Screening Panels for Monoclonal Gammopathies: Time to Change

Abstract

The introduction of quantitative assays for serum free light chains (FLC) has changed the approach to screening for monoclonal gammopathies. Recent guidelines from the International Myeloma Working Group have recommended the use of serum protein electrophoresis (SPEP), immunofixation electrophoresis (IFE) and FLC as the screening panel unless primary amyloidosis (AL) is suspected. If screening for AL, then urine IFE should also be performed. We discuss the background for these recommendations as well as data showing that SPEP and FLC alone may provide a simplified screening panel for detecting multiple myeloma (MM) and Waldenström’s macroglobulinaemia (WM) with little loss of diagnostic sensitivity. If AL is suspected, we support the recommendation to include serum and urine IFE. The selective use of serum and urine IFE requires that clinical judgment be integral to ordering screening panels for detection of monoclonal gammopathies.

Introduction

Based on the secreted monoclonal immunoglobulin, plasma cell proliferative disorders (PCPDs) are generally classified as monoclonal gammopathies. The PCPDs include malignant disorders such as MM, plasmacytoma, plasma cell leukaemia, and WM; protein or low tumour burden diseases such as AL, light chain deposition disease (LCDD), and POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes) syndrome; and premalignant diseases such as monoclonal gammopathy of undetermined significance (MGUS) and smouldering multiple myeloma (SMM). Because of the wide range of biology and disease presentations, the identification of the monoclonal immunoglobulin may often be the first clue to the diagnosis. The recognition of the monoclonal protein (M-protein) may be trivial or may require multiple approaches. The malignant disease of MM, for example, usually presents with a significant amount of M-protein, but even within this malignant diagnosis, there is a sub-classification of non-secretory MM (NSMM) which secretes little or no M-protein. The low tumour burden diseases such as AL or LCDD may have small numbers of clonal plasma cells and small amounts of M-protein, but their clinical implications may be severe and they require early diagnosis and intervention.

The Serum Free Light Chain Assay – A New Tool to Screen for Monoclonal Gammopathies

Established diagnostic screening panels for patients suspected of MM, AL, and related monoclonal gammopathies have included protein electrophoresis and IFE of both serum and urine.¹ Urine studies were required because the unbound immunoglobulin light chains can rapidly transit the blood, and the electrophoretic methods of urine protein electrophoresis (UPEP) and IFE can often detect FLC in urine more easily than in serum. However, the recent introduction of quantitative serum assays for κ and λ FLC has increased the sensitivity of serum testing strategies for identifying monoclonal gammopathies,^2,3 especially the monoclonal light chain diseases.^3–5 Detection of low concentrations of serum FLC enables calculation of a κ/λ ratio to identify clonal expansions of plasma cells that secrete excess FLC. Moreover, in PCPD patients who do not secrete a quantifiable M-protein, the sensitivity of the FLC assays enables disease monitoring.^4–6 Surprisingly, FLC quantification has also proved to be prognostic for progression of the premalignant conditions of MGUS⁷ and SMM⁸ as well as for survival in MM,^9–11 plasmacytoma,¹² and AL.¹³

In spite of some reported limitations to the FLC assays such as dependence on a single manufacturer, non-linearity, potential for antigen excess, and apparent incomplete coverage of lambda light chains,^{14, 15, 16, 17} the increased diagnostic sensitivity for FLC diseases as well as the prognostic and monitoring capabilities has led many clinicians to include serum FLC in screening panels for monoclonal gammopathies. New procedural recommendations from the manufacturer appear to have removed antigen excess as an issue, and because FLC is not recommended for monitoring if there is a quantifiable M-protein, κ non-linearity is less of a disadvantage. Because the quantitative serum FLC assay is relatively new and the PCPDs contain such a broad range of biology, the benefits and drawbacks of various testing strategies is only recently becoming clear. The quantification of FLC in urine, however, is neither sensitive for detection of monoclonal proteins^15,18 nor recommended for disease monitoring:^3,18,19 and urine FLC should not be used as part of a diagnostic screening panel. A number of approaches for diagnostic screening have been proposed.^16,20–25 The most recent diagnostic screening recommendations from the International Myeloma Working Group are that SPEP and IFE plus FLC is a sufficient screening panel for PCPDs other than AL.²⁶ It is recommended that screening for AL should also include urine IFE.

