Why does plasma p-tau217 show higher accuracy for amyloid-PET (AUC=0.95) than AD diagnosis (AUC=0.81)?

This performance discrepancy suggests p-tau217 may detect amyloid pathology before clinical AD manifestation, but the temporal relationship and underlying biology remain unclear. Resolving this could optimize biomarker interpretation for early intervention.

Gap type: unexplained_observation Source paper: High-sensitivity plasma proteomics reveals disease-specific signatures and predictive biomarkers of Alzheimer’s disease phenotypes in a large mixed dementia cohort. (2025, Research square, PMID:40678230)

{“resolution_pipeline”: “scidex.atlas.gap_closure_pipeline”, “task_id”: “f4f7b129-0f43-4c84-abd8-20d4e701842d”, “evaluated_at”: “2026-04-28T19:10:51.494823+00:00”, “resolution_summary”: “Resolved by hypothesis h-793f9f273d: Temporal Decoupling of CSF p-tau217 Normalization from Amyloid PET Negativity Creates a Dual Threshold Stopping Rule. Supporting evidence includes debate sess_SDA-2026-04-08-gap-pubmed-20260406-062202-5c32c50a_task_9aae8fc5.”, “match_counts”: {“hypothesis_matches”: 1, “debate_matches”: 5, “paper_matches”: 0}, “hypothesis_matches”: [{“id”: “h-793f9f273d”, “title”: “Temporal Decoupling of CSF p-tau217 Normalization from Amyloid PET Negativity Creates a Dual Threshold Stopping Rule”, “score”: 0.312, “reason”: “9 token overlaps; entity overlap: pet”, “analysis_id”: null, “target_gene”: “NA - Biomarker validation hypothesis”, “target_pathway”: null, “disease”: “neurodegeneration”, “composite_score”: 0.595, “confidence_score”: 0.58, “status”: “proposed”, “pubmed_evidence_ids”: [“30679382”, “30895603”, “33283854”, “34310832”, “37340743”]}], “debate_matches”: [{“id”: “sess_SDA-2026-04-08-gap-pubmed-20260406-062202-5c32c50a_task_9aae8fc5”, “title”: “AD patients with TDP-43 pathology show worse cognitive impairment, but how TDP-43 mechanistically contributes to this severity is unknown. Understanding this could identify TDP-43 as a therapeutic target for cognitive preservation in AD.\n\nGap type: unexplained_observation\nSource paper: TDP-43 Pathology in Alzheimer’s Disease. (2021, Mol Neurodegener, PMID:34930382)”, “score”: 0.388, “reason”: “9 token overlaps; entity overlap: pmid”, “analysis_id”: “SDA-2026-04-08-gap-pubmed-20260406-062202-5c32c50a”, “quality_score”: 0.734, “status”: “completed”, “target_artifact_id”: null, “target_artifact_type”: null}, {“id”: “sess_SDA-2026-04-07-gap-pubmed-20260406-062207-e4ce5cf0_task_9aae8fc5”, “title”: “The abstract mentions that pathological seeds have different characteristics and conformations, but the underlying molecular mechanisms that generate this diversity remain unclear. Understanding these mechanisms is critical for developing targeted therapeutic interventions.\n\nGap type: unexplained_observation\nSource paper: Protein transmission in neurodegenerative disease. (2020, Nat Rev Neurol, PMID:32203399)”, “score”: 0.363, “reason”: “9 token overlaps; entity overlap: pmid”, “analysis_id”: “SDA-2026-04-07-gap-pubmed-20260406-062207-e4ce5cf0”, “quality_score”: 0.65, “status”: “completed”, “target_artifact_id”: null, “target_artifact_type”: null}, {“id”: “sess_SDA-2026-04-15-gap-pubmed-20260411-084510-fbfafe2c_20260416-033316”, “title”: “This study reveals SYNGAP1 expression and function in radial glia before synaptogenesis, contradicting its classification as purely a synaptic protein. The molecular mechanisms underlying this non-synaptic role remain unexplained, which is critical for understanding ASD pathophysiology.\n\nGap type: contradiction\nSource paper: Non-synaptic function of the autism spectrum disorder-associated gene SYNGAP1 in cortical neurogenesis. (None, None, PMID:37946050)”, “score”: 0.351, “reason”: “9 token overlaps; entity overlap: pmid”, “analysis_id”: “SDA-2026-04-15-gap-pubmed-20260411-084510-fbfafe2c”, “quality_score”: 0.78, “status”: “completed”, “target_artifact_id”: null, “target_artifact_type”: null}, {“id”: “sess_SDA-2026-04-06-gap-pubmed-20260406-041439-ec89b1e4_task_9aae8fc5”, “title”: “While single-cell sequencing reveals microglial heterogeneity across regions, ages, and sexes, the functional consequences of this diversity remain unclear. Understanding these differences could explain variable disease patterns and inform personalized therapeutic approaches.\n\nGap type: open_question\nSource paper: Beyond Activation: Characterizing Microglial Functional Phenotypes. (2021, Cells, PMID:34571885)”, “score”: 0.351, “reason”: “9 token overlaps; entity overlap: pmid”, “analysis_id”: “SDA-2026-04-06-gap-pubmed-20260406-041439-ec89b1e4”, “quality_score”: 0.73, “status”: “completed”, “target_artifact_id”: null, “target_artifact_type”: null}, {“id”: “sess_SDA-2026-04-07-gap-pubmed-20260406-062202-094b44bf_task_9aae8fc5”, “title”: “TDP-43 inclusions occur in AD, ALS, and FTLD but the pathogenic mechanisms leading to TDP-43 pathology may differ between diseases. Understanding disease-specific drivers could reveal why TDP-43 shows limbic distribution in AD versus other patterns in ALS/FTLD.\n\nGap type: unexplained_observation\nSource paper: TDP-43 Pathology in Alzheimer’s Disease. (2021, Mol Neurodegener, PMID:34930382)”, “score”: 0.344, “reason”: “8 token overlaps; entity overlap: pmid”, “analysis_id”: “SDA-2026-04-07-gap-pubmed-20260406-062202-094b44bf”, “quality_score”: 0.697, “status”: “completed”, “target_artifact_id”: null, “target_artifact_type”: null}], “paper_matches”: []}