Plasma p-tau217 demonstrates high diagnostic accuracy for Alzheimer's disease across diverse clinical cohorts and biomarker definitions

A comprehensive meta-analysis involving 19,652 participants from clinical or biomarker-defined Alzheimer's disease cohorts evaluates the diagnostic utility of plasma p-tau217. The study aggregates data to determine sensitivity, specificity, likelihood ratios, and diagnostic odds ratios against various reference standards. Results indicate that p-tau217 achieves an 85.4% sensitivity for biomarker-defined Alzheimer's disease, with a 95% prediction interval ranging from 81.4% to 88.7%. Specificity for this definition stands at 88.0%, with a prediction interval of 85.1% to 90.6%. These metrics suggest the biomarker effectively identifies true cases while minimizing false positives in a broad population.

The analysis further examines performance regarding amyloid and tau positivity. For amyloid positivity, sensitivity reaches 87.3% and specificity is 85.5%. Regarding tau positivity, sensitivity is 84.9% while specificity is notably high at 93.8%. When compared against clinical Alzheimer's disease diagnoses, sensitivity drops to 72.9%, yet specificity remains strong at 89.5%. These variations highlight the importance of defining the reference standard when interpreting test results in practice.

Likelihood ratios provide additional context for clinical decision-making. The positive likelihood ratio is 7.13, indicating that a positive test result significantly increases the probability of disease. Conversely, the negative likelihood ratio is 0.167, suggesting that a negative result substantially lowers the probability of disease. The diagnostic odds ratio is calculated at 42.7, reflecting a strong overall discriminative ability of the biomarker across the included studies.

Secondary outcomes explore correlations with cognitive decline, frailty, and behavioral impairment. While specific correlation coefficients are not detailed in the primary results, the biomarker's association with disease progression supports its potential for monitoring disease trajectory. The study acknowledges heterogeneity across assays, populations, and reference definitions as key limitations. Some studies utilized optimized cut-offs, which may affect generalizability to different clinical settings.

Publication bias assessments were limited in power, necessitating cautious interpretation of the pooled estimates. Funding sources and potential conflicts of interest were not reported for the included studies. Despite these limitations, the practice relevance is clear: plasma p-tau217 is well-suited as a triage biomarker to guide confirmatory testing. However, widespread clinical implementation should await further large, prospectively designed studies with standardized assays and externally validated thresholds.

Clinicians should view these results as supportive evidence rather than definitive proof for immediate universal adoption. The high specificity is particularly valuable for ruling out disease in patients with negative results, reducing unnecessary imaging or invasive procedures. The moderate sensitivity for clinical diagnosis suggests that some patients with Alzheimer's disease may still test negative, requiring alternative diagnostic pathways. Future research must address assay heterogeneity to ensure consistent performance across different laboratories.

In conclusion, plasma p-tau217 represents a significant advancement in blood-based biomarkers for Alzheimer's disease. Its performance characteristics support its role in stratifying patients for confirmatory testing. While not a standalone diagnostic tool, it offers a practical first-line screening option that aligns with current guidelines for early detection. Continued validation in diverse populations will be essential to fully realize its potential in clinical workflows.