Questions and answers about FLCs
From Wikilite
|
30 |
SECTION 6 - Appendices |
| Questions and answers about FLCs |
Contents |
30.1 Questions about urine testing for free light chains
There are several reasons. UPE with scanning of the bands is inaccurate if there is accompanying proteinuria. Urine IFE is non-quantitative but is more sensitive than UPE. Dye uptake tests, dipsticks and many other widely used tests for proteinuria are unreliable as they largely fail to measure small, cationic proteins such as free κ and λ. 24-hour urine collections are frequently not collected reliably so measurements of FLC excretion may be inaccurate.
No. FLC assays will produce direct, quantitative results for uFLCs but are not always more sensitive than UPE tests. It is preferable to measure serum rather than uFLCs because the removal of FLCs by the renal tubules has a huge effect on the amount of FLCs entering the urine. Also, FLC κ/λ ratios in urine are much more variable than in serum because of renal tubular metabolism.
UPE involves scanning gels for FLC bands but these can be obscured or confused by other proteins present in urine. In contrast, FLC assays are specific but will also measure polyclonal FLCs in urine. Polymerisation or fragmentation of FLC molecules will affect measurements by either assay, but to different extents.
Moderate to heavy proteinuria makes interpretation of UPE difficult particularly when obscuring immunoglobulins are present. FLC assays are affected only by increased monoclonal and polyclonal FLC excretion and not by other proteins.
FLC assays measure both polyclonal and monoclonal FLCs. Renal impairment will lead to increases in both polyclonal κ and λ urine FLC concentrations and may produce abnormal urine κ/λ ratios. Large amounts of polyclonal FLCs may be excreted in diseases such as SLE and may prevent low concentrations of monoclonal FLCs from being detected as abnormal κ/λ ratios.
Urinary FLCs immunoassays are more sensitive than urine IFE and are quantitative. Clonality is judged by κ/λ ratios for FLCs and visually by IFE. There is a poor correlation between FLC concentrations and the intensity of staining on IFE because of the IFE antibody. However, clonality is better judged by uIFE than by immunoassays because of variable levels of polyclonal FLCs.
Ladder banding is the term used to describe repeating bands of κ and/or λ seen in concentrated urine samples on IFE gels. It is due to minor differences in charge on polyclonal FLC molecules. The pattern may be confused with, and potentially obscure, genuine monoclonal FLC bands. Ladder banding does not indicate the presence of monoclonal FLCs. uFLC immunoassays are not affected by ladder banding but the presence of polyclonal FLCs will affect measurements of monoclonal FLCs.
There is a poor correlation between serum and uFLC concentrations in individual patients. This is because the kidney metabolises large amounts of FLC molecules, preventing them from entering the urine. There is a much better correlation between changes in serum and uFLC concentrations.
Yes. There is considerable evidence that serum FLC tests alongside serum SPE and IFE identify all significant patients with monoclonal proteins. Urine tests do identify extra monoclonal light chains but there is no evidence that they are clinically significant. As regards sFLC tests replacing urine tests for monitoring MM, the jury is still out. Serum and urine tests are complementary in terms of sensitivity but it is unclear if positive urine results are clinically significant when serum tests are negative.
30.2 Clinical questions about serum testing for free light chains
Monoclonal elevations of free κ and λ occur in the same diseases that produce monoclonal gammopathies with intact immunoglobulins. The number of diseases is extensive but clearly MM and AL amyloidosis are important. A list of diseases associated with monoclonal gammopathies is given in Chapter 29.
Since the kidney can metabolise between 10 and 30g of FLCs per day in the proximal tubules, urinary FLC results do not accurately reflect FLC production by the tumour. This results in a poor correlation between serum and uFLC concentrations and makes urine testing unreliable.
