Amyloid-β and APOE genotype predict memory decline in cognitively unimpaired older individuals independently of Alzheimer’s disease polygenic risk score

Table 2 Baseline and annual change effects of APOE genotype, PRS and amyloid-β status in memory composite scores

Composite memory score
Model	Fixed effects Baseline and interaction effects with time	β (SE)	p	pFDR
Model 1	APOE APOE × time	-0.02 (0.07) -0.05 (0.02)	0.82 0.008	0.97 0.01
Model 2	AD-PRS AD-PRS × time	-0.002 (0.05) -0.04 (0.01)	0.97 0.006	0.97 0.01
Model 3	Amyloid-β Amyloid-β × time	-0.12 (0.10) -0.08 (0.04)	0.23 0.02	0.72 0.02
Model 4	APOE × AD-PRS APOE × AD-PRS × time	-0.13(0.07) 0.002 (0.02)	0.07 0.94	0.72 0.94
Model 4	APOE AD-PRS APOE × time AD-PRS × time	-0.03 (0.07) 0.002 (0.05) -0.05 (0.02) -0.04 (0.01)	0.66 0.97 0.01 0.008	0.97 0.97 0.03 0.03
Model 5	APOE × Amyloid-β APOE × Amyloid-β × time	-0.11 (0.15) -0.12 (0.05)	0.47 0.02	0.93 0.047
Model 5	Amyloid-β negative group: APOE APOE × time Amyloid-β positive group: APOE APOE × time	0.06 (0.08) -0.02 (0.02) -0.04 (0.18) -0.13 (0.06)	0.44 0.46 0.82 0.03	0.93 0.46 0.97 0.0495
Model 6	AD-PRS × Amyloid-β AD-PRS × Amyloid-β × time	-0.12 (0.10) 0.02 (0.04)	0.24 0.64	0.72 0.94
Model 6	AD-PRS Amyloid-β AD-PRS × time Amyloid-β × time	0.004 (0.05) -0.15 (0.10) -0.04 (0.01) -0.07 (0.04)	0.94 0.14 0.01 0.04	0.97 0.72 0.03 0.05

Relationship between APOE genotype dosage, AD-PRS, and baseline amyloid-β status and longitudinal memory performance. Linear Mixed Models with subject specific random intercepts and fixed slopes, corrected for age, sex, education and center. APOE genotype or PC1-3 were added depending on using APOE genotype or AD-PRS as predictor. Amyloid-β status was based on visual read [¹⁸F]flutemetamol PET standardized uptake value images. Abbreviations: amyloid-β = amyloid-beta; APOE = apolipoprotein E; FDR = False Discovery Rate; p = p value; PRS = polygenic risk score; PC = principal components; SE = standard error

ISSN: 1471-2377