Biological Age Outpaces Chronological Age in Predicting Cerebral Small Vessel Disease Progression

Background

Cerebral small vessel disease (CSVD) represents one of the most significant contributors to vascular cognitive impairment, accounting for approximately 25% of all strokes worldwide and contributing substantially to dementia burden in aging populations. The condition manifests through characteristic neuroimaging findings including white matter hyperintensities (WMH), lacunes, cerebral microbleeds (CMBs), enlarged perivascular spaces, and brain atrophy. While chronological age remains the strongest risk factor for CSVD, substantial heterogeneity exists among individuals of the same age, suggesting that biological aging processes may better capture the underlying pathophysiology driving vascular brain injury.

Traditional risk stratification has relied heavily on chronological age, cardiovascular risk factors, and imaging findings. However, the concept of biological age, derived from circulating biomarkers, physiological measurements, and clinical indicators, offers a more nuanced approach to understanding individual susceptibility to age-related cerebrovascular damage. This study investigates whether biomarker-based biological age provides complementary predictive information beyond chronological age regarding CSVD presence and progression.

Study Design and Population

This investigation utilized data from the population-based Polyvascular Evaluation for Cognitive Impairment and Vascular Events (PRECISE) cohort, a well-characterized community-dwelling population designed to examine vascular contributions to cognitive decline. Participants underwent comprehensive baseline brain magnetic resonance imaging (MRI) between 2017 and 2019, with follow-up imaging conducted between 2022 and 2024.

The study employed three established methods for estimating biological age. The Klemera-Doubal method (KDMAge) combines multiple biomarkers using regression-based algorithms to generate a biological age estimate. PhenoAge calculates biological age based on clinical chemistry markers and physical measures that capture phenotypic aging. Homeostatic dysregulation (HDAge) assesses biological age through the magnitude of deviation from reference values across multiple physiological parameters.

Progression of CSVD was systematically evaluated through standardized neuroimaging protocols assessing development of new lacunes, incident CMBs, progression of enlarged perivascular spaces in the basal ganglia (BG-EPVS), and advancement of white matter hyperintensity burden. The residual values representing the difference between biological and chronological age were used in regression analyses to determine associations with CSVD outcomes.

Key Findings

The study enrolled 3,050 participants at baseline, with a mean age of 61.22 ± 6.67 years and 53.51% female representation. Notably, 2,662 participants (87.3%) completed follow-up MRI examinations over a median follow-up period of 4.7 years, while 388 participants were lost to follow-up.

The principal finding demonstrated that higher biological age residual, particularly KDMAge and PhenoAge residuals, was significantly associated with greater CSVD burden at both baseline and follow-up examinations. This association persisted after adjustment for chronological age and conventional cardiovascular risk factors.

Most clinically relevant, KDMAge residual showed significant positive associations with progression of total CSVD burden (odds ratio [OR] = 1.18, 95% CI 1.08-1.30; p = 0.001). The effect was particularly pronounced for new lacune development, with an odds ratio of 1.31 (95% CI 1.13-1.51, p < 0.001). Additionally, KDMAge residual predicted incident cerebral microbleeds (OR = 1.13, 95% CI 1.01-1.25; p = 0.028).

Importantly, no significant associations were observed between KDMAge residual and progression of basal ganglia enlarged perivascular spaces or white matter hyperintensities during the follow-up period. PhenoAge residual demonstrated consistent findings with KDMAge residual across all outcome measures. In contrast, HDAge residual showed no significant associations with progression of CSVD or its individual imaging markers.

Clinical Significance

These findings carry substantial implications for stroke and dementia prevention strategies. The development of lacunes represents covert brain infarction, which despite being clinically silent, contributes significantly to cognitive decline and future stroke risk. The identification of biological age as a predictor of these outcomes suggests that the underlying physiological burden captured by biomarker-based aging measures may reflect vascular damage processes that precede radiographic changes.

The differential associations across CSVD imaging markers are particularly noteworthy. While lacunes and microbleeds showed significant relationships with accelerated biological aging, white matter hyperintensities and enlarged perivascular spaces did not. This pattern suggests that biological age may preferentially capture small vessel pathology affecting penetrating arteries and arterioles rather than diffuse white matter changes. Such specificity could inform targeted screening strategies for individuals with elevated biological age.

Expert Commentary

The study represents a methodologically rigorous contribution to understanding vascular contributions to brain aging. Several strengths merit emphasis: the population-based design enhances generalizability compared to clinic-based cohorts; the longitudinal follow-up with systematic imaging allows assessment of true disease progression rather than cross-sectional associations; and the application of multiple biological age estimation methods enables comparison of predictive validity.

Limitations warrant consideration. The follow-up period of approximately 5 years, while substantial, may underestimate associations with slower-progressing manifestations. The predominantly middle-aged cohort limits extrapolation to older populations where CSVD burden is highest. Additionally, the single-center design, despite population-based recruitment, may introduce regional factors affecting generalizability.

The mechanistic pathways connecting accelerated biological aging to CSVD progression likely involve multiple interrelated processes. Biomarkers incorporated into biological age calculations—including inflammatory markers, metabolic indicators, and vascular risk factors—may collectively reflect endothelial dysfunction, blood-brain barrier compromise, and small vessel arteriolosclerosis. The Klemera-Doubal and PhenoAge methods may capture these processes more comprehensively than homeostatic dysregulation assessment alone.

Conclusion

This landmark cohort study demonstrates that biomarker-based biological age residuals, particularly KDMAge and PhenoAge, independently predict progression of cerebral small vessel disease over approximately five years of follow-up. The associations with incident lacunes and microbleeds highlight the potential for biological age assessment to identify individuals at elevated risk for clinically relevant cerebrovascular damage before overt symptoms develop.

These findings support the incorporation of biological age estimation into stroke and dementia risk stratification algorithms. Individuals with biological age substantially exceeding their chronological age may benefit from intensified neuroimaging surveillance, aggressive vascular risk factor management, and early intervention strategies. Future research should validate these findings in more diverse populations with extended follow-up periods and explore targeted interventions for those with accelerated biological aging profiles.

Funding

Study supported by the National Natural Science Foundation of China and related Chinese research funding agencies. Clinicaltrials.gov registration applicable.

References

Wang M, Cai X, Gao P, Yang Y, Ding Y, Zhang Z, Sun J, Zhang Y, Liu D, Wang Y, Wang Y, Pan Y. Associations of Accelerated Biological Aging With the Presence and Longitudinal Progression of Cerebral Small Vessel Disease. Neurology. 2026-04-10;106(9):e214818. PMID: 41962120.