Evaluation of New Screening Panels for Monoclonal Gammopathies

In order to evaluate the use of laboratory tests and test panels, it is necessary to understand their sensitivity and specificity as well as their costs and patient compliance. For many tests and diseases these are dif cult parameters to define precisely. In deciding the “gold standard” to use for determining sensitivity and specificity, by far the best is disease association: if an assay is abnormal and the patient has the “disease”, the test result is a true positive. Approximately two-thirds of the monoclonal gammopathies that we identify in our Mayo Clinic practice, however, are laboratory-defined disorders. The premalignant conditions of MGUS and SMM represent 61% and 3% of the monoclonal gammopathies that we have identified since 1960 (Table 1).²⁷ These “diseases” have no clinical symptoms but are clearly monoclonal gammopathies and PCPDs, and there is considerable increased risk for development of MM. The commonly accepted gold standard for monoclonal gammopathies is the identification of M-proteins by the use of IFE even though IFE has some limitations. There are small numbers of patients among disease categories such as NSMM, AL, LCDD and plasmacytoma that may not be detected as abnormal by urine or serum IFE. Some of these patients, however, may have abnormal serum FLC κ/λ ratios. The gold standard therefore needs to include a combination of disease diagnosis and IFE. In addition, the choice of test populations will affect specificity and patient compliance. Some of the early publications on the use of the serum FLC assay evaluated patient results from populations that originated predominantly from haematologists, and the assay had 100% specificity.²⁰ Subsequent studies have evaluated results from more general hospital and outpatient populations and have documented specificities of 96–98.5%.^22–24 A majority of the false-positive results are slightly elevated FLC κ/λ ratios, and a high percentage of these fall in the newly defined reference interval for those with chronic kidney disease.²⁸ As renal clearance declines significantly, there is no longer a preferential removal of κ FLC from the blood, and the FLC κ/λ ratio rises slightly. These false-positives have led to some wariness about the use of the serum FLC assay in screening panels. The FLC specificity, however, should be compared to the specificity of SPEP which has been the backbone of screening for many decades. Most laboratories define a screening SPEP as abnormal if there is a quantifiable M-protein, a fuzzy restricted-migration band (not quantifiable as a M-protein), hypogammaglobulinaemia, or elevated beta or alpha fractions above defined cutoffs. If the SPEP is abnormal by any of these criteria, an IFE is performed to confirm and characterise the monoclonal protein. In our own practice, the SPEP specificity is 92.5%. Of the SPEP samples identified as having fuzzy gamma bands, hypogammaglobulinaemia, beta ≥16 g/L, or alpha 2 ≥ 15 g/L, 7.5% are normal by IFE. No large studies evaluating various screening panels have documented the specificities of both the SPEP and FLC portions of the testing panel.

Table 1.

Distribution of plasma cell proliferative disorders from the Mayo Clinic dysproteinaemia database, 1960–2002 (n = 29,528).

MGUS	61%
MM	18%
AL	9%
Lymphoproliferative Disease	3%
SMM	3%
Solitary or Extramedullary Plasmacytoma	2%
Macroglobulinaemia	2%
Other	2%

MGUS, monoclonal gammopathy of undetermined significance; MM, multiple myeloma; AL, amyloidosis; SMM, smouldering multiple myeloma.

