Axonal degeneration and amyloid pathology predict cognitive decline beyond cortical atrophy

Table 3 Multivariate analysis. Multivariate linear regression model with the residual measure as dependent variable and the variables statistically significant in bivariate analyses (Table 2) as independent variables, controlling for presence/absence of subjective cognitive decline, baseline mean whole brain cortical thickness, and baseline mPACC5. Standardized beta coefficients with bootstrapped 95% CIs are presented

	Standardized beta coefficient (95% CI)	P value
Age at baseline	0.02 (−0.10 to 0.16)	0.768
Male sex	0.01 (−0.13 to 0.11)	0.813
ICV	−0.10 (−0.21 to 0.03)	0.108
WML volume	0.003 (−0.10 to 0.11)	0.960
APOEe4 allele	−0.04 (−0.15 to 0.06)	0.405
Abnormal CSF Aβ42/40	−0.11 (−0.23 to −0.004)	0.049
CSF P-tau181	−0.03 (−0.19 to 0.12)	0.687
CSF NfL	−0.20 (−0.34 to −0.05)	0.009
CSF GFAP	−0.01 (−0.11 to 0.12)	0.920

Abbreviations: CI confidence interval, ICV intracranial volume, WML white matter lesion, Aβ β-amyloid, P-tau phosphorylated tau, NfL neurofilament light, GFAP glial fibrillary acidic protein, mPACC5 modified Preclinical Alzheimer’s Cognitive Composite 5

ISSN: 1758-9193