The associations between ATN biomarkers and episodic memory diminish as age increases

The relationship between age, biomarkers, and cognitive function at advanced ages remains underexplored. By utilizing a naturalistic memory clinic cohort, we aimed to clarify how the A/T/N biomarkers are associated with verbal episodic memory performance and the moderating effect of age.

This cross-sectional study included 676 patients from memory clinics in the Stockholm metropolitan region, comprised of 141 with Alzheimer's disease dementia, 403 with mild cognitive impairment, and 132 with subjective cognitive impairment. We used weighted least-squares regressions to evaluate the association between episodic memory and the biomarkers and the moderating effect of age.

There was a significant interaction between age and A (Aβ42), where the negative impact of Aβ42 abnormality on Rey Auditory Verbal Learning (RAVLT) delayed recall was reduced with higher age (β = 0.22, p < 0.001). In addition, a corresponding interaction effect was observed between age and N (MTA) on RAVLT delayed recall (β = 0.13, p < 0.01). Additional exploration analyses showed that at age 80 or older, the ATN biomarkers showed weaker associations with episodic memory performance.

Although interpretations should be conducted with caution, these findings raise questions about the justification for assessing CSF in patients older than 80 years, given its weak associations with core clinical measures, such as learning and free recall of episodic memory. Further clinical studies are needed to validate these results and also evaluate other core cognitive measures. The naturalistic nature of the memory clinic population, with its clinically driven diagnoses, may help clarify these findings.