Answers are grounded in this paper and cite their sources.
To the Editor: In a recent study by Little et al. (1), the story of C-peptide standardization was left somewhat uncertain with the problem of 2 reference methods providing different C-peptide concentrations. Although very well correlated (r2 = 0.986), there was a bias of approximately 20% between the Diabetes Diagnostic Laboratory (DDL)1 method (operated in 2 different laboratories in the US) and the National Metrology Institute of Japan (NMIJ; y = 0.7824x − 0.0202). There were differences in the procedures and in the labeled internal standards (ISs). The NMIJ method was traceable to the NMIJ Certified Reference Material (6901-b) (2), whereas the DDL method was traceable to the IS developed in Dr. Stein's Laboratory in New York (3). Subsequent experiments have resolved this issue. Our first step was to verify that the DDL method was traceable to the NMIJ Certified Reference Material 6901-b, which was chosen as the primary reference material for the traceability scheme. Serum with zero C-peptide was enriched with native NMIJ reference material in a serial dilution and analyzed as routine samples by the DDL LC-MS method. The theoretical vs measured C-peptide results matched (y = 1.0016x + 0.239, r2 = 0.9996); therefore, both methods appeared to be traceable to the NMIJ Certified Reference Material. Our next step was to ship a new set of both pooled (n = 10) and single-donor (n = 17) samples from the DDL to NMIJ along with the DDL IS that had been developed by Stein et al. (2). As in the original studies, each sample was analyzed in duplicate or triplicate by each method, and the mean result was used for data analysis. While our original comparison showed that NMIJ results were approximately 20% lower, the current comparison (Fig. 1) showed virtually no bias between NMIJ and DDL with use of the same IS. In addition, sample analyses in the DDL were performed with the original instrument (SCIEX API 4000) as well as a new SCIEX QTRAP 6500+. Clearly, when the same ISs were used by each laboratory, the same results were obtained regardless of the specific method and instruments used. We therefore hypothesize that there was a possible interference with the NMIJ IS that was used in the initial comparison. Circles indicate analyses by DDL API 4000 and triangles indicate analyses by DDL QTRAP 6500+. Closed symbols and open symbols correspond to pooled and single-donor samples, respectively. We conclude that either method can be used in the traceability scheme as long as the choice of IS is validated by use of serum samples. Diabetes Diagnostic Laboratory National Metrology Institute of Japan internal standard. R. Little gratefully acknowledges the support of NIH/NIDDK (Grant Number 1UC4DK096587-01). We also thank Dr. Daniel Stein for his help with providing labeled internal standard.
No takes yet. Share an insight, caveat, or question.
Little et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: