A decade ago, a novel method for studying Alzheimer’s disease was introduced, known as “Alzheimer’s in a dish”. This model uses mature brain cells cultured in a gel, mimicking the progress of the disease in the human brain over a decade in just six weeks. But how accurately does this model reflect the changes seen in Alzheimer’s patients? A recent study conducted by Mass General Brigham, in collaboration with Beth Israel Deaconess Medical Center (BIDMC), offers some answers.

The researchers developed an algorithm to objectively assess how closely these Alzheimer’s disease models resemble the gene expression patterns and function seen in patients’ brains. The results, published in the scientific journal Neuron, identified key common pathways, confirming the model’s effectiveness for drug assessment and revealing potential new drug targets.

“Our goal is to find a model that closely mirrors the activity we see in the brains of Alzheimer’s patients,” said co-senior author Doo Yeon Kim, PhD, from the Department of Neurology at Massachusetts General Hospital (MGH). “We developed this 3D cell culture model for Alzheimer’s a decade ago. Now, we have proof that it can help speed up drug discovery.”

The study was a collaboration between researchers specializing in neurology and data systems, all united in their mission to find better treatments for Alzheimer’s disease. Traditional mouse models for Alzheimer’s research have limitations, as they do not develop the same Alzheimer’s markers as humans. While Kim and his colleagues have developed different models, determining their accuracy in reflecting molecular and functional changes has been a challenge.

Co-senior author Winston Hide, PhD, from BIDMC’s Department of Pathology, articulated a central dilemma: “Understanding which models truly reflect the complexity of Alzheimer’s in the human brain was a significant hurdle. By focusing on broader biological pathways instead of individual genes, we’ve developed a system that revolutionizes how we discover and test drugs.”

To solve this issue, the research team developed an innovative analytical platform called Integrative Pathway Activity Analysis (IPAA). It can identify which models most closely mirror the functional changes seen in Alzheimer’s and reveal the most promising pathways for drug discovery.

The researchers identified 83 disrupted pathways common to both brain samples from deceased Alzheimer’s patients and 3D cell models. They tested drugs targeting one particular pathway as a proof of concept and found a yet untested clinical inhibitor was highly effective at reducing Alzheimer’s pathology in the dish model, showing potential for future clinical trials.

Excitingly, the study’s findings go beyond a single pathway. The platform’s ability to identify promising drug targets, combined with the speed and scalability of the Alzheimer’s in a dish model, enables the simultaneous testing of multiple drugs. The researchers have already tested hundreds of approved drugs and natural products, paving the way for future clinical trials.

“We now have a system that not only allows us to test new drugs rapidly, but also a platform that can predict which drugs will be most effective,” said co-senior author Rudolph Tanzi, PhD, from MGH. “These advancements bring us one step closer to discovering better drugs and making them available to patients.”