LCMM comprises between 15-18% of all MM. Approximately, half to two thirds of these patients will have abnormalities by SPE. This especially applies to patients who are producing large quantities of FLCs, are in renal failure or have associated hypogammaglobulinaemia. The remaining LCMM patients with normal SPE (5-7%) and 70% of those with NSMM (2%) will be detected by sFLC assays. Most patients with AL amyloidosis and other rare monoclonal gammopathies will also be detected. In addition, normal individuals with FLC MGUS will be identified. Intact immunoglobulin MGUS occurs in approximately 3% of people aged over 70 and more frequently with increasing age. Therefore, at least 10-15% additional patients with monoclonal gammopathies will be detected using sFLCs. The study by Bakshi et al., (Chapter 23) detected an additional 50% of patients and included B-CLL and other plasma cell dyscrasias.
Yes, for two reasons. sFLC analysis will detect monoclonal gammopathies that are missed by IFE. Also, quantification of sFLCs, at the time of clinical presentation provides a base line result for subsequent disease monitoring.
FLC serum assays are more sensitive for FLC detection than both SPE and IFE so approximately 5% extra patients with low concentration monoclonal FLCs will be detected. Some of these patients will have LCMM, NSMM or AL amyloidosis. Many of these patients also have FLCs in the urine but some do not. If monoclonal FLCs are detected in the serum then there is little value in performing additional urine tests. However, there are rare patients that have normal sFLC concentrations but have detectable FLCs in concentrated urine by IFE. These include patients with AL amyloidosis but not patients with LCMM.
MGUS have historically comprised intact immunoglobulins and these are detected using SPE and sIFE. FLC only MGUS have been observed in the urine but only rarely. FLC MGUS have recently been detected in serum using FLC immunoassays at a frequency of approximately 20% of intact immunoglobulin MGUS (3% of samples from individuals over 70 years of age). Since MGUS comprises 60-80% of monoclonal gammopathies, many new FLC MGUS patients will be detected. These may be the precursor of LCMM and AL amyloidosis and are a focus of considerable clinical interest.
Yes, serum is preferable to urine for monitoring patients for several reasons. Serum samples are easier to collect than 24-hour urine samples and serum is a more reliable fluid for assessing changes in production of FLCs. Also, normal serum concentrations of FLCs are less variable than urine concentrations so abnormal results are more easily assessed. Since serum is more sensitive for detection of FLCs than urine, sFLCs are more effective for assessing minimal residual disease. The new international guidelines for MM monitoring include the use of sFLCs for meeting stringent remission criteria.
By definition, these patients have no detectable monoclonal proteins in their serum and urine by conventional electrophoresis tests and have to be monitored by marrow biopsies or bone scans. sFLC tests are clearly useful in these patients and more accurate than bone marrow biopsies that can miss patchy tumour deposits. sFLC concentrations assess FLC production by the whole bone marrow (and extramedullary sites) so are probably a better reflection of overall tumour activity than isolated bone marrow aspirations. sFLC tests have led to a reduced number of bone marrow biopsies in these patients. However, as for all tumours, a tissue diagnosis, obtained by biopsy, is essential to establish the initial diagnosis, even if the FLC test results are grossly abnormal.
AL amyloidosis is a difficult disease to diagnose because existing serum and urine tests are not sufficiently sensitive to identify all patients. sFLC measurements not only identify more patients than electrophoresis tests of serum and urine but also they are important for monitoring disease progress. The short serum half-life of FLCs makes them a mandatory test for evaluating responses to treatment and identifying relapse.
Publications from Birmingham University and The Mayo Clinic have shown similar results. The data from The Mayo Clinic contains results from older people and indicates that sFLC concentrations increase in normal individuals aged >70 years due to deteriorating glomerular filtration. Results from patients should be compared with the age-matched, normal range data. However, all laboratories should make some assessment of normal ranges in their own laboratories since there will be minor variations resulting from differences in race, age, exposure to infections, the use of different instruments, etc.
