Genomic Evidence on Antihypertensive Drug Targets and Oral Disease Outcomes
Aiswarya Puzhakkara Chennas, Ignacio Leiva‐Escobar, Stefan Lars Reckelkamm, Birte Holtfreter, Thomas Kocher, Sebastian‐Edgar Baumeister, Michael Nolde, Zoheir AlayashABSTRACT
Aim
Antihypertensive medications are widely prescribed for hypertension, yet evidence regarding their oral health safety remains limited. We aimed to investigate whether genetically proxied pharmacological actions of antihypertensive drug classes influence the risk of periodontitis and dental caries using a two‐sample drug‐target Mendelian randomization framework.
Methods
Genetic instruments were obtained for 12 antihypertensive drug classes using variants within or near target genes associated with systolic blood pressure (SBP) in two genome‐wide association study (GWAS) meta‐analyses: one combining data from the International Consortium of Blood Pressure and UK Biobank, and another additionally including the Million Veteran Program (MVP) and Vanderbilt University's biorepository. Summary statistics for chronic periodontitis were obtained from the MVP GWAS, and for dental caries from the Gene‐Lifestyle Interactions in Dental Endpoints consortium using the decayed, missing, and filled surfaces index. Causal effects were estimated using the inverse‐variance weighted method with Benjamini–Hochberg correction for multiple testing. Colocalization analyses assessed whether drug targets and outcomes shared causal variants.
Results
Genetically proxied diazoxide acting through the KCNJ11 locus was associated with chronic periodontitis after multiple‐testing correction (odds ratio [OR] per 1 mmHg lower SBP: 1.06, 95% confidence interval [CI]: 1.02–1.09), with consistent estimates after excluding variants associated with type 2 diabetes. No other associations were observed. Colocalization analyses provided no evidence of shared causal variants.
Conclusion
We found little evidence that genetically proxied on‐target pharmacological actions of antihypertensive drug classes were associated with periodontitis or dental caries. The diazoxide–periodontitis association should be interpreted with caution, given possible residual pleiotropy at the KCNJ11 locus and the absence of colocalization.