Beyond the genome: the dominant role of biological age over polygenic risk in post-PCI mortality prediction.

Background

Polygenic risk scores (PRS) are widely used to predict long-term outcomes, but their value in secondary prevention remains unclear. Biological age (BA) may enhance risk stratification. This study investigated BA acceleration (BAA) as a dynamic complement to static PRS in post-PCI patients and examined whether BA provides additional prognostic value when genetic risk alone fails to differentiate mortality outcomes.

Methods

We included 3,484 patients undergoing PCI. PRS for coronary artery disease was categorized into low (bottom 20%), intermediate (20–80%), and high (top 20%) groups. BA was estimated using KDMAge, Light BioAge, and PhenoAge, and BAA was defined as the residual from regressing BA on chronological age. BA acceleration was analyzed as continuous, binary, and quartile-based variables. Multivariable Cox models were used to assess associations with all-cause mortality within each PRS stratum, and interaction terms were tested to evaluate effect modification by genetic risk.

Results

During a median follow-up of 10 years, the cumulative incidence of all-cause mortality did not differ significantly across the three genetic risk groups (P > 0.05). In multivariable Cox regression analyses, neither the intermediate-risk group (HR = 1.03, P = 0.804) nor the high-risk group (HR = 1.14, P = 0.839) showed a significant independent association with all-cause mortality. These findings suggest that, in patients undergoing PCI, traditional genetic susceptibility stratification has limited discriminatory ability for long-term survival outcomes.

Interaction analysis of genetic modulation effects demonstrated a significant statistical interaction between KDM-defined BAA and PRS (P for interaction = 0.024). Stratified analyses further showed that in the low genetic risk group, BAA was consistently associated with a significantly increased risk of all-cause mortality (HRcontinuous = 1.26; HRbinary = 1.81; HRQ4 vs. Q1 = 2.00; all P < 0.05). Similar associations were observed in the intermediate genetic risk group (HRcontinuous = 1.22; HRbinary = 1.44; HRQ4 vs. Q1 = 2.05; all P < 0.05). However, in the high genetic risk group, the predictive effect of BAA was completely attenuated (all P > 0.05).

Although analyses based on LightBioAge and PhenoAge did not demonstrate statistically significant interactions, their trends were consistent with those observed for KDM, with all three forms of BAA showing significant predictive value for all-cause mortality in the low- to intermediate-risk groups (Table 1).

Conclusion

In patients undergoing PCI, traditional PRS showed limited ability to discriminate mortality risk. In contrast, biological age acceleration provided significant prognostic value, particularly in low- to intermediate-genetic risk groups. These findings support integrating biological aging assessment with genetic risk to improve post-PCI risk stratification.