Sex and Education Modify the Association Between Subjective Cognitive Decline and Amyloid Pathology

BackgroundSubjective cognitive decline (SCD) may be an early risk factor for dementia, particularly in highly educated individuals and women. This study examined the effect of education and sex on the association between SCD and Alzheimers disease (AD) biomarkers in non-demented older adults. MethodVanderbilt Memory and Aging Project participants free of clinical dementia or stroke (n=156, 72{+/-}6 years, 37% mild cognitive impairment, 33% female) completed fasting lumbar puncture, SCD assessment, and Wide Range Achievement Test-III Reading subtest to assess reading level at baseline as a a proxy for educational quality. Cerebrospinal fluid (CSF) biomarkers for AD ({beta}-amyloid42 (A{beta}42), A{beta}42/40 ratio, phosphorylated tau (p-tau), tau, and neurofilament light (NfL)) were analyzed in batch. Linear mixed effects models related SCD to CSF AD biomarkers and follow-up models assessed SCD x sex, SCD x reading level, and SCD x education interactions on AD biomarkers. ResultIn main effect models, higher SCD was associated with lower A{beta}42 and A{beta}42/40 ratio (p-values<0.004). SCD was not associated with tau, p-tau, or NfL levels (p-values>0.38). SCD score interacted with sex on A{beta}42/40 ratio (p=0.03) but no other biomarkers (p-values>0.10). In stratified models, higher SCD was associated with lower A{beta}42/40 ratio in men (p=0.0003) but not in women (p=0.48). SCD score interacted with education on A{beta}42 (p=0.005) and A{beta}42/40 ratio (p=0.001) such that higher education was associated with a stronger negative association between SCD and amyloid levels. No SCD score x reading level interaction was found (p-values> 0.51) though significant associations between SCD and amyloid markers were seen in the higher reading level group (p-values<0.004) but not the lower reading level group (p-values>0.12) when stratified by a median split in reading level. ConclusionAmong community-dwelling older adults free of clinical dementia, higher SCD was associated with greater cerebral amyloid accumulation, one of the earliest pathological AD changes. SCD appears most useful in detecting early AD-related brain changes in men and individuals with higher quantity and quality of education. SCD was not associated with CSF markers of tau pathology or neurodegeneration. These findings suggest that considering sex and education is important when assessing SCD in older adults.