Normal ranges are usually established by reference to standards (comprising purified proteins) that have been agreed by international committees. Since there is no international agreement, at present, and because of the novelty of the FLC tests, no definitive statement on the accuracy of the normal ranges can be made. In time, agreed reference materials will be manufactured and international agreement established. Existing normal ranges will then be adjusted to take into account any recommendations.
sFLC abnormalities should be assessed from the concentrations of the clonal FLC, the alternate FLC and the κ/λ ratio. This is optimally performed using a κ/λ log plot for each patient's result (see Figure 30.1) which is compared with normal range and disease group data. This elegantly distinguishes monoclonal FLC diseases from polyclonal FLC abnormalities and individuals that are in the normal range. Since renal function is frequently affected in MM patients, the alternate FLC concentrations are often elevated but the κ/λ ratio remains abnormal.
If the patient has bone marrow suppression, the individual FLC concentrations might be low, in which case the κ/λ ratio may be more helpful. This is typical of patients undergoing chemotherapy and in those patients with NSMM and AL amyloidosis with bone marrow suppression. Clinical judgement should be used in these cases. If either the FLC concentrations or the κ/λ ratios are outside the normal range then the cause should be investigated. Consideration should be given to assessment of renal function and causes of increased immunoglobulin production such as autoimmune diseases, chronic infections and some malignant tumours.
FLC results are more useful than existing SPE or IFE because they are more sensitive and, therefore, more likely to detect residual disease. Also, sFLC concentrations are more likely to be elevated than uFLC levels when patients are in remission. Thus, some patients apparently in full remission by existing tests might have abnormal FLC concentrations and their clinical status will need to be revised (Chapter 25.4, Table 25.4). Some patients who relapse between monitoring periods may be more easily identified using sFLC assays rather than SPE or IFE.
As patients develop renal impairment the concentrations of κ and λ polyclonal serum FLCs increase. This is associated with an increase in the monoclonal FLC but the κ/λ ratio will also increase slightly because of the relative reduction in clearance of κ molecules (Chapter 13.2) . However, changes in the κ/λ ratio are a better guide to changes in clinical status when glomerular filtration rates are changing than the concentrations of the monoclonal FLC. Changes in the concentrations of creatinine or cystatin C should be assessed, under these circumstances, in order to provide an independent assessment of renal function and a correction factor for the κ/λ ratios.
Bone marrow suppression, either because of bone marrow replacement by tumour or resulting from chemotherapy, leads to a reduction in the concentrations of polyclonal FLCs. Typically, at the time of diagnosis, the alternate FLC is suppressed and there is a grossly distorted κ/λ ratio. κ/λ ratios play an important role in assessing changing FLC concentrations in these patients and may even be useful when the concentrations of the monoclonal FLCs are below normal levels.
This is an important issue. At present, intact monoclonal immunoglobulins in serum or FLCs in urine are used to monitor the progress of patients. The half-life of IgG is 3-4 weeks, so reductions in tumour mass with chemotherapy may not be reflected in serum monoclonal protein changes for several weeks.
The half-life of FLCs is only a few hours - extending to 2-3 days when renal function is impaired. Thus, reductions in tumour mass with chemotherapy can be identified earlier when the patients are being monitored using sFLC tests. Indeed, changes in tumour mass can be assessed between each cycle of chemotherapy allowing subsequent treatments to be specifically tailored to individual patients. Changes in urine and serum levels of FLCs broadly correspond but renal tubular reabsorption prevents accurate assessment of tumour responses from urine measurements.
Approximately 95% of patients with MM, excreting intact immunoglobulins have abnormal sFLCs. It is likely, because of the short half-life of FLCs, discussed above, that disease monitoring in many of these patients will be better using sFLC levels rather than intact immunoglobulin concentrations.
Yes, sFLC assays are useful for monitoring post transplantation progress in AL amyloidosis and MM and may indicate relapses and responses earlier.
Serum polyclonal FLC concentrations increase if there is increased production or reduced glomerular filtration of κ and λ proteins. Increased production results from any disease that stimulates B-cell proliferation such as infections, autoimmune diseases, various tumours etc. For example, in active SLE, total immunoglobulin production increases 3-4 fold with a corresponding increase in sFLC concentrations.
