Background: Alzheimer’s Disease and Related Dementia (ADRD) is a major public health

challenge, with few disease-modifying therapies available. Drug repurposing targeting biological

mechanisms implicated in ADRD, including inflammation and proteotoxicity, is a promising

approach. In prior work, we identified tricyclics as a drug class with protective effects in ADRD

models. A subsequent cross-sectional pharmacoepidemiologic analysis of Medicare claims found

tricycic exposure associated with lower ADRD prevalence. To strengthen causal inference and

address temporality, we conducted a five year longitudinal study to evaluate the association

between tricyclic exposure and incident ADRD

Methods: We used a longitudinal cohort of 1,000,000 Medicare fee-for-service beneficiaries

aged 65-75 in 2012, randomly sampled and followed through 2016 or until death. Individuals

with ADRD at baseline were excluded. Tricyclic exposure was identified using Part D prescription

claims, with beneficiaries classified as exposed after receipt of at least one phenothiazine

prescription. The primary outcome was incident ADRD, identified using a modified Bynum

algorithm applied to inpatient and outpatient claims. We will estimate hazard ratios (HRs) and

95% confidence intervals (CIs) using Cox proportional hazards models, adjusting for age, sex,

race/ethnicity, ADRD comorbidities and risk factors. Sensitivity analyses will assess cumulative

exposure, confounding by indication

Results: The eligible study sample included 956,057 Medicare fee-for-service beneficiaries aged

65-75 in 2012 (M = 70.3, SD = 2.8). During follow-up, 62,973 individuals developed ADRD.

Among the 35,411 beneficiaries (4.2%) exposed to tricyclics, 1,128 (3.2%) developed ADRD,

compared with 14,915 (1.9%) of 800,546 unexposed individuals (95.8%). Compared to

unexposed beneficiaries, those exposed to tricyclics were slightly younger, more likely to be

female, and had higher rates of ADRD risk factors. 

Prochlorperazine exposure remains significantly associated with reduced ADRD In the full cohort as well as in

beneficiaries without psychiatric conditions

Conclusion: In this large longitudinal analysis of Medicare beneficiaries, 4.2% were exposed to

tricyclics. We will evaluate whether phenothiazine exposure is associated with ADRD risk over

five years of follow-up. These findings extend prior cross-sectional results and assess

hypotheses related to ADRD pathophysiology in humans. Despite limitations inherent to claims

data and potential residual confounding, this study provides real-world evidence to inform

whether tricyclics warrant further investigation as repurposed agents for ADRD prevention or

delay

In a large, nationally representative cohort of older adults, exposure to prochlorperazine was associated with a

reduced risk of incident ADRD over five years. These findings reinforce and extend preclinical evidence supporting

tricyclic compounds as modulators of neurodegenerative pathways. More broadly, this work demonstrates the power

of validating experimentally generated hypotheses in real-world human data to accelerate translation from

bench to bedside. Carefully designed prospective studies are now warranted to determine whether these associations

reflect true neuroprotective effects and to define their potential role in ADRD prevention or treatment strategies.

In a large, nationally representative cohort of older adults, exposure to prochlorperazine was associated with a

reduced risk of incident ADRD over five years. These findings reinforce and extend preclinical evidence supporting

tricyclic compounds as modulators of neurodegenerative pathways. More broadly, this work demonstrates the power

of validating experimentally generated hypotheses in real-world human data to accelerate translation from

bench to bedside. Carefully designed prospective studies are now warranted to determine whether these associations

reflect true neuroprotective effects and to define their potential role in ADRD prevention or treatment strategies.