Abstract
I am very pleased to be part of a debate about the serum free light chain (FLC) assay. I am also encouraged that we all agree that measuring monoclonal proteins is difficult and that it is impossible to measure FLCs with the same accuracy and precision as complement C3! The FLC assays (κ and λ) and the derived κ/λ ratio consistently show coefficient of variations (CV) in UKNEQAS of over 50% and often over 100%. A completely normal serum was distributed last year and the stated method group CVs ranged between 6.7% and 146% for free kappa and 11.3% and 42.1% for free lambda. I remain completely astonished that these results are generated by reagents from one company using reagents sold for the various platforms and calibrated using a common standard. The Binding Site may prefer not to debate the analytical issues but they cannot be ignored.
The issue of clinical sensitivity of the FLC approach versus urine analysis for Bence Jones protein has also been raised. Hill et al. 1 did report 15 urine samples showing monoclonal kappa or lambda yet only 11 of the matched sera showed abnormal kappa/lambda ratios hence 4/15 (26.6%) false-negatives. They went on to classify three of these as ‘potential’ false-negatives by a combination of the urinary Bence Jones protein being at low concentration (approximately 50 mg/L), it disappearing on subsequent analysis or the patient having no clinical indications of B-cell malignancy. The guidelines for the analysis of Bence Jones protein 2 state that the lower limit of detection for Bence Jones protein should be below 10 mg/L so even small amounts of Bence Jones protein (approximately 50 mg/L) should not be readily dismissed. A large study of patients with monoclonal gammopathies of undetermined significance reported the cumulative probability of progression to a B-cell malignancy was 10% at 10 years and 21% at 20 years 3 suggesting that ‘follow-up’ needs to be long term rather than a one-off.
The Binding Site puts much emphasis on the importance of urine analysis (for Bence Jones protein) and they with Hill and colleagues suggest that we should use the FLC assay because of the ‘reality of not getting urines’. So two questions. What is the value of serum FLC assays if we do get a urine? And then the tricky one – just how hard is it to get a urine sample? We can CT or MRI scan patients and biopsy their bone marrow with (frighteningly) large needles … but we cannot get a urine sample? What is the risk to the patient of collecting a urine? Is it more difficult to get urine for Bence Jones protein than it is to get urine for albumin or for microscopy, culture and sensitivity or for organic acids? I think that we should educate the clinical staff to send urines when investigating suspected B-cell malignancy and specifically request them when we, the trained laboratory professionals, consider it relevant. We do not need to introduce another test.
The evidence for the clinical benefit of serum FLC analysis seems to be restricted to the monitoring of patients with non-secretory myeloma and amyloid 4 – a tiny proportion of the workload of most labs running serum and urine electrophoresis. The Binding Site gives us FLC sound bites of ‘measurement identified every patient with a disease requiring treatment’ and FLC ‘could replace urinary BJP analysis without loss of diagnostic sensitivity’ but this does NOT equate to significantly improved diagnostic sensitivity (or specificity).
We have hard decisions to make at a local level but also in the wider health-care community. If we introduce something new it should be better than current practice and offer the patients clinical benefit. In June last year, National Institute for Health and Clinical Excellence (NICE) reported that bortezomib (Velcade) could be used for the treatment of myeloma in certain situations. One of the conditions of this treatment is that the manufacturer rebates the full cost of bortezomib for patients whose disease has not demonstrated a complete or partial response after four cycles. 5
We are scientists and analytical issues are at the core of our practice – if there is any valid reason to measure free light chains, the analytical issues will remain a subject of debate. Urine analysis is important to the diagnosis and management of patients with B-cell malignancies, so we should ensure that we get a urine sample – in reality, it is not difficult. I am uncertain of the clinical benefit of the FREELITE assay in the majority of patients but maybe the Binding Site should consider a rebate the full cost of the FREELITE assay in patients where the tests did not contribute to diagnosis or monitoring.