International Myeloma Working Group Recommendations for Screening Panels

A number of studies in defined disease cohorts have shown that the addition of serum FLC to SPEP and IFE can replace the need for urine screening studies in light chain MM,²⁹ NSMM⁴ and SMM,⁸ and that the FLC added to serum and urine IFE increases diagnostic sensitivity when used in AL.^5,20 The most recent screening recommendations from the International Myeloma Working Group are that SPEP, IFE, and FLC is a sufficient screening panel for PCPDs other than AL, and that screening for AL should also include urine IFE.²⁶ The recommendation to base screening panels on serum tests is based primarily on the disease cohort studies listed above, as well as a screening study from Mayo Clinic.²¹ In this study we specifically addressed the question of whether urine IFE was needed in the initial diagnostic screening panel for monoclonal gammopathies. We identified all newly diagnosed Mayo Clinic patients that had a monoclonal gammopathy, had an M-protein identified in the urine, and who also had concurrent SPEP, IFE and FLC assays. The cohort consisted of 428 patients with MM (n = 148), AL (n = 123), MGUS (n = 69), and SMM (n = 59) as well as small numbers of osteosclerotic myeloma, macroglobulinaemia, lymphoproliferative disease, LCDD, and plasma cell leukaemia. This cohort is of course skewed due to the specific study requirement for abnormal urine studies. Although 61% of monoclonal gammopathies in our practice are MGUS, they were only 16% of this cohort. Only a minority of MGUS patients have urine studies performed and 43% of MGUS patients have excess monoclonal FLC²⁰ compared to patients with MM who almost always have urine studies and in whom >95% have abnormal FLC. If these 428 patients only had SPEP and IFE performed, 28 patients (6.5%) would have been missed. These included 2 MM, 19 AL, 3 plasmacytoma, 1 SMM, and 3 MGUS. These 28 patients reinforce the rationale for having included urinary studies in the screening panel of SPEP and UPEP and IFE. With the inclusion of the serum FLC assay, however, only 2 MGUS patients remained undetected. One of the MGUS patients had no serum abnormalities, but had a monoclonal IgAκ detected in the urine. The urinary IgAκ was no longer apparent at a 3 month follow-up. This finding was most likely a contamination artefact or specimen mix-up. The second MGUS had a small λ M-protein in the urine (83 mg/24 hr.). We concluded from this study that urine studies are not required as part of the screening panel. If an M-protein is diagnosed, however, subsequent urine studies to evaluate renal function or to monitor disease may still be needed. At the time of this 2006 analysis, the US Medicare cost for urine studies was $71 against $38 for the FLC assay (κ and λ). In these 428 patients, the substitution of serum FLC for urine studies would have reduced the cost of the diagnostic screen with no loss of diagnostic sensitivity. The subsequent study by Paladini et al. identified 5 of 115 AL patients (4.3%) in whom serum IFE and FLC were normal and yet the urine IFE contained a monoclonal λ light chain identified by IFE.¹⁶ Because early intervention is critical to prevent organ failure in AL, these findings are important and are the reason for the exception in the International Myeloma Working Group recommendations.

This group of studies was undertaken to determine if a urine specimen was needed for the initial screening panel. The conclusions are that for all suspected diagnoses except AL, a panel of SPEP, IFE, and FLC is recommended, and when AL is suspected, the urine IFE should be added to this panel. Because many patients do not submit urine for assessment, the overall diagnostic success should increase with this new screening strategy.

Time to Change: A New Screening Panel of Serum Protein Electrophoresis and Serum Free Light Chain Assay

These studies have not addressed the question about the need for serum IFE in a screening panel for monoclonal gammopathies. A panel of SPEP and FLC would reduce costs by eliminating the serum IFE, and would potentially eliminate much of the manual testing effort. The use of capillary electrophoresis for SPEP and nephelometry for FLC quantification would essentially automate screening for monoclonal gammopathies. A number of studies have used SPEP and FLC as a screening panel.^22–24 Bakshi et al. performed capillary zone electrophoresis on 1003 consecutive samples, identified 39 patients with monoclonal gammopathies, and found that the FLC assay identified an additional 16 patients for a 41% increase in detection rate.²² In a study of a US Veterans Administration population, 312 patients were tested by SPEP and FLC.²³ The SPEP and FLC assays identified 9 and 17 false-positives respectively. The serum FLC assay increased the detection of monoclonal gammopathies from 68 to 88 patients for a 29% increase. In neither study, however, was IFE performed to define the total number of cases and therefore the sensitivity of the screening panel.

