Scientists have identified a potential new treatment strategy for Alzheimer’s disease using a combination of two existing cancer drugs. The research suggests that targeting multiple types of brain cells simultaneously offers a more effective approach than focusing on a single target. These findings were published in the journal Cell.
The study was led by Yaqiao Li and Yadong Huang from Gladstone Institutes, alongside Marina Sirota from the Bakar Computational Health Sciences Institute at the University of California, San Francisco. The team initiated this research to address the high failure rate of Alzheimer’s drug development. Most current treatments focus on a single aspect of the pathology, such as amyloid plaques, or a single cell type, such as neurons.
However, Alzheimer’s is a complex condition that affects various biological networks. The researchers aimed to develop a strategy that acknowledges this complexity. They sought to correct the dysregulated gene expression networks across multiple cell types simultaneously. This approach relies on the idea that restoring the health of both neurons and the glial cells that support them is necessary for effective treatment.
“Alzheimer’s is a complex disease and currently lacks effective treatment. It affects multiple brain cell types through diverse and interconnected molecular pathways,” said study author Yaqiao Li, a postdoctoral fellow at Gladstone Institutes.
“Rather than focusing on a single gene or protein, our approach embraces a systems biology perspective—aiming to correct dysregulated gene expression networks across multiple cell types and restore them toward healthy-state behaviors.”
“Recent advances in computational power, large-scale transcriptomics, and network modeling make it possible to map these cell-type-specific regulatory programs with high resolution and design interventions that target the underlying system-level dysfunction rather than isolated components. The transformative power of combining the advancement of computational capability, real-world large clinical database, and improved animal models naturally led us to explore this topic.”
The researchers utilized a computational method driven by human data. They began by integrating large datasets of single-nucleus RNA sequencing from post-mortem human brains. This technology allows scientists to examine genetic activity at the level of individual cells rather than in bulk tissue.
This granular analysis revealed specific gene expression changes associated with Alzheimer’s across distinct cell populations. The team identified unique pathological signatures in excitatory neurons, inhibitory neurons, and glial cells like microglia and astrocytes. This step established a detailed map of the cellular dysfunction present in the disease.
With these disease signatures in hand, the researchers employed a computational drug repurposing pipeline. They queried a database known as the Connectivity Map, which contains information on how various drugs affect gene expression. The algorithm looked for drugs that produced gene expression patterns opposite to those seen in Alzheimer’s disease.
The goal was to find compounds that could essentially flip the diseased gene networks back to a healthy state. This screening process identified letrozole and irinotecan as top candidates. Letrozole is an aromatase inhibitor used to treat breast cancer, while irinotecan is a topoisomerase inhibitor used for colorectal cancer.
The computational prediction indicated that letrozole primarily targets defects in neurons. Conversely, irinotecan appeared to target the support networks in glial cells. The researchers hypothesized that combining these two drugs would provide a synergistic effect by addressing the disease on multiple fronts.
Before testing in animals, the team sought validation in real-world human data. They analyzed electronic medical records from over one million patients within the University of California health system. They looked for a correlation between the use of these specific cancer drugs and the incidence of Alzheimer’s disease.
The analysis compared patients who had taken letrozole or irinotecan for cancer against matched control groups who had not. The data indicated that individuals exposed to these drugs had a significantly lower risk of being diagnosed with Alzheimer’s. This real-world evidence strengthened the case for further experimental testing.
The researchers then proceeded to validate the combination therapy in a mouse model. They utilized mice engineered to express both amyloid beta and tau pathologies. These mice develop memory deficits and brain changes that closely mimic the human condition.
The mice were divided into four groups for testing. One group received a vehicle solution, two groups received either letrozole or irinotecan alone, and the final group received the combination of both. The treatment was administered over a period of three months.
Behavioral testing showed that the combination therapy produced the most significant benefits. Mice treated with both drugs demonstrated improved learning and memory retention in spatial navigation tasks. While single-drug treatments offered some benefit, the combination outperformed them consistently.
Pathological examination of the mouse brains corroborated the behavioral results. The combination treatment significantly reduced the area covered by amyloid plaques. It also lowered the levels of phosphorylated tau, a protein associated with the formation of neurofibrillary tangles.
The researchers also assessed the health of the neurons and glial cells. The combination therapy prevented the loss of neurons in the hippocampus, a brain region essential for memory formation. Furthermore, the treatment reduced signs of neuroinflammation driven by microglia and astrocytes.
To understand the molecular mechanisms, the researchers performed single-nucleus RNA sequencing on the treated mice. This analysis confirmed that the drugs worked as the computational model had predicted. The combination therapy reversed the disease-associated gene networks in a cell-type-specific manner.
The data provides evidence that letrozole helped correct neuronal pathways related to synaptic activity. Simultaneously, irinotecan appeared to modulate inflammatory and metabolic pathways in glial cells. This dual action likely contributes to the superior efficacy observed with the combination.
“We discovered that two FDA-approved cancer drugs that work together to reverse key features of Alzheimer’s disease in mouse models by targeting different brain cell types,” Li told PsyPost. “This combination therapy improved memory and reduced brain damage better than untreated or either drug alone.”
“The real surprise is that the top two drugs are both cancer treatment drugs. It is very exciting, with a bit of surprise, that the combination therapy with one drug targeting Alzheimer’s-related genetic changes in neurons and another targeting glia performed remarkably well in mice genetically engineered to develop aggressive Alzheimer’s-like symptoms.”
“It reduced brain pathology and improved memory more effectively than either drug alone or no treatment,” Li continued. “These results suggest that addressing the disease’s complexity by targeting multiple cell types and pathways simultaneously with cell-type-precision therapy may be key to unlocking a cure.”
Despite the promising results, the study has some limitations. The initial drug screening utilized databases based on cancer cell lines rather than brain cells. This difference means the predicted effects require extensive validation in neurological contexts.
Additionally, the study noted sex differences in the mouse models. Male mice showed more robust behavioral improvements compared to female mice. The researchers suggest this could be due to hormonal factors or specific characteristics of the mouse model used.
“Current treatment options for Alzheimer’s disease remain very limited,” Li noted. “Because these two drugs are already approved for other indications, they provide a strategic head start toward future clinical testing, and we are actively pursuing opportunities and partnerships to advance this combination therapy into clinical trials.”
“Looking ahead, our long-term goal is to build on this work by integrating artificial intelligence (AI) technology with large-scale molecular, clinical, and drug databases to enable a fast-track therapeutic development pipeline toward true precision medicine for Alzheimer’s—where therapies are guided not only by symptoms, but by each patient’s unique molecular and cellular signatures as well as clinical profiles.”
“This study exemplifies the power of interdisciplinary collaboration in biomedical research—combining large-scale drug screening and computational tools developed at UCSF with deep expertise in disease biology and experimental validation at the Gladstone Institutes,” Li added. “The seamless integration of computational and experimental biology was central to the success of this study.”
The study, “Cell-type-directed network-correcting combination therapy for Alzheimer’s disease,” was authored by Yaqiao Li, Carlota Pereda Serras, Jessica Blumenfeld, Min Xie, Yanxia Hao, Elise Deng, You Young Chun, Julia Holtzman, Alice An, Seo Yeon Yoon, Xinyu Tang, Antara Rao, Sarah Woldemariam, Alice Tang, Alex Zhang, Jeffrey Simms, Iris Lo, Tomiko Oskotsky, Michael J. Keiser, Yadong Huang, and Marina Sirota.
