Plasma Phosphorylated Tau 217 Cutoffs for Amyloid Pathology and Kidney Function, Body Mass Index, and Anemia

Abstract

Importance Plasma phosphorylated p-tau 217 levels vary with biological factors such as kidney dysfunction, body mass index (BMI), and anemia. It remains unclear whether a biological subgroup-specific optimal cutoff or a double-cutoff strategy could enhance diagnostic accuracy and cost efficiency beyond the standard single cutoff. Objective To compare the diagnostic and economic performance of 3 plasma p-tau217 classification strategies for detecting amyloid-beta (A beta) positivity: standard single cutoff, subgroup-specific optimal cutoff, and double cutoff. Design, Setting, and Participants This cohort study was a multicenter cross-sectional study conducted from 2016 to 2023; analyses were completed in 2025. Participants were recruited from multiple memory clinics and community-based cohorts. All participants had amyloid positron emission tomography (PET) imaging, clinical evaluation, and p-tau217 testing with measures of estimated glomerular filtration rate (eGFR), BMI, and hemoglobin. Measurements of p-tau217 were made using UGOT Simoa and Roche Elecsys, and the %p-tau217 ratio was assessed using a tau multianalyte assay (C2N Diagnostics LLC). Exposures Kidney function (chronic kidney disease [CKD], eGFR <60 mL/min/1.73 m(2); advanced CKD, eGFR <45 mL/min/1.73 m(2)), underweight (BMI <18.5), obesity (BMI >= 27.5), and anemia (hemoglobin <12 g/dL in women, <13 g/dL in men). Main Outcomes and Measures Plasma p-tau217 concentration; standard single cutoff, optimal cutoffs, and double cutoff for A beta positivity (Centiloid >= 25.5); accuracy; and cost-effectiveness estimated from false-positive, false-negative, and confirmatory imaging costs. Results A total of 2571 participants were analyzed with UGOT, 1578 with Roche, and 304 with the C2N %p-tau217 ratio. The mean (SD) age was similar across cohorts (71.3 [8.6] years in the UGOT cohort, 71.3 [8.5] years in the Roche cohort, and 71.8 [7.8] years in the C2N cohort); there were 1633 (63.5%), 1006 (63.8%), and 191 (62.8%) women and 938 (36.5%), 572 (36.2%), and 113 (37.2%) men, respectively. In the UGOT cohort, the optimal cutoff improved diagnostic accuracy compared with the standard single cutoff, particularly in CKD and anemia (CKD: from 0.65; 95% CI, 0.57-0.72; to 0.83; 95% CI, 0.76-0.89; anemia: from 0.80; 95% CI, 0.76-0.84; to 0.86; 95% CI, 0.82-0.90), with consistent findings in the Roche cohort. In all biological subgroups, the double-cutoff strategy also increased accuracy relative to the single cutoff and reduced false classifications but yielded 12% to 39% intermediate results. When compared directly, the optimal cutoff provided higher accuracy than the double cutoff for CKD while lowering total diagnostic costs. For anemia, the double cutoff showed slightly higher accuracy but required confirmatory PET in up to 25% of cases, offsetting its economic advantage. In obesity, the double cutoff remained superior for both diagnostic accuracy and cost efficiency. Conclusions and Relevance This study found that both the optimal cutoff and the double-cutoff strategy outperformed the standard single cutoff, each showing distinct strengths across subgroups. These findings support biologically informed thresholds to improve diagnostic accuracy and cost efficiency when implementing plasma p-tau217.