Sufficient evidence has not been provided to support the hypothesis that the use of phosphodiesterase-5 (PDE5) inhibitors reduces the risk for incident Alzheimer disease and related dementia (ADRD), according to study findings published the journal Brain Communications.
In the trial, Drug Repurposing for Effective Alzheimer’s Medicines (DREAM; ClinicalTrials.gov Identifier: NCT05039086) researchers compared the incidence of ADRD following initiation of a PDE5 inhibitor, such as sildenafil and tadalafil, vs initiation of an endothelin receptor antagonist (ERA; bosentan, ambrisentan, macitenan) among patients with pulmonary arterial hypertension (PAH), after controlling for 76 confounding variables through propensity score matching.
A new-user, active-comparator cohort study design was implemented, in which patients were required to have 365 days of continuous enrollment in Medicare Parts A, B, and D (considered the baseline period) prior to cohort entry (defined as the date of initiating either a PDE5 inhibitor or an ERA). All participants needed to have 2 or more claims with a diagnosis of PAH during the baseline period, including the cohort entry date.
A total of 9968 initiators of PDE5 inhibitors and 3053 initiators of ERA fulfilled study inclusion criteria. For confounding adjustment, 2888 initiators of PDE5 inhibitors (ie, 73.6% treated with sildenafil; 26.4% treated with tadalafil) were pair-matched with 2888 initiators of comparable ERA. The average participant age was 74 years (range, 65 to 96 years). Overall, 69% of the patients were women.
In this cohort, the prevalence of comorbid conditions was high, with 90% of participants having hypertension, 75% having heart failure, and 43% having atrial fibrillation. In accordance with recommended dosing for PAH, more than 99% (2107 of 2123) of those treated with sildenafil received a 20-mg, 3-times-daily regimen and 98% (745 of 761) of those treated with tadalafil received a 40-mg daily regimen.
A total of 4 separate analytic approaches that were designed to address specific types of biases, including:
- Analysis 1: An “as-treated” follow-up approach
- Analysis 2: An “as-started” follow-up approach that incorporated a 6-month “induction” period
- Analysis 3: Incorporating a 6-month “symptoms to diagnosis” period
- Analysis 4: High-specificity outcome definition
No evidence of a decreased risk for ADRD was observed with PDE5 inhibitor use in all 4 analyses. Hazard ratios were 0.99 (95% CI, 0.69-1.43), 1.00 (95% CI, 0.71-1.42), 0.67 (95% CI, 0.43-1.06), and 1.15 (95% CI, 0.57-2.34), respectively.
Additionally, no evidence demonstrated that sildenafil improved diverse molecular abnormalities relevant to AD in most cell culture–based phenotypic assays.
Several limitations of the present analysis warrant mention. To begin, it is possible that the study might have been underpowered to detect differences that are of a small magnitude. Further, the average length of follow-up was short, which could lead to underestimation of any effects that might necessitate a lengthier treatment period. Additionally, the population in the current study is atypical, experiencing a high cardiometabolic disease burden that may not be representative of older individuals at risk for ADRD in routine care.
The researchers concluded that “While wider use of routinely collected health care data to evaluate biological hypotheses for drug repurposing is a welcome development, caution is warranted to avoid common pitfalls and consequent overinterpretation of estimates generated from these data.”
Disclosure: Some of the study authors have declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.
This article originally appeared on Neurology Advisor
Desai RJ, Mahesri M, Lee SB, et al. No association between initiation of phosphodiesterase-5 inhibitors and risk of incident Alzheimer’s disease and related dementia: results from the Drug Repurposing for Effective Alzheimer’s Medicines study. Brain Commun. Published online October 4, 2022. doi: 10.1093/braincomms/fcac247