Introduction: Screening for multiple myeloma requires both serum and urine protein electrophoresis, because in about 20% of patients with myeloma, monoclonal free light chain (FLC) is the only paraprotein found, and it is commonly missed by serum protein electrophoresis. However, many requests for testing do not include a urine sample (>80% of requests in our experience). This risks missing clinically significant disease. Recent availability of serum FLC assays has raised the possibility that these assays may replace testing for urinary FLC in screening for monoclonal gammopathies, and that the serum kappa:lambda light chain ratio (LCR) may be more sensitive for detecting monoclonal FLC than serum and urine protein gel electrophoresis.


  • To identify how many additional patients with monoclonal gammopathies would be detected if serum FLC assays were incorporated into the routine myeloma screen.

  • To evaluate the ability of serum FLC assays to identify all patients identified by urine protein electrophoresis.

Method and Setting: We analysed data from a consecutive cohort of 753 serum blood samples submitted for myeloma screening to Hull Royal Infirmary Immunology Laboratory between 03/23/07 and 05/31/07. During this period all myeloma screen requests received serum capillary zone protein electrophoresis (CE) (SEBIA Capillarys 2, Analytical Technologies) and serum FLC analysis using a latex-enhanced immunoassay (The Binding Site, Birmingham, UK on a Beckman-Coulter IMMAGE nephelometric analyzer). When available, urine protein CE was also perfomed (SEBIA Capillarys 2). Samples with an abnormal serum CE or serum LCR were tested by immunofixation (SEBIA Hydrasys, Analytical Technologies). Repeat samples were requested from patients with LCR outside the reference interval (0.26–1.65) before referral, but an immediate hematology referral was recommended if LCR >3.5 sd from the mean (ie 0.18–2.01).

Results: Of 753 patients, 118 had features on serum CE requiring immunofixation. Of these, 76 had a monoclonal paraprotein identified. A further 46 samples had normal serum CE with abnormal LCR and 25 of these had LCR outide mean±3.5 sd. Of 6 patients so far referred as a result of abnormal LCR but normal serum CE, 4 (67%) had a lymphoproliferative disease (free kappa myeloma, free kappa MGUS, free lambda MGUS, and a chronic lymphocytic leukaemia). Urine samples were received from 128 (17%) patients, of whom 8 (6%) had a monoclonal FLC identified in the urine. All of these patients had an abnormal serum LCR, though in one patient with acute renal failure and raised kappa and lambda results the LCR was borderline abnormal (1.75), with a very small band in the urine, visible only by agarose gel immunofixation. For the 2 patients with normal serum CE, but with urinary monoclonal FLC present, serum LCR was abnormal (LCR= 33 and 1.75).

Discussion and Conclusions:

  • Use of serum FLC assays increased the detection of monoclonal paraproteins (by 5% so far but further follow up is required to quantify this exactly).

  • Serum FLC analysis had a sensitivity of 100% for identifying patients with urinary FLC.

Author notes

Disclosure:Employment: Ewan Robson is an employee of The Binding Site, manufacturer of the free light chain assays used in the study. Research Funding: The Binding Site provided free light chain assay kits free of charge for this research and funded travel for the presenting author to attend ASH 2007.