Background The soluble biomarker sub-committee of OMERACT has published validation criteria for biomarkers reflecting structural damage end-points. Within these, categories of Discrimination and Feasibility relate to assay reproducibility and performance. 14–3-3eta (η) is a joint-derived biomarker that drives joint damage processes and informs radiographic progression independently of acute phase reactants. As an ELISA assay, 14–3-3η is currently available for clinical use as a laboratory developed test (LDT) in the United States (US) and as an in-vitro diagnostic (IVD) in Canada. The ELISA is (CE) marked for Europe and is Therapeutics Good Administration (TGA) approved for Australia. Previous studies have described 14–3-3eta assay equivalence at independent laboratories in the US and Canada.
Objectives The purpose of this study was to demonstrate the reproducibility of 14–3-3eta measurements at two independent laboratories in Japan and one in Canada.
Methods A total of 212 samples from the University of Occupational and Environmental Health were provided for this study. Serum 14–3-3eta measurements were performed using the 14–3-3eta ELISA provided by Augurex. Testing in Canada occurred in August 2015; these 14–3-3eta values were set as the “standard”. The 212 samples were shipped from Canada to Japan in January 2016 with testing being performed in February 2016. Testing at MBL was performed on a blinded-basis. Upon completion of testing, results were sent to the investigators at which point in time, MBL was unblinded. 14–3-3eta positivity was defined at ≥0.19ng/ml. Contigency and Spearman analyses were performed to assess the strength of the results between the two testing centres. Values above the linear range of the assay i.e. ≥20ng/ml were excluded for the Spearman analyses and determination of the median (IQR). A p-value of <0.05 established statistical significance.
Results In Canada, of the 212 samples tested, 146 (68.8%) were 14–3-3eta positive and in Japan, 147 (69.3%) were positive. In Canada, 187 patients had reportable values in the linear range and in Japan 186 did. Median 14–3-3eta levels in Canada were 0.51 ng/ml (IQR: 0.11–2.09) and in Japan they were 0.58 ng/ml (IQR: 0.02–2.68), respectively. Spearman correlation analysis revealed a highly significant correlation between the two testing sites, r=0.92; p<0.00001. Contingency analysis revealed a strong concordance between the two sets of results delivering a likelihood ratio (LR) of 156, p<0.0001. Between the two sites, 58 of the samples were 14–3-3eta negative and 139 were positive; there was agreement in 197 (93%) of the 212 samples.
Conclusions As defined by the soluble biomarker sub-committee of OMERACT, reliability and standardization of biomarker assessments in routine clinical practice across the globe and in clinical studies is of high importance. The results presented herein demonstrate that the 14–3-3eta assay is highly reliable, that testing/delivery of results can be standardized between two independent laboratories, and that it is very stable over time and when shipped between continents.
Disclosure of Interest S. Hirata: None declared, M. Kaneda Employee of: Medical & Biological Laboratories, T. Isayama Employee of: Medical & Biological Laboratories, Y. Gui Employee of: Augurex Life Sciences Corp., A. Marotta Employee of: Augurex Life Sciences Corp., W. Maksymowych Consultant for: Augurex Life Sciences Corp, Y. Tanaka: None declared