Machine Learning-Based Multiclass Classification of Cognitive Stages Using Plasma Biomarkers, Clinical Assessments, and Genetic Features: A Repeated Nested Cross-Validation Study in ADNI with External Validation in CNTN

Background: Plasma biomarkers are widely promoted as scalable tools for staging Alzheimer’s disease (AD), yet head-to-head comparisons against the clinical scales used to define diagnostic labels remain scarce. Reported gains from machine-learning fusion of clinical and biomarker features may therefore reflect label circularity rather than biological signal. Methods: From the Alzheimer’s Disease Neuroimaging Initiative (ADNI), we assembled 655 participants (CN = 296, MCI = 168, AD = 191) with concurrent plasma biomarkers (pT217, Aβ42/40, NfL, GFAP), clinical scales (MMSE, CDR-SB, FAQ), APOE genotype, and demographics. Three pre-specified feature sets (clinical-only, biomarker-plus-demographic-genetic, and full fusion) were compared across four classifiers (Logistic Regression, SVM, Random Forest, XGBoost) using repeated nested cross-validation (5-fold × 3 outer, 5-fold inner) with balanced class weighting. External validation used the Center for Neurodegeneration and Translational Neuroscience (CNTN) cohort (n=130). Results: Clinical scales alone reached a three-class AUC-OvR of 0.9539±0.0041, and fusion reached 0.9559±0.0046, an indistinguishable gain. Because MMSE, CDR-SB, and FAQ partly determine ADNI diagnostic labels, both estimates are circularity-inflated upper bounds. Independently of this circularity, the plasma-plus-demographic-genetic model still achieved AUC-OvR =0.7455±0.0150, with pT217 the dominant contributor. Pairwise discrimination was excellent for CN vs. AD (1.0000) and MCI vs. AD (0.9739), but markedly weaker for CN vs. MCI (0.9302 fused, 0.69 plasma-only). The reduced biomarker model transferred to CNTN with AUC-OvR =0.702 (95% CI 0.635–0.764). Conclusions: Apparent fusion gains in ADNI are largely a consequence of label circularity. After removing the circular clinical features, plasma pT217 supports three-class CN/MCI/AD screening at AUC ≈0.74 internally and 0.70 externally, which establishes a realistic performance ceiling for blood-based AD staging. MCI detection remains the principal bottleneck.