OC 08.5 - Impact of common medication use on platelet transcript levels.

Background: Associations between specific drugs and protein, transcript or even DNA methylation levels have been investigated in many cell-types, but little is known about the impact of common medications on the platelet transcriptome. Such associations could reveal direct or indirect gene-drug interactions in the context of platelets, and provide guidance for future investigations of these platelet transcripts.

Aims: To identify platelet transcript levels impacted by common medication use.

Methods: Platelet RNA-sequencing was performed for 646 Framingham Heart Study participants during the 2016-2019 exam. Abundances were obtained for 14,471 transcripts with the nf-core/rnaseq quantification pipeline. Medication use was assessed with a medicine bag brought by each participant. Medication categories (such as anti-depressants) and more refined sub-categories (such as serotonin-affecting or norepinephrine-affecting within anti-depressants) were selected for analysis if at least 40 (6.2%) users were identified within the (sub-)category. Association analyses were performed with GMMAT, which accounts for relatedness, and adjusted for age, sex, cohort, and principal components. Significant associations were identified after multiple testing correction with the Benjamini-Hochberg FDR (5%) procedure.

Results: We analyzed 8 medication categories: anti-hypertensives, anti-lipids, anti-depressants, anti-platelets, lung medicines, proton-pump inhibitors, anti-diabetes and hormone therapy, and 16 additional sub-categories. Overall, we observed 51 significant associations (at P< 6.76E-6), which involved 33 transcripts and the following medication categories or their sub-categories: anti-hypertensives (N=20 associations), anti-depressants (N=13), anti-lipids (N=9), anti-diabetes (N=5), anti-platelets (N=3) and lung medication (N=1). Notably, we observed several associations previously reported in other cell-types, such as the use of statins which was associated with decreased expression of ABCA1 and ABCG1.

Conclusion(s): Medication use appears to have a very narrow impact on the platelet transcriptome, with only 33 transcripts involved in significant associations. However, with this new resource, investigators of the platelet transcriptome will be better equipped to understand signals involving these transcripts knowing the potential interaction with medication use.