Background

In previous studies, plasma neurofilament light chain (NfL) and progranulin (PGRN) levels are associated with cognitive and physical impairment in older individuals. However, evidence of their relationships with frailty is lacking. This study aims to explore the associations of plasma NfL and PGRN levels with frailty in community-dwelling older adults.

Methods

We included 507 older adults (mean [standard deviation] age, 76.7 [4.5] years) with plasma NfL and PGRN measurements from the Multidomain Alzheimer Preventive Trial (MAPT). The timepoint of biomarker measurements, either 12 or 24 months after study enrollment, was defined as the baseline for each participant. Frailty phenotype (robust, pre-frail, and frail) was assessed at 12, 24, 36, 48, and 60 months by Fried's frailty criteria. The cross-sectional associations between plasma neurodegenerative biomarkers and frailty severity were examined using logistic regressions. We further used Cox proportional hazard models to evaluate the associations between plasma biomarkers and incident frailty among robust or pre-frail participants at baseline (n = 403).

Results

At baseline, participants with high plasma NfL levels (>93.11 pg/ml [the upper quartile]) had a higher likelihood of pre-frailty or frailty compared to their normal NfL counterparts (odds ratio = 1.68; 95% confidence interval = 1.10–2.57); however, this association did not remain significant after controlling for covariates. Neither NfL nor PGRN levels showed prospective associations with incident frailty over 4 years.

Conclusions

This study failed to find associations of circulating NfL and PGRN levels with frailty among community-dwelling older adults in adjusted analyses. Whether plasma neurodegenerative markers serve as potential biomarkers of frailty requires further investigation.