Reduced glomerular filtration of FLCs occurs in renal damage because almost all FLC removal is via the glomerular pores and the proximal renal tubules. In renal failure, sFLC concentrations may rise 10-20 fold but in all cases both free κ and λ are elevated, so the free κ/λ ratio remains normal. In many diseases, particularly SLE, increases in polyclonal FLC concentrations are due to a combination of increased FLC production and reduced renal filtration.
In situations of increased polyclonal FLC production or reduced filtration, there may be moderate distortions of κ/λ ratios (e.g., 3-4 standard deviations from the normal range). These patients are borderline abnormal and should be investigated appropriately.
30.3 Laboratory questions about serum testing for free light chains
No. SPE detects intact immunoglobulin monoclonal proteins and some FLC monoclonal proteins. In contrast, sFLC assays detect FLC monoclonal gammopathies, either alone, or in association with intact monoclonal immunoglobulins. Approximately 95% of patients with intact monoclonal immunoglobulins have abnormal sFLCs, but not all, particularly those patients with low concentration MGUS.
Yes. Monoclonal proteins can be intact immunoglobulins or FLCs. Since sFLC assays are >100 times more sensitive than electrophoretic tests it is most unlikely that FLCs will be detected in serum by SPE or IFE yet be normal by FLC assays. However, intact monoclonal immunoglobulins by tradition (especially MGUS) are detected by IFE and SPE, but may have normal FLC concentrations. Studies indicate that all LCMM and ~95% of AL amyloidosis patients are correctly identified by sFLC assays.
The sensitivity of SPE for monoclonal bands depends upon the width of the band and its position in the gel in relation to other plasma proteins. Narrow monoclonal bands in the gamma region bands in association with hypogammaglobulinaemia will be visible at 200-400mg/L. The same band in a beta position, perhaps superimposed on transferrin, will be invisible. The monoclonal protein may need to be over 2,000mg/L to be visible in this area of the gel. In addition, monoclonal FLCs may be polymerised to different extents and then they migrate on electrophoresis gels as diffuse bands. This is frequently found in NSMM and is well-documented in LCMM. In these patients, even 5,000mg/L of monoclonal FLC may be difficult to detect above the background of the other plasma proteins.
Yes, the sFLC assays will detect all patients with LCMM and most patients with NSMM and AL amyloidosis. IFE will not detect many of these patients. Since sFLC immunoassays are 100-fold more sensitive than serum IFE, sFLC abnormalities, visible by IFE, will be exceptionally rare if the sFLC tests are normal. Urine IFE or uFLC measurements may be helpful in these rare cases.
No. Renal damage never increases the glomerular filtration rate of small molecules such as FLCs or creatinine since they normally pass relatively unhindered through the glomerular pores. Molecules as large as albumin are not normally filtered by the kidney but they are cleared in nephrotic syndrome as the glomerular pores become damaged. The extra protein leakage overwhelms the proximal tubular reabsorption mechanisms allowing many different proteins to appear in the urine. The protein leakage damages the tubules in the process which become non-functional with glomerular death. Renal clearance of all small proteins is then reduced. This leads to an increase in sFLC levels (and creatinine). In the early stages of the process, FLCs are increased in the urine because of increased competition with albumin for reabsorption by the proximal tubules.
Renal impairment leads to increases in both κ and λ FLCs in the serum. Therefore, when both are elevated the likely cause is a reduction in glomerular filtration. There is a correlation between changes in the concentrations of serum creatinine, cystatin C and FLCs during changes in renal function.
Using electrophoresis methods, clonality is judged by the appearance of a narrow protein band. Using FLC assays, clonality is judged by the numerical ratio of free κ to free λ concentrations. In a similar manner, B-cell clonality in leukaemia is assessed by cellular κ/λ ratios using flow cytometry. Arguably, numerical FLC ratios are more accurate than visual assessments of stained bands on electrophoresis gels. Furthermore, in NSMM, clonality may not be apparent by any electrophoretic procedure but is usually identified by serum κ/λ ratios. In the situation of biclonal gammopathies, with increased synthesis of both free κ and λ molecules, free κ/λ ratios may be normal but the concentrations of both FLCs will be raised.