We have recently studied the sensitivity of screening panels in order to define the sensitivity of various testing combinations for the detection of the spectrum of PCPDs. We identified a cohort of 1877 untreated patients with an assortment of plasma cell proliferative diseases.³⁰ These patients all had UPEP and IFE as well as SPEP, IFE, and FLC performed within 30 days of diagnosis. The purpose of the study was to assess the sensitivity of the various tests or combinations of tests for detecting monoclonal gammopathies in patients with PCPDs. In this study an abnormal FLC result was defined as an abnormal FLC κ/λ ratio (normal = 0.25–1.65). An abnormal agarose SPEP was defined by the presence of a quantifiable M-protein, a fuzzy restricted migration band, hypogammaglobulinaemia (<5.5 g/L), an elevated beta fraction (≥16 g/L), or an elevated alpha 2 fraction (≥15 g/L). There were some SPEP abnormalities that were not subsequently abnormal by serum IFE (e.g. hypogammaglobulinaemia); they were still considered abnormal in this study if the urine IFE or serum FLC assay was abnormal and therefore the SPEP had agged the abnormality. The serum IFE assessed migration patterns for γ, α, μ, κ, and λ immunoglobulin chains. Urine samples were concentrated to a maximum of 200-fold to attempt to achieve final concentrations of urine protein between 20 and 80 g/L. The 1877 patients were grouped into 9 disease groups (MM, SMM, MGUS, plasmacytoma, extramedullary plasmacytoma, WM, AL, LCDD, and POEMS syndrome). The MM group of 467 patients included 451 MM, 4 NSMM, 4 plasma cell leukaemia, 1 osteosclerotic (non-POEMS) myeloma, and 7 indolent myelomas. The 26 macroglobulinaemia patients included 18 WM, 5 smouldering WM, and 3 cryoglobulinaemias. The 524 MGUS patients included 41 patients with idiopathic Bence Jones proteinuria. The results from these 1877 patients (Table 2) can be summarised as follows:

Table 2.

Sensitivity of monoclonal gammopathy screening panels.

		Serum	Serum	Serum	Serum	Serum	Serum	Serum
		SPEP	SPEP	SPEP	SPEP		SPEP
		+IFE	+IFE	+IFE		IFE
		+FLC		+FLC	+FLC			FLC
		Urine	Urine
		IFE	IFE
Diagnosis	N	N	N	N	N	N	N	N
All	1877	1851	1821	1828	1770	1632	1482	1395
		%	%	%	%	%	%	%
		98.6	97.0	97.4	94.3	87.0	79.0	74.3
Multiple Myeloma (MM)	467	100.0	98.7	100.0	100.0	94.4	87.6	96.8
Smouldering MM	191	100.0	100.0	100.0	99.5	98.4	94.2	81.2
MGUS	524	100.0	100.0	97.1	88.7	92.8	81.9	42.4
Plasmacytoma	29	89.7	89.7	89.7	86.2	72.4	72.4	55.2
Extramedullary Plasmacytoma	10	20.0	20.0	10.0	10.0	10.0	10.0	10.0
Macroglobulinaemia	26	100.0	100.0	100.0	100.0	100.0	100.0	73.1
Primary Amyloid	581	98.1	94.2	97.1	96.2	73.8	65.9	88.3
Light Chain Deposition	18	83.3	77.8	77.8	77.8	55.6	55.6	77.8
POEMS Syndrome	31	96.8	96.8	96.8	74.2	96.8	74.2	9.7

SPEP, serum protein electrophoresis; IFE, immunofixation electrophoresis; FLC, free light chains.

1. A panel of all urine and serum tests identified 1851 patients as abnormal (98.6%). There were 26 patients who had no abnormality in any of these tests: 11 patients with AL (1.9% of all 581 AL); 8 extramedullary plasmacytoma (80%); 3 plasmacytoma (10.3%); 3 LCDD (16.7%); and 1 POEMS syndrome (3%).

2. The panel of urine IFE plus SPEP and IFE (omitting FLC) missed 30 additional patients (56 in all): 6 MM, 23 AL, and 1 LCDD.

3. A panel of SPEP, IFE and FLC (omitting urine IFE) missed 23 additional patients (49 in all): 15 MGUS, 1 extramedullary myeloma, 1 LCDD, and 6 AL.

4. A panel of SPEP plus FLC missed 58 patients compared to the panel of SPEP plus FLC plus serum IFE: 44 patients with MGUS, 7 POEMS, 5 AL, 1 plasmacytoma and 1 SMM. Omitting serum IFE did not miss any patients with MM, macroglobulinaemia, or LCDD.

