeing in the lab—which often looks at proxies and combines different data to form a complete picture—but also build simulations that played like a movie, allowing – us to understand what is happening with direct observation.”

Basically, AI enables a transition to direct observation to determine which hypotheses are correct and which targets are relevant with a better understanding of the mechanisms at work – and according to Harhen, this is one of the main challenges in this space.

“The hard part is determining the mechanism of action. In my opinion, you have to start with a hypothesis or use all available data to generate one. That’s what AI allows us to do today – combining data in a logical way to get the right hypothesis.”

This is made possible by what we call large quantitative models (LQMs), which we hear less about than large language models (LLMs) that get a lot of hype because of their ability to ingest large amounts of text data.

But in biotech and more specifically in the Alzheimer’s space, LQMs have an advantage over LLMs. “Quantitative AI—mathematics, physics, molecules—is not based on words, and LLMs cannot solve these problems. This is where LQMs come in. LQMs, trained by mathematical models, allow us to understand the physics behind molecular interactions and how molecules behave with their targets,” said Harhen.

This makes LQMs extremely versatile and undoubtedly hold potential far beyond Alzheimer’s and neurodegenerative diseases. Harhen also pointed out that because LQMs are based on physics, they do not hallucinate; it is based on the laws of reality.

“I like to think of the AI ​​landscape as a bell curve, with many popular and useful tools in the middle. With LQMs, we focus on cutting edge cases where most of these common tools are not enough – like in Alzheimer’s disease,” said Harhen.

AI’s drug discovery capabilities extend beyond simple screenings. AI-based platforms can rapidly simulate molecular interactions, assess blood-brain barrier permeability, and predict adverse effects, significantly reducing the time and cost traditionally required.

Beyond SandboxAQ simulations, for example, Exscientia’s AI platform Centaur Chemist uses predictive modeling to streamline candidate compound selection. This platform has already brought three AI-engineered drug candidates to clinical testing, including DSP-0038 for Alzheimer’s psychosis, which targets serotonin receptors to help alleviate behavioral symptoms associated with the disease.

In partnership with Cambridge University, and Insilico Medicine application its artificial intelligence model, PandaOmics, to target proteins implicated in Alzheimer’s disease through a process known as “protein phase separation.” They identified therapeutic targets such as MARCKS, CAMKK2 and p62 – proteins likely involved in Alzheimer’s progression due to their tendency to form abnormal protein aggregates.

Although no AI drug candidate for Alzheimer’s disease has yet completed the full clinical journey, it may only be a matter of time before new solutions are brought to patients by LQM. And what if the next big thing in the Alzheimer’s space was already on the market, but we just didn’t know it had such potential?

Reuse of drugs already approved with AI

Discovering a new drug is not the only way to bring new solutions to patients. “We saw with GLP-1 agonists that they were originally developed for diabetes, but then we discovered a beneficial side effect for weight loss. Instead of competing for a single treatment, companies are now pursuing various indications in parallel, expanding the market. drug repurposing is becoming a big field, and I think we’ll see something similar happen with Alzheimer’s treatments,” Harhen said.

Harhen mentioned Every Cure, but it’s not the only company aiming to find the next big thing in “old” drugs. The Harvard Framework for Drug Repurposing in Alzheimer’s Disease (DRIAD). application machine learning to screen and identify existing drugs that could treat Alzheimer’s disease by repurposing them for neuroprotection.

The DRIAD platform screened numerous anti-inflammatory and neuroprotective drugs, prioritizing those that affect pathways relevant to Alzheimer’s pathology. The DRIAD team applied the screening method to approximately 80 potential candidates and shortlisted the most promising ones. According to the AI ​​model and the scientists who operate it, Janus Kinase (JAK) inhibitors. could have serious potential in the Alzheimer’s space.

Another example is the Dream (Drug Repurposing for Effective Alzheimer’s Medicines) study. Led by the National Institute on Aging (NIA), it study aims to identify and validate drugs originally approved for other conditions but showing promise against Alzheimer’s disease. The researchers identified 35 FDA-approved drugs that target 20 metabolic pathways associated with Alzheimer’s, narrowing them down to 15 candidates for further analysis. If these drugs prove effective, they may provide faster and more affordable treatment options for Alzheimer’s disease compared to traditional methods of drug development.

This is another promising application of artificial intelligence, but the revolutionary tool still faces some challenges in the neurodegenerative field and more specifically in Alzheimer’s disease.

What is AI missing from finding the next big Alzheimer’s treatment?

AI is not just hype, the hype is likely to create unreasonable expectations. To enable AI to have the transformative impact it has promised in Alzheimer’s disease, issues still need to be addressed.

“One challenge is feeding the data; we don’t have that many. It’s a very common challenge in AI – there’s not enough data, it’s not in the same place, and it’s not organized. The other challenge is that it is not generalizable. “When you look at genetic diversity, the differences are so small that they are extremely hard to see,” noted Harhen.

The effectiveness of AI in Alzheimer’s research is hampered by the limited availability of high-quality standardized data. Data sets are often small, heterogeneous, or collected under different conditions, which complicates the training of AI models that require reliable and diverse data to effectively generalize across populations.

However, Haren explained that this lack of data is not a problem in all AI applications for Alzheimer’s disease. “For the drug discovery process, we don’t need to ingest third-party data for training because we use data from molecule-to-molecule interactions and that can be synthetically generated. However, when you need data, I advise you to carefully consider which partner to go with in order to have the most ethically collected data.”

Beyond the technical limitations, AI can still be a touchy subject. “I think there is still mistrust of AI because of explainability. And it is exacerbated in clinical settings because health is a subject of low trust.”

Indeed, there are still concerns about the explainability of how the AI ​​reaches its conclusions. For now, AI is like a black box where the complicated mechanisms are quite opaque, and experts are gradually trying to make more sense of it. In addition to the black box problem, data collection is very complex, especially when it comes to particularly sensitive data such as medical data.

The first AI-discovered Alzheimer’s treatment has yet to reach the market and patients, and its arrival may be the final step needed for the public to have full confidence in these tools that are transforming biotechnology—not just the neurodegenerative landscape.