The authors of a recent review analyzed the results of 14 studies that included 139 million people to identify patterns that connect dementia risk to commonly used medications [1].
Finding a new purpose
Despite its prevalence in the elderly population, there is a lack of effective clinical treatments for dementia, and there is ongoing research to find new medications and therapies that can cure or slow down dementia. While this is essential, there is also another approach that can be simultaneously applied: repurposing existing drugs to slow down the progress of dementia. The increased availability of routinely collected medical data makes it possible to conduct studies involving millions of patients and hundreds of drugs.
The researchers point out that some drugs that are currently prescribed for different conditions are already known to affect dementia risk. For example, some diabetes drugs have been linked to dementia risk reduction [2].
Driven by data
The authors of this study conducted a systematic review of studies that researched the association between prescribed medications and dementia risk. They limited their analysis to the data-driven, rather than hypothesis-driven, approach. They define data-driven as “an exploratory approach that analyzes large datasets to extract insights and patterns by applying analytical techniques and modes of reasoning.”
Such an approach has its advantages and disadvantages. While it can exclude some high-quality studies that are hypothesis-driven, which limits the study to only a subset of data that pertains to a given hypothesis, it also has some advantages. Since the data-driven approach uses all available data of all currently prescribed drugs, the researchers can identify previously unidentified associations missed by hypothesis-based approaches.
In their analysis, they included fourteen studies from the USA, Japan, South Korea, Germany, and Wales, including 139 million people and 1 million cases of dementia, and investigated an estimated 200 pharmacological subgroups, including more than 2000 ingredients.
Some drugs reduce dementia risk
In general, the researchers found that inconsistencies between the studies made it difficult to analyze the impact of individual drugs on dementia risk. Nevertheless, they found some general trends in different classes of medicines. For example, antimicrobials, vaccines, and anti-inflammatories were linked to reduced dementia risk. Antimicrobials and vaccines may protect against dementia because they address viral and bacterial infections, which have been linked to an increase in dementia risk [3-5].
The inclusion of anti-inflammatories is not surprising, since inflammation is a widely known pathogenic pathway in aging. Suppressing inflammation has already been investigated as a possible target for Alzheimer’s disease [6]. Even though clinical trials of these agents have not been successful so far [7], the authors believe that “using the right agent at the right time point in disease progression, perhaps prior to manifestation of cognitive decline, may be crucial.”
Other drugs increase dementia risk
On the other hand, antipsychotics and drugs for diabetes were linked to increased dementia risk.
The authors believe that for antipsychotics and antidepressants, some of the associations can be explained by reverse causation. For example, antidepressants can be prescribed in the early stages of dementia to treat mood disorders. Including these patients in the dataset means that antidepressants become associated with an increased likelihood of dementia, even though dementia can lead to antidepressant use.
However, other drugs may have dementia as a potential side effect, and in these cases, it is crucial to investigate the mechanisms of action in order to better avoid prescribing such drugs to people who are at risk for dementia.
Conflicting results
There were also classes of medications including antihypertensives, antidepressants, and, to a lesser extent, drugs to manage blood glucose levels that showed conflicting results, as some data suggests that they increase dementia risk, while other data suggests that they decrease it.
The authors discuss a few possible explanations for the conflicting results. One drawback to this line of research is that it analyzes classes of drugs and not specific drugs by themselves. Different members of the same class of drugs might act on various molecular targets and exert distinct effects on organisms.
For example, some antihypertensives lead to the upregulation of autophagy, a biological process linked to longevity [8]. However, this is not the case for all antihypertensives. Since all antihypertensives are analyzed together, such beneficial effects of single drugs might be missed, leading to false negative results. Nevertheless, the authors believe that in their research, false negatives are unlikely since their data includes large numbers of both people and drugs.
The authors also bring up an important caveat: suboptimal datasets used for research. The datasets used in the study were created for clinical purposes, not research, and they may not include important information, such as drug administration. Those drugs were prescribed to patients, but we cannot be confident that patients took them.
