Effects of antidiabetic agents on platelet characteristics with implications in Alzheimer's disease: Mendelian randomization and colocalization study

Background: Observational studies have found a potential link between the use of thiazolidinediones (TZDs) and a lower risk of Alzheimer's disease (AD) development. Platelets were the great source of amyloid-β (Aβ) and involved in the development of AD. This study aimed to assess the correlation between antidiabetic agents and platelet characteristics, hoping to provide a potential mechanism of TZDs neuroprotection in AD. Method: Drug-targeted Mendelian randomization (MR) was performed to systematically illustrate the long-term effects of antidiabetic agents on platelet characteristics. Four antidiabetic agent targets were considered. Positive control analysis for type 2 diabetes (T2D) was conducted to validate the selection of instrumental variables (IVs). Colocalization analysis was used to further strengthen the robustness of the results. Result: Positive control analysis showed an association of four antidiabetic agents with lower risk of T2D, which was consistent with their mechanisms of action and previous evidence from clinical trials. Genetically proxied TZDs were associated with lower platelet count (β[IRNT] = −0.410 [95 % CI -0.533 to −0.288], P = 5.32E-11) and a lower plateletcrit (β[IRNT] = −0.344 [95 % CI -0.481 to −0.206], P = 1.04E-6). Colocalization suggested the posterior probability of hypothesis 4 (PPH4) > 0.8, which further strengthened the MR results. Conclusion: Genetically proxied TZDs were causally associated with lower platelet characteristics, particularly platelet count and plateletcrit, providing insight into the involvement of platelet-related pathways in the neuroprotection of TZDs against AD. Future studies are warranted to reveal the underlying molecular mechanism of TZDs’ neuroprotective effects through platelet pathways.