Investigating drug effects on aging and alzheimer's disease using causal inference approaches

The geroscience hypothesis offers a compelling approach to extending human lifespan and healthspan by targeting fundamental aging mechanisms. Recent advances in biomarkers for biological aging (BA biomarkers) present promising tools for measuring the aging process in humans and may serve as valuable endpoints for assessing the impact of interventions, including pharmacological treatments, on aging. Alzheimer's disease (AD), one of the most prevalent age- related diseases, imposes immense suffering on patients and families and poses significant healthcare burdens in aging societies. While recent amyloid-targeted immunotherapies show progress in slowing cognitive decline in early-stage AD, emerging evidence suggests that combined therapies targeting multiple pathological pathways, as well as early interventions in preclinical or earlier stages, may hold the greatest promise for preventing or managing AD progression. This doctoral thesis evaluates the effects of various drugs on aging and AD onset, aiming to provide insights into drug candidates with repurposing potential as anti-aging agents or AD-modifying treatments.

Studies I and II employed a longitudinal design using conditional generalized estimating equation (cGEE) models to analyze within-individual variations across two aging cohorts from Sweden and the United States. A broad range of BA biomarkers-spanning molecular, physiological, and functional levels-served as outcome measures. Multiple drug classes were evaluated for associations with BA biomarker changes, identifying calcium channel blockers, vitamin D analogues, bisphosphonates, thyroid hormone replacements, thiazides, and proton pump inhibitors as associated with reductions in one or more BA biomarkers.

Study III utilized a Mendelian randomization design to investigate the genetically predicted effects of antidiabetic drugs on AD risk. Findings suggested that the genetically predicted effects of sulfonylureas and glucagon-like peptide-1 (GLP- 1) analogues were associated with approximately 60% and 70% reductions in AD risk per 1 mmol/L decrease in blood glucose levels, respectively.

Study IV evaluated the effects of GLP-1 analogues, sulfonylureas, and dipeptidyl peptidase-4 (DPP-4) inhibitors on dementia onset in a real-world setting using Swedish register data specifically comparing their effectiveness among old individuals with type 2 diabetes mellitus (T2DM). Results indicated that GLP-1 analogues were associated with 30 and 23% reduced risks of dementia compared to sulfonylureas and DPP-4 inhibitors, respectively.

In conclusion, Studies I and II identified several drug classes that may benefit certain aspects of human aging, though these findings are exploratory and warrant further validation in larger, diverse populations. Studies III and IV highlighted the potential neuroprotective effects of sulfonylureas and GLP-1 analogues. The limited brain penetration of current sulfonylureas suggests an avenue for future research into the neuroprotective role of sulfonylurea targets in the brain. While GLP-1 analogues emerge as the most promising candidates for neuroprotection from these studies, their effects may vary between individuals with and without type 2 diabetes. Future randomized controlled trials are essential to clarify the neuroprotective effects of GLP-1 analogues across these populations.

List of scientific papers

I. Tang B, Li X, Wang Y, Sjölander A, Johnell K, Thambisetty M, Ferrucci L, Reynolds CA, Finkel D, Jylhävä J, Pedersen NL, Hägg S. Longitudinal associations between use of antihypertensive, antidiabetic, and lipid-lowering medications and biological aging. Geroscience. 2023 Jun;45(3):2065-2078. https://doi.org/10.1007/s11357-023-00784-8

II. Tang B, Kuo P, Moore A, Thambisetty M, Ferrucci L, Hägg S, Longitudinal Associations Between Medication Use and Phenotypic Aging: Insights from the Baltimore Longitudinal Study of Aging. [Submited]

III. Tang B, Wang Y, Jiang X, Thambisetty M, Ferrucci L, Johnell K, Hägg S. Genetic Variation in Targets of Antidiabetic Drugs and Alzheimer Disease Risk: A Mendelian Randomization Study. Neurology. 2022 Aug 16;99(7):e650-e659. https://doi.org/10.1212/wnl.0000000000200771

IV. Tang B, Sjölander A, Wastesson J, Maura G, Blotiere P-O, Szilcz M, Mak JK, Qin C, Alvarsson M, Religa D, Johnell K, Hägg S. Comparative effectiveness of glucagon-like peptide-1 agonists, dipeptidyl peptidase-4 inhibitors, and sulfonylureas on the risk of dementia in older individuals with type 2 diabetes in Sweden: an emulated trial study. eClinicalMedicine. 2024 Jun 1;73:102689. https://doi.org/10.1016/j.eclinm.2024.102689