P-499: DETECTION OF BENCE JONES PROTEINURIA BY SERUM FREE LIGHT CHAINS: DELIMITATION OF 24H URINE STUDIES IN PATIENTS WITH MONOCLONAL GAMMOPATHIES.

Introduction: Bence Jones proteins (BJP) are free light chains of monoclonal immunoglobulins that can appear in urine of patients with monoclonal gammopathies (MG). Currently, detection of BJPs is performed by 24h urine collection, which is inconvenient for the patient, and analysis by proteinogram and immunofixations, which is complex and costly for the laboratory.
The aim of this study is to investigate the diagnostic performance of proteinuria and measurement of serum free light chains (sFLC) in the detection of BJP in order to optimize 24h urine collection.

Methods: Retrospective data of laboratory analysis were collected from 206 serum and urine samples of 44 patients with MM who had measured proteinogram, immunofixation and protein quantification in 24-hour urine, as well as serum measurement of sFLC.
To validate the results obtained, data from 1020 serum and urine samples from patients who underwent the same determinations in routine laboratory activity during the year 2022 were evaluated.
Urine proteinogram and immunofixation were performed on Hydrasis (Sebia), quantification of sFLC on an Optilite using Freelite™ reagent (Binding Site). Statistical analysis was performed using MedCalc v.18.2.

Results: Total urine protein (0.115 g/l), sFLC κ/λ ratio (monoclonality Lambda < 0.82 and >1.99 monoclonality Kappa) and difference of involved and uninvolved free light chain (dFLC) (>11.93 mg/l) cutoffs were established to predict presence of BJP.
Urine protein obtained an AUC=0.656, sensitivity (S)=60.7%, specificity (E)=70.6% and a false negative (FN) rate of 23%. sFLC κ/λ ratio showed a better ability to detect BJP, both to detect Bence Jones kappa (BJK): AUC= 0.913, S=91.5%, E=91.1% and FN=3.39%, as well as Bence Jones Lambda (BJL): AUC=0.890, S=90.2%, E=87.7% and FN=2.42. Usefulness of dFLC obtained AUC= 0.831, S=87.9%, E=78.4% and FN=7.76%.
The results of the validation of these cutoffs to detect BJP with the largest sample showed that, if the protocol based on sFLC κ/λ ratio had been applied the analysis of 55% of the urine studies would have been avoided; whereas with the dFLC based it would be 45%, achieving similar S results and less than 5% of FN.

Conclusions: We propose an algorithm that allows 45-55% of 24h urine studies to be avoided based on sFLC, a more sensitive and a faster alternative to 24h urine testing for BJP screening.
Both sFLC κ/λ ratio and dFLC analysis achieve high sensitivity in the detection of BJP with a low false negative rate, which allows them to be used as optimal screening test.
Use of dFLC allows working with a single cut-off point, simplifying its application.
Recently, important studies have appeared that support the inclusion of sFLC to replace 24h urine for the assignment of response in MM and modified response IMWG criteria. Until clinical guidelines change, it is necessary to simplify urinary testing.