22 Detecting Antigen Excess in Serum Free Light Chain Measurement. (11th January 2018)

• Record Type: Journal Article
• Authors: Zaydman, Mark; Logsdon, Nicole; Scheller, Katherine; Gronowski, Ann
• Is Part Of: American journal of clinical pathology. Volume 149(2018)Supplement 1
• Page Start: S175
• Page End: S175
• Publication Date: 2018-01-11
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqx149.391
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: The FreeLite assay (Binding Site, Birmingham UK) is a homogeneous immunoassay utilizing polyclonal antibodies to detect serum free light chains (sFLC) by targeting light chain epitopes that are inaccessible in intact immunoglobulins. This assay has clinical utility in diagnosing and prognosticating monoclonal gammopathies, and has largely eliminated the need for a 24-hour urine sample. Unfortunately, the FreeLite assay has been reported to produce falsely low results due to antigen excess in 0.12%-5.40% and 0.0%-1.2% of kappa and lambda results, respectively. We identified one such case by noting discordance between the sFLC and serum immunofixation (iFix) results; specifically, a normal value for kappa sFLC despite a dense monoclonal free kappa band on iFix. The objective of this study was to evaluate two strategies for improving antigen excess detection in the FreeLite assay, ie, a discordance check with iFix and a novel kinetic flag based upon a higher-order derivative (curvature) of the FreeLite reaction data. First, to evaluate the performance of the established linearity and kinetic flags, three months of FreeLite results (1, 332 samples) using the Cobas 501 (Roche Diagnostics) were retrospectively reviewed. The numbers of samples that underwent autodilution, and had discordant sFLC and iFix results were quantified. Next, to prospectively evaluate if discordance with iFix can identify otherwise undetected antigen excess, all consecutive samples within a one-month period that had a monoclonal free light chain on iFix were measured using the FreeLite assay neat and at a 1:100 initial dilution (if not performed automatically). Finally, the curvature-based kinetic flag was tested against serial dilutions of the index patient's serum. In three months, the Cobas instrument had automatically flagged 33.3% and 23.9% of kappa and lambda sFLC results for autodilution and properly corrected for antigen excess in these samples. During this time, 41 and 95 samples for kappa and lambda, respectively, exhibited discordance between sFLC and iFix results. However, in one month of prospective data collection, 44 discordant samples were repeated at the 1:100 dilution without identifying a single additional case of antigen excess. Finally, the curvature-based kinetic flag successfully identified all dilutions of the index patient's serum that exhibited antigen excess. These data are revealing in several ways. First, they illustrate that the Cobas instrument already successfully detects many cases of antigen excess in the FreeLite assay. Second, identifying samples by discordance between sFLC and iFix leads to costly assay repetition and only rarely identifies additional cases of antigen excess. Finally, the curvature-based kinetic flag was able to successfully identify antigen excess in the index case that had evaded the previously established detection methods. Ongoing work will further evaluate this novel kinetic flag using a larger set of patient samples.