Performance characteristics of an ECL immunoassay for the quantitation of serum periostin

Introduction: Periostin is under investigation as a biomarker to select patients likely to benefit from treatment with lebrikizumab, an IL-13-targeted monoclonal antibody. A clinical trial version of the Elecsys ® Periostin assay was used in the lebrikizumab Phase IIb LUTE/VERSE trials to stratify patients by baseline serum periostin, based on a 50ng/ml cut-off established in the Phase II study MILLY. We present technical performance data for the final Elecsys ® Periostin assay used in the pivotal lebrikizumab Phase III LAVOLTA trials. Methods: Limits of blank [LoB], detection [LoD] and quantitation [LoQ], linearity, interferences and sample stability were evaluated according to CLSI guidelines. Precision was assessed at 3 US sites, using 3 reagent lots (each site used 2 different lots) and a sample panel (11 human serum pools, 3 controls) identical across sites. Results: The specified values for LoB (2ng/ml), LoD (4ng/ml) and LoQ (10ng/ml) were met. The assay was linear throughout the measuring range (10–160ng/ml) with recoveries within ±10% of target at concentrations >30ng/ml and ±3ng/ml at ≤30ng/ml. Recovered periostin concentrations were within ±10% of target for 43 potentially interfering substances and drugs. Samples were stable across various storage conditions and durations (8h at room temperature, 7d at 4°C, 12w at –20°C, and 3 freeze/thaw cycles). Repeatability over sites and lots (N=150) varied across samples and controls from 0.9–1.5% CV. Intermediate precision was 1.2–1.7% CV and reproducibility was 1.7-3.1% CV. Conclusion: The final version of the Elecsys ® Periostin immunoassay exhibits analytical performance that makes it ideally suitable for clinical testing of serum periostin.