A Multimodal Latent Severity Axis for Alzheimer’s Disease: Probabilistic PCA, Bayesian Trajectories, and Stage-Aware Timing Effects

INTRODUCTION: Multimodal Alzheimer's disease (AD) cohorts capture cognition, function, neuroimaging, and fluid biomarkers, yet overall disease severity remains difficult to summarize on a single clinically meaningful scale. The apolipoprotein E ε4 (APOE ε4) allele is the strongest common genetic risk factor for late-onset AD, but its association with "progression" has been inconsistent because earlier placement along the disease continuum is often conflated with faster within-stage decline. METHODS: Using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), we analyzed an amyloid-positive baseline cohort (N = 1058) and a longitudinal subset (N = 932; ≥ 2 visits and ≥ 4 of 13 measures per visit). Measures included standardized cognitive and functional assessments, structural and functional neuroimaging, cerebrospinal fluid biomarkers of amyloid‑beta and tau pathology, and plasma neurofilament light protein as a marker of neuroaxonal injury. Magnetic resonance imaging (MRI) volumes were adjusted using amyloid-negative cognitively normal controls with quadratic age and intracranial volume terms. Probabilistic principal component analysis (PPCA) was used to derive a latent severity coordinate, defined as the first principal component (PC1). Hierarchical Bayesian random-intercept and random-slope models were used to estimate individual trajectories, partition APOE ε4 effects into baseline severity and within-stage rate, and generate genotype-stratified ages at prespecified severity landmarks. Axis stability was assessed with 100 bootstrap refits, and predictive performance was assessed with participant-level fivefold cross-validation. RESULTS: The PC1 explained 38.7% of baseline variance and produced a clinically interpretable multimodal severity axis. Stability was high across bootstrap refits, and residual association with age was minimal after MRI volume adjustment. Higher APOE ε4 dose was associated with greater baseline latent severity, whereas within-stage rate differences were smaller than the baseline severity-position effect. A latent symptomatic landmark was reached approximately 3.0-3.3 years earlier per ε4 allele. Adding APOE improved out-of-sample prediction by about 10% without loss of calibration. CONCLUSIONS: Probabilistic principal component analysis provides a stable, multimodal, biologically informed severity axis for longitudinal modeling in amyloid-positive ADNI. Within this framework, APOE ε4 was associated primarily with latent severity position and model-implied timing along the continuum, whereas within-stage rate differences were smaller. These findings support stage-aware longitudinal inference and methodological applications within this cohort, while external clinical calibration and validation remain necessary.