A-004 Commutability Assessment for high-sensitivity Cardiac Troponin I Reference Material 8121: A human plasma-based material consisting of a set of four relevant hs-cTnI concentrations

Abstract Background Harmonizing/standardizing cardiac troponin I (cTnI) has been a goal for over 20 years. A cTnI standard reference material (SRM2921) was produced in 2004 by the National Institute of Science and Technology. However, SRM2921 is human heart tissue-based and is non-commutable, i.e., cTnI results in human blood do not yield the same numeric relationship across cTnI assays. Thus, a commutable higher-order secondary calibrator is required for harmonizing cTnI clinical assays. A candidate secondary material, coined RM8121, was produced from human plasma as a 4-concentration sample-set (A-D) spanning from a cTnI level below the female 99th percentile upper reference limit (URL) to a 100-fold higher concentration. Methods Here commutability of RM8121 was assessed with hs-cTnI methods on the Access2 and Unicell DxI systems (Beckman Coulter) in accordance with the manufacturer’s instructions. For this study, RM8121(Levels A-D) and clinical samples (CS) from 40 consented individuals having native hs-cTnI levels across the concentration range of RM8121. Commutability was assessed with the method of Miller et al. (Clin Chem 69;966–975:2023) and approved by IFCC’s Scientific Division. To achieve 90% statistical power, a three-run measurement grid was designed that consists of interspersed RM8121 and CS specimens for a grand total of 556 measurements with each hs-TnI assay. Specifically, three sets of RM 8121 (A,B,C&D) were included in each run (n=12), all measured in triplicate for n=36 measurements per run, and a total of n=36X3=108 hs-cTnI measurements of RM8121 with 3 runs per assay. For CS, run 1, included 40 samples, each measured in quadruplicate for n=40X4=160 measurements per assay. Runs 2 and 3 each included 36 CL specimens, measured in quadruplicate for a total n=160+144+144=448 measurements. Thus, the total n=108+448=556 measurements per assay for 90% statistical power. RM8121 and CS hs-cTnI data from the Access 2 versus Unicell DXI devices were plotted and analyzed in a pair-wise comparison with linear regression. Those points within the 95% confidence interval (CI) of the regression line are considered commutable. Analysis also examined the Access2 and DXI hs-cTnI assays and their non-commutability bias plus uncertainty falling within the maximums allowable interval of ±6.5% at the 99th female URL. Results The figure’s left panel shows that the pairwise comparison of the Access2 and DXI methods by linear regression for RM8121 (red dots) and CS (green dots) measured points. All data fell with the 95%CI bounds of the regression line for the hs-cTnI methods. The material is therefore commutable. The figure’s right panel shows that with the exception of one point, all data fell entirely within the maximum allowable non-commutability bias interval of ±6.5%, which includes at the 99th upper reference limit concentration of 14.9ng/L for females, 19.9ng/L for males and 17.9ng/L overall. Conclusion The RM8121 hs-cTnI material, a human based LiHep plasma matrix, is commutable with the Access2 and Unicell DXI hs-cTnI assays. Although this work transparently demonstrates the methodology for evaluating commutability, a full commutability study must be conducted that includes 7 more FDA-cleared hs-cTnI methods before this characteristic is fully assessed.