SUBJECTIVE COGNITIVE DECLINE VERSUS SUBTLE DEFICITS AS INDICATORS OF PRECLINICAL AD: RESULTS FROM THE MULTICENTER OBSERVATIONAL STUDY ON PREDEMENTIA ALZHEIMER’S DISEASE (DELCODE) STUDY

Abstract

In preclinical AD, subtle neuropsychological deficits occur in groups without mild cognitive impairment (MCI), although the detection of a subtle decline in individual cases may be difficult. Subjective cognitive decline (SCD) has been emphasized as being one of the first symptomatic manifestations of preclinical AD, possibly even preceding subtle deficits because compensation of slight impairments is still possible. Subjective and subtle cognitive decline are both mentioned in the NIA-AA-criteria-2018 as indicators for the Alzheimer's continuum in cognitively normals, but their overlap and their possibly complementary contribution in this regard is unclear. By applying an operational definition for subtle cognitive decline in SCD patients, we aim to clarify the degree of overlap of these definitions and their association with CSF-AD biomarkers. We analyzed 240 memory-clinic patients with a worrisome subjective cognitive decline (SCD) and 209 controls (CON) without SCD. Both groups performed within the normal range on a comprehensive test battery (>-1.5SD). Presence of subtle deficits was defined by mild neuropsychological or functional impairments (Table1). Prevalence of subtle impairment was calculated in SCD and CON to gauge the extent of overlap. With logistic regression, we analyzed whether subjective decline (SCD vs. CON) would predict CSF-AD markers, controlling for age. We repeated this analysis after excluding all subjects with subtle deficits. Most CON and SCD participants had no subtle cognitive impairments and groups didn't differ significantly in this regard (Table1). Subjective decline was associated with amyloid load (SCD>CON; p<.05) in cases with and without subtle impairment. Subtle impairment was associated with amyloid in CON and SCD (Figure1). SCD were 2.979 more likely Aβ42-positive compared to CON, even if excluding all subjects with subtle decline from analysis (Exp(B)=2.762).