Context.—Nonsecretory plasma cell myeloma is characterized by an absence of detectable monoclonal protein in both the serum and urine. It is generally reported to comprise approximately 1% to 5% of all cases of plasma cell myeloma and, because of its rarity, requires a high index of suspicion and bone marrow biopsy to establish the diagnosis.
Objective.—Review the diagnostic strategy when nonsecretory plasma cell myeloma is a clinical consideration in light of a relatively new serum free light chain assay.
Data Sources.—Case study and review of the literature.
Conclusions.—Initial data using a recently developed nephelometric serum free light chain assay suggests that only about one fourth of nonsecretory plasma cell myeloma cases may be truly nonsecretory. The definition of nonsecretory plasma cell myeloma should be modified to exclude cases with evidence of clonality using the serum free light chain assay.
In November 2002 our laboratory encountered a case of nonsecretory plasma cell myeloma (PCM) in a 64-year-old woman presenting as a pituitary plasmacytoma that was nearly misdiagnosed as an adenoma. There were lytic calvarial lesions and hypogammaglobulinemia. Conventional serum and urine electrophoresis studies, including immunofixation, were negative, but the serum free light chain (FLC) assay (FREELITE, The Binding Site Ltd, Birmingham, UK) showed 884 mg/L κ with a κ-λ ratio of 329:1 (reference range, 0.26–1.65). The histologic findings were then referred for a second opinion and the correct diagnosis was made. Notably, κ and λ immunohistochemical staining of the pituitary mass was negative, as was CD79a. Fortunately, CD138 was positive, supporting plasma cell differentiation. The patient received radiation therapy, but 1 year later developed a low-level immunoglobulin GK monoclonal protein with the bone marrow showing 27% intermediate grade plasma cells. Initially there were only 1.4% plasma cells.
As illustrated by this case, nonsecretory PCM enters into the differential diagnosis when serum and urine protein electrophoresis studies do not detect a monoclonal protein, but the clinical picture continues to have features of a plasma cell dyscrasia. The most important findings are lytic bone lesions and/or hypogammaglobulinemia, which are seen in nearly all cases.1 In a series of 29 patients with conventional nonsecretory PCM, investigators at Mayo Clinic reported that 92% had suppression of at least 1 uninvolved immunoglobulin.2 Notably, none of the patients in that series had a serum creatinine value greater than 2.002 mg/dL, compared with approximately 20% of PCM patients overall.
Results using the serum FLC assay suggest the majority of patients with conventional nonsecretory PCM do have evidence of clonal immunoglobulin production. In the largest series to date, which is from the United Kingdom, 19 of 28 patients presenting with nonsecretory PCM as determined by conventional methods had abnormal κ-λ ratios.3 Mayo Clinic authors reported a small cohort of 5 patients with untreated nonsecretory PCM, all of whom had an abnormal κ-λ ratio, and 9 of 15 treated patients who also had abnormal ratios.4 All 6 of the patients with normal κ-λ ratios had received a stem cell transplant, and 5 of these patients were in morphologic remission based on bone marrow evaluation. Besides helping establish the diagnosis of PCM in these cases, the serum FLC assay also can be a useful adjunct in monitoring therapy.3
The nonsecretory variant is widely quoted as comprising approximately 1% to 5% of all PCM cases. However, even before the advent of the serum FLC assay, the proportion of PCM cases that were considered nonsecretory was likely diminishing as laboratories began using more sensitive detection techniques. Most of the laboratories in North America have converted from immunoelectrophoresis to immunofixation electrophoresis for the identification of monoclonal proteins. In the February 2005 College of American Pathology Survey, 601 laboratories used immunofixation electrophoresis compared with only 21 that used immunoelectrophoresis.
