Decomposing Heterogeneity in Disease Progression Speeds and Pathways

Understanding why patients with the same diagnosis exhibit markedly different disease progression--some progressing rapidly, others slowly, and through distinct symptom patterns--remains a major challenge in medicine. Here, we developed a machine learning framework called DiSPAH (Disease-progression Speed and Pathway Analysis based on a Hidden Markov model) to estimate both the pathway and speed of disease progression in individual patients. DiSPAH models disease progression as transitions of latent states evolving over continuous time with a patient-specific progression speed. We applied DiSPAH to longitudinal clinical scores from an amyotrophic lateral sclerosis (ALS) cohort and successfully inferred each patients hidden disease trajectory and progression speed. These individualized dynamics were significantly associated with baseline clinical features and enabled prediction of future disease course from data available at the first clinical visit. Our results highlight that jointly modeling progression pathway and speed improves prediction of heterogeneous disease courses, offering a powerful tool for personalized care and research in ALS and other chronic conditions. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=104 SRC="FIGDIR/small/26345194v1_ufig1.gif" ALT="Figure 1"> View larger version (22K): org.highwire.dtl.DTLVardef@fb65ccorg.highwire.dtl.DTLVardef@d861b7org.highwire.dtl.DTLVardef@1f7670dorg.highwire.dtl.DTLVardef@18e95d9_HPS_FORMAT_FIGEXP M_FIG Graphical Abstract: Schematic illustration of the computational framework proposed in this paper C_FIG