Light‐initiated chemiluminescent assay of 17β‐estradiol metrological traceability system established by manufacturer according to ISO17511:2020 and basic performance evaluation performed by clinical end‐users

BACKGROUND: In order to ensure the accuracy of the product, we established 1(st) model of metrological traceability hierarchy for light‐initiated chemiluminescent assay (LICA) of 17β‐estradiol (E(2)) at the manufacturer, based on International Organization for Standardization (ISO) 17511:2020. Moreover, we verified/validated the basic performance (such as matrix effect and long‐term stability of end‐user IVD MD calibrator, precision, linearity interval, accuracy/ trueness, and detection capability) at the clinical end‐user. METHODS: Human serum samples were used in this study. E(2) was detected by mass spectrometry (MS) and LICA. The metrological traceability of LICA for E(2) was established according to ISO 17511: 2020 standards, and pools of human samples were used as the m.3. secondary calibrator. Precision was validated according to Clinical and Laboratory Standards Institute (CLSI) EP05‐A3. The linear interval was verified according to CLSI EP06‐ED2. Comparison of accuracy and trueness of E(2) with MS and Roche according to CLSI EP09‐A3. The detection capability was validated according to EP17‐A2. Matrix effect and long‐term stability evaluation of end‐user IVD MD calibrator were carried out according to CLSI EP14‐A2, EP25‐A. Statistical software was used for data analyses. RESULTS: The use of pools of human samples and fine adjusting calibrators ensured the accuracy of end‐user test results. The metrological traceability of LICA for E(2) was established. It showed excellent precision, meeting the requirements of allowable imprecision (7.5%). The allowable deviation from linearity (ADL) of 5% was allowed to show a good linear interval (12.52–4167.25 pg/ml). The accuracy/ trueness was verified, and relative deviation in the medical decision level met the performance specification of 10.03% compared with MS or Roche. The validated limit of blank, limit of detection, and limit of quantitation of E(2) were 4.95 pg/ml, 8.93 pg/ml, and 9.88 pg/ml, respectively (the allowed imprecision is 20.00%). The interference rate of E(2) ranged from −5.5% to 6.6%. CONCLUSION: LICA showed high sensitivity, high specificity, excellent precision, wide linearity interval, IVD MD calibrator has long‐term stability, and no matrix effect. The metrological traceability of E(2) established by using pools of human samples as M.3. can deliver accuracy to the end‐user IVD MD and show good consistency with MS and Roche.