Specificity of a High-Sensitivity Cardiac Troponin I Assay Using Single-Molecule–Counting Technology

Cardiac troponin is the standard for diagnosis or exclusion of acute myocardial infarction. Recent guidelines recommend a cutoff value at the upper 99th percentile (99th%)1 of values for a reference population of healthy individuals and assay imprecision (CV) ≤10% at this cutoff (1). Many commercial assays cannot meet these criteria, and values below the 99th% cutoff appear to supply diagnostic and prognostic information (2)(3). Recently the Erenna™ cardiac troponin I (cTnI) immunoassay (Singulex) has been shown to provide sensitivity and precision that meet these goals (4), with a preliminary 99th% cutoff value at 8 ng/L and 10% CV of 1.8 ng/L (5). This novel assay uses single-molecule counting technology and has been previously described. However, we could not fully exclude the possibility that, with such sensitive limits of detection, low-level nonspecific binding (NSB) events might contribute to cTnI measurements. We explored this issue by testing potential sources of NSB events. To test for NSB of serum or plasma constituents, cTnI was quantified in 4 specimen types from 20 healthy individuals, serum, EDTA, lithium heparin, and sodium-citrated plasma. The protocol was approved by the University of California Committee on Human Research. All 20 individuals were free of cardiac disease or symptoms (13 female, 7 male, mean age 43 years, range 23–64 years) and provided signed written consent. Blood was centrifuged within 30 min of collection, divided into aliquots, and stored frozen at −70 °C until analysis at Singulex with the Erenna system (4)(5). Values of cTnI were quantifiable in 93% of specimens (Table 1 ), and no trends based on age or sex were observed. For each donor, the mean value assessed for each specimen type fell within 2 SDs of the mean, indicating no significant differences between specimen types (95% CI). For 1 donor increased cTnI occurred in all specimen types. Grub’s test results (P < 0.05) indicated that this donor was a statistical outlier, and all data from this donor were excluded. No further clinical data on this individual are available at this time. The combined adjusted mean (SD) value for cTnI was 2.78 (1.93) ng/L. cTnI concentrations across specimen types and capture antibody cross-reactivity in healthy human specimens analyzed with the Singulex MP-based cTnI assay. This serum sample from a 64-year-old woman was identified as an outlier and removed. ND, nondetectable; LoD, limit of detection; DNQ, detectable but nonquantifiable; LLoQ, lower limit of quantitation. One serum sample displayed 55% cross-reactivity with amyloid-β-42 (1.26 ng/L) compared to cTnI (2.36 ng/L). cTnI concentrations across specimen types and capture antibody cross-reactivity in healthy human specimens analyzed with the Singulex MP-based cTnI assay. This serum sample from a 64-year-old woman was identified as an outlier and removed. ND, nondetectable; LoD, limit of detection; DNQ, detectable but nonquantifiable; LLoQ, lower limit of quantitation. One serum sample displayed 55% cross-reactivity with amyloid-β-42 (1.26 ng/L) compared to cTnI (2.36 ng/L). To test capture antibody NSB, we obtained 8 additional serum samples from healthy blood bank donors (4 male, 4 female, mean age 25.5 years, range 22–30 years). Frozen serum was assayed as described previously (4)(5), with the following changes. Microparticles (MPs) were coated with one of the following nonspecific capture antibodies, amyloid-β-42 (Covance), macrophage inflammatory protein-1 α (R&D Systems), granulocyte–colony-stimulating factor (R&D Systems), or prostate-specific antigen (BiosPacific). As a positive control MPs were coated with a cTnI capture antibody (R&D Systems), and as a negative control MPs were left uncoated (blank MPs). Additional experimental details are available on request. Across all 28 donors, cTnI was detectable in all (limit of detection 0.2 ng/L) and quantifiable in 93% (lower limit of quantitation 1 ng/L) of specimens (Table 1 ). Serum and plasma from only 1 volunteer (5%) were quantifiable with a nonspecific antibody, which displayed 55% cross-reactivity with amyloid-β-42 (1.26 ng/L) compared to cTnI (2.36 ng/L). No other NSB events were quantifiable. Some low-level NSB (<lower limit of quantitation and >limit of detection) was observed for blank MPs (7%) and for nonspecific capture antibodies amyloid-β-42 (5%), macrophage inflammatory protein-1 α (25%), and prostate-specific antigen (13%). No NSB was observed with granulocyte–colony-stimulating factor. Analyte NSB events were tested with standard curves and linear regression analysis of cTnI standard (NIST) and nonspecific skeletal troponin (Hytest) assayed over 0.1–100 ng/L. Back-interpolated standard curves of the NIST standard cTnI (y = 1.053x – 0.229; R2 = 0.999) correlated poorly with skeletal troponin (y = 0.000x + 0.156; R2 = 0.001), indicating an absence of NSB when skeletal troponin was used as the analyte. Interassay precision was determined by independent assay of 2 control sera (14 runs over 4 days), and showed interassay imprecisions (CVs) of 7% and 10%, with mean cTnI measurements of 8.3 and 2.2 ng/L respectively. These data support the specificity of this novel single-molecule assay for quantification of cTnI in apparently healthy persons. Even in individuals with documented NSB, binding was below the limit of quantitiation in all but 1 case (Table 1 ), in which it accounted for 55% of the cTnI signal, a value that would not have moved this specimen out of the normal range (4)(5). As more sensitive detection techniques are developed, such studies will be essential to confirm the specificity of detection. Cross-reactivity studies that evaluate troponin orthologs (i.e., skeletal muscle troponin I and troponin T) and other NSB events (matrix, antibodies, and MPs) are critical for the validation of these assays. These data, taken together with those reported previously (4)(5) provide strong evidence that this particular cTnI assay maintains the specificity of cTnI detection. Similar data will be necessary for other “high-sensitivity” troponin I assays. Author Contributions:All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data or analysis, and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors’ Disclosures of Potential Conflicts of Interest:Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest: Employment or Leadership: S.J. Agee, Singulex; Q.A. Lu, Singulex; J. Todd, Singulex. Consultant or Advisory Role: A.H.B. Wu, Singulex; A.S. Jaffe, Siemens, Beckman-Coulter, Singulex, Critical Diagnostics, Bayer, and Novartis. Stock Ownership: S.J. Agee, Singulex; Q.A. Lu, Singulex; J. Todd, Singulex. Honoraria: None declared. Research Funding: A.H.B. Wu, Singulex; A.S. Jaffe, Beckman-Coulter. Expert Testimony: None declared. Role of Sponsor: The funding organizations played a role in the design of study, choice of enrolled patients, review and interpretation of data, preparation or approval of manuscript. Acknowledgments: We would like to thank Douglas Held, Ph.D. for preparation of reagents.