AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Mitral Annulus-to-Leaflet Area: A Novel Quantitative Endophenotype for Mitral Valve Prolapse Enables Risk Stratification and Genetic Discovery
By Makarem Y, Balasubramanian AA, Chilazi M et al.
Background: Mitral valve prolapse (MVP), the most common valvular disorder, affects 2–3% of the population and is characterized by progressive remodeling of the mitral valve leaflets and annulus culminating in leaflet displacement and mitral regurgitation. Clinical diagnosis often occurs after significant valvular dysfunction has developed, despite leaflet deformation and annular remodeling evolving gradually over years. Objective: A quantitative measure of mitral leaflet prolapse that captures the continuous spectrum of mitral valve remodeling may improve MVP risk stratification and empower genetic discovery. Methods: We trained and validated a deep learning model to segment the mitral valve annulus and leaflets from 3-chamber cardiac magnetic resonance images in 63,000 UK Biobank (UKB) participants and measured the annulus-to-leaflet area at peak systole, a novel quantitative measure of the extent of leaflet prolapse. We then evaluated its associations with cardiovascular disease, characterized its genetic architecture through genome-wide analyses, and developed and externally validated a polygenic risk score ( Figure 1 ). Results: Smaller annulus-to-leaflet area was associated with both prevalent (OR /1-SD decrease = 2.46; 95% CI 1.60–3.78) and incident (HR /1-SD decrease = 3.16; 95% CI 2.27–4.41) MVP independent of measures of cardiac structure and function. In genome-wide association analysis, we identified 25 independent loci implicating genes associated with MVP ( WNT4, CASQ2, CPED1, SPTBN1, SYN3, SLC12A2, HMGA2, and CAMK2D ), extracellular matrix remodeling ( ADAM15, ADAMTS5 and PDZRN3 ), intracellular signaling ( AKAP13, NPR3, RXFP4, PLCE1, PRAG1, BLK, PTPRR, and IER3 ), and papillary muscle fibrosis ( TBX3 and SLC35F1 ) ( Figure 2 ). The annulus-to-leaflet area demonstrated significant genetic correlation with MVP ( rg = –0.44; p = 7.59х10 ⁻ ¹ ² ), and Mendelian randomization supported a causal role of the annulus-to-leaflet area in MVP susceptibility (β = –1.64; p = 1.4х10 ⁻ ¹ ¹ ). A polygenic risk score (PRS) of annulus-to-leaflet area was specifically associated with incident MVP in 418,420 non-overlapping UKB participants (HR /1-SD decrease = 1.19; 95% CI 1.12–1.28) and in the Mass General Brigham Biobank (MGBB) (HR /1-SD decrease = 1.16; 95% CI 1.09–1.24), with no significant associations observed across other cardiovascular outcomes ( Figure 3a ) Individuals in the bottom fifth percentile of the PRS had a 65% increased incidence of MVP in MGBB (HR = 1.65; 95% CI 1.29–2.11) ( Figure 3b ). Conclusions: These findings establish a population-scale quantitative imaging endophenotype of MVP that identifies individuals at risk of incident MVP, elucidates the genetic architecture underlying disease susceptibility, and reveals biological pathways that may represent therapeutic targets for MVP.