SPEP, IFE, and FLC assays did not perform well as single tests. SPEP and IFE missed patients in every disease category except macroglobulinaemia, while FLC did not identify 100% of the patients in any category. The 6 AL patients that were missed by serum FLC but captured by urine IFE all had monoclonal λ light chains, confirming the lower sensitivity of the λ assay¹⁶ and suggesting a gap in the λ FLC antiserum. Among the total 57 AL patients that were missed by the serum FLC assay but identified by urine and/or serum IFE (Table 2), 52 (91%) expressed λ light chains.

These recent findings are not dissimilar to previous publications, but are more definitive due to comprehensive testing of patients with various plasma cell disorders. Previous data prompted the International Myeloma Working Group to recommend that screening panels for monoclonal gammopathies should include SPEP, IFE, and FLC; and that urine studies should be added for diagnosing AL. Our more recent series confirms these recommendations for urine. Of the 581 AL patients, 11 (1.9%) had no detectable abnormality in any of our serum and urine assays. Among the 570 patients with a detectable monoclonal protein, 6 patients (1.0%) would have been missed if urine was omitted from the screening panel. Although this is a small number of patients with a relatively rare disease, the consequence of missing them is severe.

Compared to SPEP, IFE and FLC, omission of IFE missed a large number of patients with MGUS (n = 44), a high percent of POEMS (23%), and 5 AL, 1 plasmacytoma, and 1 SMM. Population-based screening for MGUS (a common premalignant disorder with a prevalence of 3% in the population over 50 years of age³¹) is not recommended; however, because MGUS progresses to MM at a rate of 1% per year,³² patients should be monitored for progression once they have been identified. The 44 MGUS patients that were missed by using a screening panel of SPEP plus FLC are among the lowest risk for progression to MM, having a serum M-protein <15 g/L and a normal serum FLC κ/λ ratio:⁷ such patients have a 2% 20-year risk of developing MM. The consequence of missing these low-risk MGUS patients should not be severe since 98% of these patients will never develop MM. Indeed, the savings in periodic testing and patient anxiety may actually be beneficial.

The insensitivity of all of the serum assays used in isolation is a reminder that SPEP and/or IFE should be used in combination with either urine IFE and/or serum FLC. Urine IFE and quantitative serum FLC assays both detect small numbers of abnormalities that the other may miss.^{5,16,20,33–35} In this most recent study, the urine assay in combination with serum IFE detected 15 MGUS, 6 AL, 1 LCDD, and 1 extramedullary plasmacytoma that were missed by the serum FLC assay in combination with serum IFE. Conversely, the serum FLC assay in combination with serum IFE detected 6 MM, 23 AL, and 1 LCDD that were missed by the urine assay in combination with serum IFE.

Conclusions

Serum IFE, SPEP, and FLC combined with urine IFE and UPEP is the most comprehensive and inclusive panel to screen for monoclonal gammopathies. However, because of the small incremental sensitivity provided by urine studies and serum IFE, the use of SPEP plus FLC provides a simple and efficient initial diagnostic screen for the high tumour burden monoclonal gammopathies like MM and WM. Urine studies and serum IFE can be ordered more selectively (Table 3).

Table 3.

Screening panels for different plasma cell disorders.

	SPEP	Serum FLC	Serum IFE	UPEP/Urine IFE
MM	✓	✓
WM	✓	✓
MGUS	✓	✓
SMM	✓	✓
Plasmacytoma	✓	✓	✓
Extramedullary Plasmacytoma	✓	✓	✓
POEMS	✓	✓	✓
AL	✓	✓	✓	✓
LCDD	✓	✓	✓	✓

SPEP, serum protein electrophoresis; FLC, free light chains; IFE, immunofixation electrophoresis; UPEP, urine protein electrophoresis; MM, multiple myeloma; WM, Waldenström’s macroglobulinaemia; MGUS, monoclonal gammopathy of undetermined significance; SMM, smouldering multiple myeloma; POEMS, polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy, and skin changes; AL, amyloidosis; LCDD, light chain deposition disease.