Analytical and Clinical Validation of a High Accuracy Fully Automated Digital Immunoassay for Plasma Phospho-Tau 217 for Clinical Use in Detecting Amyloid Pathology

ABSTRACT Objectives Phospho-tau 217 (p-Tau 217) in plasma has emerged as a promising biomarker for detecting amyloid pathology in plasma as an aid in diagnosis of Alzheimer’s disease. Because of its low abundance in plasma, reliable quantitation in blood is challenging. Digital single molecule array (Simoa) technology provides higher analytical sensitivity than possible with conventional analog immunoassays, enabling precise measurement of plasma p-Tau 217. We report the analytical and clinical validation of a Simoa digital immunoassay for p-Tau 217 across highly diverse clinical cohorts that meets consensus criteria for confirmatory test performance to aid the diagnosis of Alzheimer’s disease in people with cognitive impairment. Methods A Simoa p-Tau 217 assay utilizing a 2-cutoff approach was analytically validated using industry standard protocols, and diagnostic thresholds were clinically validated across 2 clinically diverse independent cohorts (total n = 873) using amyloid PET or CSF biomarkers as comparators. Results The assay exhibited acceptable analytical characteristics, including analytical sensitivity enabling measurement of plasma p-Tau 217 in all clinical samples and a lower limit of quantitation of 0.006 pg/ml. For samples not in the intermediate zone between the 2 cutoffs, the assay gave a clinical sensitivity of 90.3%, a specificity of 91.3%, and overall accuracy of 90.7%. 30.9% of the samples fell into the intermediate zone between the lower and upper cutoffs (0.04 and 0.09 pg/mL respectively). With an amyloid prevalence of 50%, typical for mild cognitive impairment, the positive and negative predictive values were 91.4% and 90.4% respectively. Conclusions The analytical characteristics are suitable for implementation of the Simoa p-Tau 217 assay as a lab developed test under the Clinical Laboratory Improvement Act (CLIA), and the clinical performance characteristics meet consensus criteria for a confirmatory plasma test to aid in Alzheimer’s diagnosis. RESEARCH IN CONTEXT Evidence before this study Following rapid advances in plasma testing for AD-related biomarkers, the field has coalesced around p-Tau 217 as the most accurate single blood-based biomarker for detecting the presence of amyloid pathology. There have been a number of recent reports and conference presentations on different tests for plasma p-Tau 217 including large head-to-head comparisons. To date, however, none of these reports examine the analytical and clinical performance characteristics of the test--or the methodologies by which these characteristics were determined--in sufficient detail for a critical evaluation of the fitness of the test for its intended clinical use. Indeed, it has been the lack of robust analytical and clinical validation reporting that has led some to question if some of these tests are ready for prime time. Added value of this study Our work is the first to describe in detail the analytical and clinical performance characteristics of a high-accuracy laboratory developed test for plasma p-Tau 217 validated for clinical use. While the test is not currently FDA cleared, the validation studies were conducted with study protocols recommended by the FDA that generally exceed the standards required of LDTs developed and offered for clinical use under the Clinical Laboratory Improvement Act (CLIA). This report helps advance the field by describing the level of validation that is compatible with the FDA’s Final Rule (May 2024, FDA-2023-N-2177) in which LDTs were included as regulated in vitro diagnostic devices. Detailed validation reports like this can facilitate critical evaluation of whether a currently available LDT is fit for use as demonstrated by its validation robustness. More rigorous validation of these tests represents an important next step on the path toward widespread adoption and use of blood-based biomarkers for AD.