On the other hand, there might be people who take some over-the-counter medications without a prescription. Such data will be missing. There is also no information about dose responses or analysis of taking single or multiple medications simultaneously, and there is often missing data regarding confounding factors, such as socioeconomic status, genotyping, and biomarker information. There is also a question of the accuracy of dementia diagnoses, since dementia is often under- or misdiagnosed.
Setting priorities
The researchers believe that their work can aid in prioritizing which drugs should be explored in further studies for potential repurposing to treat dementia. Those studies might involve a single drug or drug combinations since dementia is a complex condition, and to treat such a condition, it might be necessary to target it with multiple drugs that address numerous molecular pathways involved in its pathology.
Literature
[1] Underwood, B. R., Lourida, I., Gong, J., Tamburin, S., Tang, E. Y. H., Sidhom, E., Tai, X. Y., Betts, M. J., Ranson, J. M., Zachariou, M., Olaleye, O. E., Das, S., Oxtoby, N. P., Chen, S., Llewellyn, D. J., & Deep Dementia Phenotyping (DEMON) Network (2025). Data-driven discovery of associations between prescribed drugs and dementia risk: A systematic review. Alzheimer’s & dementia (New York, N. Y.), 11(1), e70037.
[2] Nørgaard, C. H., Friedrich, S., Hansen, C. T., Gerds, T., Ballard, C., Møller, D. V., Knudsen, L. B., Kvist, K., Zinman, B., Holm, E., Torp-Pedersen, C., & Mørch, L. S. (2022). Treatment with glucagon-like peptide-1 receptor agonists and incidence of dementia: Data from pooled double-blind randomized controlled trials and nationwide disease and prescription registers. Alzheimer’s & dementia (New York, N. Y.), 8(1), e12268.
[3] Muzambi, R., Bhaskaran, K., Brayne, C., Davidson, J. A., Smeeth, L., & Warren-Gash, C. (2020). Common Bacterial Infections and Risk of Dementia or Cognitive Decline: A Systematic Review. Journal of Alzheimer’s disease : JAD, 76(4), 1609–1626.
[4] Levine, K. S., Leonard, H. L., Blauwendraat, C., Iwaki, H., Johnson, N., Bandres-Ciga, S., Ferrucci, L., Faghri, F., Singleton, A. B., & Nalls, M. A. (2023). Virus exposure and neurodegenerative disease risk across national biobanks. Neuron, 111(7), 1086–1093.e2.
[5] Ballard, C., Aarsland, D., Cummings, J., O’Brien, J., Mills, R., Molinuevo, J. L., Fladby, T., Williams, G., Doherty, P., Corbett, A., & Sultana, J. (2020). Drug repositioning and repurposing for Alzheimer disease. Nature reviews. Neurology, 16(12), 661–673.
[6] Cummings, J., Zhou, Y., Lee, G., Zhong, K., Fonseca, J., & Cheng, F. (2024). Alzheimer’s disease drug development pipeline: 2024. Alzheimer’s & dementia (New York, N. Y.), 10(2), e12465.
[7] Meyer, P. F., Tremblay-Mercier, J., Leoutsakos, J., Madjar, C., Lafaille-Magnan, M. E., Savard, M., Rosa-Neto, P., Poirier, J., Etienne, P., Breitner, J., & PREVENT-AD Research Group (2019). INTREPAD: A randomized trial of naproxen to slow progress of presymptomatic Alzheimer disease. Neurology, 92(18), e2070–e2080.
[8] Siddiqi, F. H., Menzies, F. M., Lopez, A., Stamatakou, E., Karabiyik, C., Ureshino, R., Ricketts, T., Jimenez-Sanchez, M., Esteban, M. A., Lai, L., Tortorella, M. D., Luo, Z., Liu, H., Metzakopian, E., Fernandes, H. J. R., Bassett, A., Karran, E., Miller, B. L., Fleming, A., & Rubinsztein, D. C. (2019). Felodipine induces autophagy in mouse brains with pharmacokinetics amenable to repurposing. Nature communications, 10(1), 1817.
