Highlights
- Plasma p-tau217 is a highly specific biomarker for amyloid-β pathology, yet its concentrations are significantly influenced by peripheral biological factors like renal clearance and hemodynamics.
- Standard single-cutoff strategies for p-tau217 yield high false-positive rates in patients with chronic kidney disease (CKD) and anemia, and potential false-negatives in obesity.
- Subgroup-specific optimal cutoffs significantly improve diagnostic accuracy, particularly for CKD patients, increasing accuracy from 0.65 to 0.83 in large multicenter cohorts.
- A double-cutoff (three-zone) strategy minimizes false classifications but necessitates confirmatory imaging for 12%–39% of the population, presenting a trade-off between accuracy and diagnostic cost.
Background
The landscape of Alzheimer’s disease (AD) diagnostics has been transformed by the emergence of high-performance blood-based biomarkers (BBMs). Among these, phosphorylated tau 217 (p-tau217) has emerged as the most robust plasma indicator of brain amyloid-β (Aβ) pathology, mirroring the precision of cerebrospinal fluid (CSF) analysis and positron emission tomography (PET). However, as p-tau217 moves from research settings into diverse clinical populations, a critical challenge arises: the influence of non-neurological comorbidities on protein concentrations.
Plasma levels of p-tau217 do not only reflect brain pathology but are also subject to peripheral physiological processes, including hepatic metabolism, renal clearance, and volume of distribution. For instance, reduced renal function leads to diminished clearance of small proteins, potentially elevating p-tau217 levels in the absence of increased brain pathology. Similarly, variations in body mass index (BMI) affect the blood volume into which biomarkers are diluted, while anemia may alter the assay’s performance or reflect underlying systemic fragility. Addressing these confounders is essential for the equitable and accurate implementation of AD biomarkers in real-world primary and secondary care settings.
Key Content
Mechanistic Insights: Why Comorbidities Matter
To interpret plasma p-tau217 correctly, clinicians must understand the tau life cycle. Once tau is released from neurons into the interstitial fluid, a fraction enters the blood. The steady-state concentration in plasma is a balance between this influx and the peripheral clearance. The kidney is a primary site for the filtration and degradation of tau fragments. Consequently, in patients with chronic kidney disease (CKD), particularly those with an estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m², p-tau217 levels can be disproportionately high relative to the brain’s amyloid burden. This creates a risk of “biological false positives” where a patient is misdiagnosed with AD due to renal impairment rather than neurodegeneration.
BMI also plays a significant role through the “dilution effect.” Individuals with a high BMI (obesity) typically have a larger plasma volume, which can lead to lower measured concentrations of p-tau217 even if brain production is high. Conversely, underweight individuals may show artificially elevated levels. Anemia, characterized by low hemoglobin, serves as another potential confounder, possibly reflecting altered protein transport or general metabolic shifts that affect the circulating proteome.
The Development of Biologically Informed Cutoffs
Recent evidence, specifically from the K-ROAD study groups and multicenter analyses (Yun et al., 2026), compared three distinct diagnostic approaches across cohorts using various assays (UGOT Simoa, Roche Elecsys, and C2N %p-tau217):
1. Standard Single Cutoff Strategy
Applying a universal cutoff to all patients regardless of health status is the simplest approach but the least accurate in comorbid populations. In the UGOT cohort, the standard cutoff achieved an accuracy of only 0.65 in patients with CKD and 0.80 in those with anemia. This failure rate is unacceptably high for a screening tool intended to determine eligibility for expensive and potentially side-effect-prone anti-amyloid therapies.
2. Subgroup-Specific Optimal Cutoff Strategy
By adjusting the threshold for specific biological groups, researchers found a dramatic improvement in performance. For CKD patients, raising the cutoff to account for reduced clearance improved diagnostic accuracy to 0.83. Similar adjustments in patients with anemia (improving accuracy to 0.86) demonstrate that “normalizing” the threshold based on the patient’s physiological profile can mitigate the impact of peripheral confounders.
3. The Double-Cutoff (Three-Zone) Strategy
This approach utilizes two thresholds: a lower one with high sensitivity (to rule out pathology) and an upper one with high specificity (to rule in pathology). Patients falling in the “intermediate zone” are referred for confirmatory PET or CSF testing. While this strategy consistently reduced false classifications across all subgroups (CKD, obesity, and anemia), it comes at the cost of diagnostic efficiency. Between 12% and 39% of patients required secondary, more invasive, and expensive testing, which may not be feasible in resource-limited settings.
Evidence by Therapeutic Class and Assay Performance
The findings across different p-tau217 platforms (Simoa vs. Elecsys vs. C2N) show remarkable consistency. While the absolute values vary between assays, the relative impact of CKD and BMI remains a shared challenge. Interestingly, the p-tau217 ratio (e.g., C2N’s %p-tau217) is often considered more resistant to dilution effects (BMI) because both the phosphorylated and non-phosphorylated forms are affected equally by volume changes. However, even with ratio-based assays, the evidence suggests that biological subgrouping provides an additional layer of diagnostic security.
Expert Commentary
The clinical implementation of plasma p-tau217 must move toward a “personalized cutoff” paradigm. Experts in the field argue that a one-size-fits-all approach ignores the basic tenets of renal and systemic physiology. The data from the K-ROAD study suggest that for patients with CKD, the optimal cutoff strategy is not only more accurate but also more cost-effective than the double-cutoff approach because it reduces the need for unnecessary confirmatory PET scans without sacrificing diagnostic integrity.
Conversely, in obesity, the double-cutoff strategy appears superior. Because obesity can mask pathology through dilution, having a high-sensitivity lower threshold ensures that early-stage AD patients are not missed, while the upper threshold maintains the specificity required for treatment initiation. A major controversy remains: how to integrate these varied cutoffs into laboratory information systems (LIS). Automation will be required to pull eGFR, BMI, and hemoglobin data from a patient’s electronic health record to automatically suggest the most appropriate p-tau217 threshold for that specific individual.
Conclusion
The shift toward blood-based biomarkers for Alzheimer’s disease is a monumental step forward, but its success depends on rigorous validation in medically complex populations. This review highlights that kidney function, BMI, and anemia are significant determinants of plasma p-tau217 levels. Moving forward, clinical guidelines must evolve to include biologically informed thresholds. Specifically, optimal subgroup cutoffs should be prioritized for patients with renal impairment to avoid false positives, while double-cutoff strategies may be best suited for obese populations to prevent false negatives. Future research should focus on longitudinal data to determine if these adjusted cutoffs better predict clinical progression and treatment response across diverse global populations.
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;83(3):269-279. PMID: 41627837.
- Ashton NJ, et al. Plasma p-tau217 a better predictor of Alzheimer’s disease pathology than p-tau181 and p-tau231. Nat Commun. 2021;12(1):2936. PMID: 34011953.
- Janelidze S, et al. Head-to-head comparison of 8 plasma amyloid-β 42/40 assays in Alzheimer disease. JAMA Neurol. 2021;78(11):1375-1382. PMID: 34515744.
