Highlight
Research published in JAMA Neurology indicates that the diagnostic performance of plasma phosphorylated tau 217 (p-tau217) for Alzheimer’s disease is significantly influenced by systemic biological factors, including kidney function, body mass index (BMI), and anemia status.
Applying a subgroup-specific optimal cutoff improved diagnostic accuracy for amyloid-β (Aβ) positivity in patients with chronic kidney disease (CKD) from 0.65 to 0.83, addressing a major limitation in clinical biomarker application.
While a double-cutoff strategy reduces false classifications, it necessitates confirmatory imaging for up to 39% of patients, suggesting that subgroup-specific thresholds may offer a more cost-effective solution for patients with kidney dysfunction and anemia.
Introduction: The Evolution of Plasma Biomarkers
The field of Alzheimer’s disease (AD) diagnostics has undergone a paradigm shift with the emergence of high-sensitivity plasma biomarkers. Among these, plasma p-tau217 has demonstrated exceptional utility in identifying brain amyloid-β (Aβ) pathology, often rivaling the accuracy of cerebrospinal fluid (CSF) analysis and positron emission tomography (PET). However, as these assays transition from controlled research environments to diverse clinical populations, the impact of non-neurological comorbidities on biomarker concentrations has become a critical concern.
Plasma p-tau217 levels are not solely reflective of cerebral pathology; they are also influenced by peripheral clearance and metabolic factors. Previous observations have suggested that renal impairment, variations in BMI, and hematological conditions like anemia can alter the steady-state concentration of tau species in the blood. Without accounting for these variables, clinicians risk misinterpreting results, leading to false positives in patients with poor renal clearance or false negatives in those with high BMI. This study critically evaluates whether personalized or stratified cutoff strategies can maintain the diagnostic integrity of p-tau217 across these biological subgroups.
Study Design and Methodology
This multicenter cross-sectional study, conducted between 2016 and 2023 with analyses finalized in 2025, utilized data from multiple memory clinics and community-based cohorts. The primary objective was to compare three distinct classification strategies for detecting Aβ positivity (defined as a Centiloid score ≥25.5 on PET):
1. Standard Single Cutoff
A uniform threshold applied to all participants regardless of their biological profile.
2. Subgroup-Specific Optimal Cutoff
Tailored thresholds adjusted for specific biological subgroups, including those with CKD (eGFR <60 mL/min/1.73 m²), underweight status (BMI <18.5), obesity (BMI ≥27.5), and anemia (hemoglobin <12 g/dL in women, <13 g/dL in men).
3. Double-Cutoff Strategy
The use of two thresholds to create an “intermediate zone.” Results falling within this zone are considered inconclusive and require confirmatory PET or CSF testing.
The study analyzed three distinct assay platforms: the UGOT Simoa (n=2571), the Roche Elecsys (n=1578), and the C2N %p-tau217 ratio (n=304). This multi-platform approach ensured that findings were robust across different technological methodologies (immunoassays vs. mass spectrometry-based ratios).
Impact of Biological Confounders on p-tau217
The researchers identified significant shifts in p-tau217 concentrations based on the participants’ systemic health. In patients with CKD, the reduced glomerular filtration rate leads to diminished peripheral clearance of tau fragments, resulting in elevated plasma levels that do not necessarily correlate with increased brain amyloid. Conversely, variations in BMI can affect the volume of distribution, while anemia may alter the protein binding and transport dynamics of tau in the blood.
In the UGOT cohort, the standard single cutoff performed poorly in the CKD subgroup, yielding an accuracy of only 0.65 (95% CI, 0.57-0.72). This suggests that using a generic threshold in a population with renal impairment would lead to an unacceptable rate of false-positive Alzheimer’s diagnoses. Similarly, accuracy was compromised in patients with anemia (0.80) when using the standard approach.
Results: Comparing Classification Strategies
The implementation of biologically informed thresholds showed marked improvements in diagnostic metrics. In the UGOT cohort, the optimal cutoff strategy raised diagnostic accuracy in the CKD subgroup to 0.83 (95% CI, 0.76-0.89). Similar gains were observed in the anemia subgroup, where accuracy rose to 0.86. These results were consistently replicated in the Roche cohort, reinforcing the validity of subgroup-specific adjustments.
The double-cutoff strategy also demonstrated superior accuracy compared to the single cutoff across all subgroups. By isolating ambiguous cases in an intermediate zone, this method significantly reduced the risk of misclassification. However, this precision comes at a cost of clinical efficiency: between 12% and 39% of participants fell into the intermediate category, necessitating follow-up with more invasive or expensive confirmatory tests.
Economic Performance and Cost-Effectiveness
A crucial aspect of this study was the economic modeling of these diagnostic pathways. The optimal cutoff strategy proved particularly advantageous for patients with CKD. By adjusting the threshold upward to account for reduced clearance, the model achieved higher accuracy than the double-cutoff approach while simultaneously lowering total diagnostic costs by reducing the need for unnecessary PET scans.
For patients with anemia, the double-cutoff strategy provided a slight edge in accuracy, but this was offset by the economic burden of confirmatory imaging in 25% of cases. In the context of obesity, the double-cutoff remained the superior choice for both accuracy and cost-efficiency, likely due to the complex ways BMI interacts with plasma volume and protein concentrations.
Expert Commentary and Clinical Implications
These findings have immediate implications for the clinical rollout of blood-based biomarkers. As p-tau217 moves toward becoming a primary screening tool in primary care and memory clinics, clinicians must be equipped with the tools to interpret these results within the context of the patient’s overall health. A “one-size-fits-all” cutoff is clearly inadequate for a significant portion of the elderly population, who frequently present with CKD or anemia.
The study supports a move toward “personalized biomarker interpretation.” Automated laboratory reporting systems could eventually incorporate eGFR and BMI data to provide a risk-adjusted interpretation of p-tau217 levels. However, experts note that while these adjustments improve accuracy, they do not entirely eliminate the need for clinical correlation and, in some cases, confirmatory imaging. The high percentage of intermediate results in the double-cutoff strategy highlights the ongoing necessity of maintaining access to PET and CSF diagnostics for complex cases.
Limitations and Future Directions
While the study is comprehensive, limitations remain. The cross-sectional design prevents conclusions about the longitudinal predictive value of these adjusted cutoffs. Furthermore, while the UGOT and Roche platforms showed consistent results, the sample size for the C2N %p-tau217 ratio was smaller, requiring further validation. Future research should also investigate whether other factors, such as liver function or specific medication use, further confound these measurements.
Conclusion
The transition of p-tau217 from a research curiosity to a clinical standard requires a sophisticated understanding of its biological limitations. This study provides a roadmap for improving the diagnostic accuracy and economic viability of Alzheimer’s screening. By adopting subgroup-specific optimal cutoffs or double-cutoff strategies, healthcare systems can ensure that the promise of blood-based biomarkers is realized for all patients, regardless of their comorbid conditions. Biologically informed thresholds represent the next step in the journey toward precision medicine in neurology.
References
Yun J, Lee J, Shin D, et al. Plasma Phosphorylated Tau 217 Cutoffs for Amyloid Pathology and Kidney Function, Body Mass Index, and Anemia. JAMA Neurol. 2026; doi:10.1001/jamaneurol.2025.5530.