All tests have borderline results. For FLCs, results should be judged against normal and disease state sera from the laboratory, and from national and international reference ranges. The normal range recommended for the free κ/λ ratios is greater than that used for most tests in order to provide a large safety margin for normal individuals.
Since FLC results are quantitative, less experience is required compared with protein electrophoresis. This leads to less subjective interpretation of results.
All tests produce false positive and false negative results and these need to be assessed for clinical significance. For FLCs, reference ranges have been developed in collaboration with The Mayo Clinic and include individuals up to 90 years of age. Some of these individuals have minor degrees of renal impairment. This increases the concentrations of the FLCs, and the κ/λ ratios, and is apparent on a κ/λ log plot. The difference between the normal and abnormal samples is then selected using standard deviations from the mean. If all of the 282 normal samples in the Mayo Clinic study are used, this represents four standard deviations from the mean and is greater than normally chosen cut off levels. Therefore, test samples outside this range will most likely indicate patients with monoclonal gammopathies.
Negative sFLC results occur in a few patients with NSMM, AL amyloidosis and LCDD. Also, rare patients have monoclonal proteins in the urine detected only by IFE. There is no evidence that they are abnormal in shape or size because they react with FLC antibodies. Their origin is unclear but it could be from minor tubular reabsorption failure or from breakdown of monoclonal immunoglobulins in the tubules. They appear to be of little clinical consequence.
Approximately twice as many k molecules are produced as λ. Since free λ is mostly in dimeric form it has a half-life (determined by glomerular filtration) that is approximately three times that of monomeric κ FLCs. This causes free λ molecules to accumulate in the serum more than free κ molecules and alters the free κ/λ ratio from 1.8 to 0.6. When the light chains are bound to immunoglobulins they are metabolised as the whole immunoglobulin, which is independent of light chain type, so total κ/λ ratio is 1.8:1.
Normal sFLC concentrations are <30mg/L, which is much lower than total light chain concentrations of 1,000-3,000 mg/L. Patients who have levels of monoclonal FLCs between these two ranges cannot be assessed using total light chain assays. Since this applies to most patients with ALamyloidosis, NSMM and many LCMMs, sFLC assays have considerable benefit in these diseases.
CZE of serum is more sensitive than SPE but less sensitive than IFE for detecting monoclonal proteins. In a recent study, it was shown that sFLC assays detected all monoclonal FLCs from patients with LCMM that were missed by CZE but detected by IFE. If CZE is used for initial detection of monoclonal proteins, sFLC assays will detect additional patients.
The FLC antisera are produced by immunisation with many different monoclonal FLC proteins. These are not representative of all monoclonal FLCs but the antibody target is the constant region of the molecule that has little structural variation. However, tumour produced monoclonal FLCs may be truncated, have amino acid substitutions or additions and may be abnormally polymerised. Therefore, occasional patients’ monoclonal FLCs may not be detected reliably by the immunoassays or may be detected differently with different antiserum batches. It is, therefore, ideal laboratory practice to assay current and previous samples alongside each other. This is no different from the situation when quantifying IgG with different antiserum batches.
To minimise batch-to-batch variation, antisera pools are large, are prepared from multiple immunisations and are carefully controlled to maintain consistency. Many monoclonal proteins are tested when new batches are prepared but there is a limit to the number of different monoclonal proteins that can be used.
Monoclonal antibodies have been assessed in some studies to measure FLCs but they have proved to be unreliable. Polyclonal antisera are superior since they detect more monoclonal FLC molecules and they detect them more reliably.
