Inverse relationship between education and amyloid burden in individuals with subjective cognitive decline plus and mild cognitive impairment

Abstract

AbstractBackgroundHigher educated patients with MCI can tolerate more neuropathology than lower educated patients with similar clinical impairment. It is not known whether this observation also accounts for potential preclinical stages of AD, namely subjective cognitive decline plus (SCD+).MethodData of 197 SCD+ individuals, 227 MCI and 157 AD patients were included, which were collected as part of the AMYPAD‐DPMS cohort (https://amypad.eu/). First, median education in years was computed across the AD‐spectrum groups for each of the 8 European sites. Next, using a median split, the SCD+, MCI and AD cohort were separately categorized into a higher and lower educated group, excluding subjects with median education. Afterwards, the higher and lower educated AD‐spectrum groups were matched for age, sex and cognitive function (MMSE) using propensity score matching in R, leading to the following sample (low/high education): 54/54 SCD+, 70/81 MCI and 56/65 AD patients. Global amyloid load was compared between education groups using Centiloid (CL) information derived from Flutemetamol and Florbetaben PET scans. To confirm the results of the stringent group comparison, partial correlations between years of education and CL values correcting for age, sex and tracer type were performed for each AD‐spectrum group. All analyses were conducted in SPSS 28 and significance level was set to p < .05.ResultHigher educated SCD+ subjects presented significantly (p <. 001) lower CL values (M(CL) = 16.48) than lower educated SCD+ subjects (M(CL) = 32.17), whereas the opposite effect (p < .046) was observed in the MCI cohort with higher educated MCI patients presenting greater CL values (M(CL) = 57.08) compared to the lower educated MCI group (M(CL) = 41.74). No difference was found in the AD group. Results were further confirmed by the correlation approach across the unmatched sample yielding a negative association between years of education and CL values in the SCD+ group (r = ‐.159, p = .027) and a positive association in the MCI group (r = .153, p = .022).ConclusionThese results indicate that sensitivity to early amyloid accumulation may possibly increase with higher education in potential preclinical stages of AD, whereas in clinical stages of AD higher education may support compensatory mechanisms against amyloid burden.