The preferred option is to measure FLCs alongside SPE/IFE at the time of the presentation blood sample. SPE will identify all IIMM patients while FLC assays will identify LCMMs, most NSMMs and other FLC diseases such as AL amyloidosis. Low concentration, intact immunoglobulin MGUS sera (less than 2-5g/L) will not be detected using these two procedures. A strategy of performing SPE/IFE as a screen for FLC monoclonal proteins and not FLC immunoassays will result in some patients with LCMM, NSMM and AL amyloidosis being missed. FLC assays, performed on patients' presentation samples, are also important for providing a baseline for subsequent disease monitoring. For easy interpretation, results should be reported using a logarithmic κ/λ plot, alongside existing clinical data.
When monitoring patients with FLC diseases, results should be reported alongside other analyses. IFE may add little to the combined use of SPE/FLC tests apart from identifying some low level intact immunoglobulin MGUS samples and rare patients with AL amyloidosis.
Yes, for two reasons. The range of monoclonal sFLCs is huge, from a few mg/L to many g/L. Hence, assay conditions causing antigen excess occur on a regular basis. In addition, the small size of FLC molecules and the variety of different shapes and sizes may produce antigen excess conditions at relatively low concentrations for some samples. For accurate results, care must be taken to dilute high concentration samples into the appropriate assay range. If in doubt, samples should be re-analysed at higher dilutions.
Nephelometers and turbidimeters have a similar level of sensitivity and precision for measuring sFLCs. Instruments vary in their ability for sample handling, in providing clean cuvettes for each test, for evaluating antigen excess, etc. Generally, the large clinical chemistry analysers are the best platforms for measuring sFLCs. The assays are available for most analytical platforms and others will be available in due course. Since the FLC kits are specifically prepared for each instrument, cross-usage will produce unreliable results.
Quantification of monoclonal FLCs by immunoassay is less accurate than scanning bands on SPE gels. This is the same situation as using nephelometry for measuring intact immunoglobulin monoclonal proteins. Studies have shown that purified monoclonal FLCs assessed by accurate quantitative protein tests may give quite different results compared with immunoassays for FLCs. The explanation is that the antibody assays cannot be expected to produce consistent results for all molecular shapes and polymeric forms of FLCs.
Since the exact amount of serum or uFLCs, at the time of diagnosis, bears little relationship to disease outcome, accurate quantification is relatively unimportant. Of greater concern is the reproducibility of the assay results in individual patients during treatment. It is apparent that FLC measurements produce consistent results during chemotherapy and this provides the basis for their value in managing patients with the various monoclonal diseases. Indeed, sFLC tests are much more reproducible than electrophoresis tests.
High concentrations of both FLCs in the urine and serum indicate a degree of renal impairment. If the kidney becomes further damaged then both serum and urine concentrations may rise further. This is because with increasing renal damage, glomerular filtration falls and sFLC concentrations increase. This leads to the circulating FLCs being filtered by the remaining nephrons. As their proximal tubules become overwhelmed by the increase in filtered FLCs there is more leakage into the urine. Improving renal function is characterised by reductions in both serum and uFLC concentrations.
The mechanism of abnormal uFLC κ/λ ratios seen in some of these patients can be explained by the renal handling of the molecules. In patients with renal damage, the glomerular pores become altered in size so that monomeric and dimeric FLCs may be filtered differentially. In addition, the proximal tubular reabsorption mechanism is partly dependent upon molecular charge, which is different for each FLC type. Thus, there may be differential clearance and adsorption of κ or λ molecules. This may lead to small distortions of serum and urine κ/λ ratios in patients with renal impairment.
There are a number of reasons why results would differ to a moderate extent. Instruments vary in their optical capabilities, different dilution capabilities, normal range usage, etc.. For example the BNII uses nephelometric, end-point reactions, the Immage assays use a turbidimetric rate reaction and the Hitachi and Olympus analysers both use turbidimetric end-point reactions. Since the normal range for Freelite was established on the Dade Behring BNII, results on other instruments will differ. While every effort is made to ensure cross-comparability of results, exactly matching results are impossible, particularly for samples with low FLC concentrations.
| Chapter 29 | Back to Contents Page | Chapter 31 |