The FLC assay is a latex-enhanced immunoassay that can be performed on a number of commonly used chemistry instruments having nephelometry capability. There are two separate measurements, one to quantify κ and the other to quantify λ. The κ-λ ratio is used to determine clonality. Both FLCs can be elevated in patients with polyclonal hypergammaglobulinemia or renal insufficiency. The FLC concentrations of 2 to 3 mg/L can be detected, compared with approximately 150 mg/L for immunofixation electrophoresis or 500 mg/L for standard serum protein electrophoresis. The typical analytical range goes up 150 to 200 mg/L. Samples containing higher concentrations are rarely misclassified but do require further dilution for accurate quantitation. Coefficients of variation are generally less than 10%. As with any immunologic assay, variable polymerization or abnormal protein structure can affect antibody-to-antigen binding; these rare cases are not well understood. Samples are stable for at least 6 weeks when refrigerated at 4°C and for more than 2 years at −20°C.5
As in our case, surprisingly high levels of serum FLC, well above the 150 mg/L detection limit of immunofixation, may rarely go undetected by that method. This is likely the result of variable polymerization of light chains causing smearing of the monoclonal bands on the electrophoresis gel; this may be more common with κ light chains, which account for the 4:1 ratio of cases with κ versus λ for nonsecretory PCM reported in the literature.3
The serum FLC assay may be a useful adjunct to other methods in screening for monoclonal gammopathies and providing a marker to follow.6,7 A recent report from the University of Michigan found 9 cases (of 1003 consecutive serum samples) of a B-lymphocyte or plasma cell proliferative process that were detected by an abnormal FLC κ-λ ratio in which capillary zone electrophoresis was negative. This included one case of nonsecretory PCM.8
In 2 large series of patients with amyloid light chain amyloidosis, the sensitivity of the serum FLC assay for detecting monoclonal protein was 91%4 and 98%9, respectively. The combination of serum FLC and serum immunofixation detected a monoclonal protein in 99% of 110 untreated patients with amyloid light chain amyloidosis, compared with 95% sensitivity for the more conventional combination of serum and urine immunofixation.4 Furthermore, in many patients with amyloid light chain amyloidosis, the serum or urine monoclonal protein levels are too low to be quantified and the serum FLC provides a better parameter for monitoring the level.10
In a conversation with R. A. Kyle, MD (September 2005), the presence of an abnormal κ-λ ratio at the time of diagnosis of solitary plasmacytoma of bone increases the risk of progression to multiple myeloma. In one small series of 13 patients with solitary plasmacytoma assessed at diagnosis, there were 2 cases of the disease that were nonsecretory as determined by conventional electrophoresis. One of these 2 patients had a monoclonal serum κ light chain by the FLC assay.11
When a plasma cell infiltrate is encountered on biopsy or aspirate material, cytologic evaluation is the important first step. A relatively simple cytologic grading system for plasma cells has the following 4 categories: mature, intermediate, immature, and plasmablast.12 Less-differentiated plasma cell infiltrates generally have larger nuclei with nucleoli and more dispersed chromatin. The nuclear-cytoplasmic ratio tends to be higher and the cytoplasm may not have a Golgi area.
Kappa and λ immunohistochemical stains may confirm clonality but, unfortunately, background staining often makes this interpretation problematic. Different pretreatment conditions (eg, EDTA) can be utilized to try to optimize light chain staining relative to the background. Although not widely available, mRNA in situ hybridization is an alternative technique for demonstrating light chain restriction with a generally cleaner background.13 Anecdotally, we have encountered occasional plasma cell infiltrates similar to those found in the case reported here in which κ, λ, and CD79a stains were all negative, but CD138 (syndecan) showed strong membrane reactivity.
In less-differentiated cases, a neoplastic plasma cell process can be confidently diagnosed, even without proving clonality. However, for more mature plasma cell infiltrates, kappa versus lambda staining is pivotal in distinguishing a polyclonal infiltrate from a monoclonal population. Immunoglobulin gene rearrangement studies have a relatively low yield for confirming clonality in plasma cell neoplasms because of postgerminal center somatic mutations. Furthermore, these studies generally take several days to complete. One needs to look carefully for amyloid and have a low threshold for performing a Congo red stain because nonsecretory PCM can be associated with amyloid light chain amyloidosis.14
CD56 expression, by flow cytometry in most studies, is seen in the majority of cases of PCM. However, it can be found in monoclonal gammopathy of undetermined significance and even in rare cases of polyclonal plasmacytosis.15,16 Further studies need to be performed to evaluate the performance of CD56 immunohistochemical staining of plasma cell infiltrates.
Immunohistochemistry studies should be performed on the plasma cells of all patients in whom nonsecretory myeloma is suspected. A cytoplasmic monoclonal protein (either κ or λ) is identified in approximately 85% of such cases.1 The remaining 15% are sometimes referred to as the nonproducer subtype. However, technical or immunologic limitations of light chain antiserum may account for many of these nonproducer cases.
The prognosis and treatment for conventionally defined nonsecretory PCM is believed to be no different than that of PCM in general.1 Patients with nonsecretory PCM tend to be diagnosed later in their disease course, but this may be offset by their comparative lack of renal insufficiency. One small series from the United Kingdom, comprising 6 patients with nonsecretory PCM, suggested that nonsecretory PCM may have a somewhat better prognosis than most other cases when treated with autologous stem cell transplantation.17
Cytogenetic analysis (both conventional and fluorescence in situ hybridization) is assuming a larger role in prognostication for PCM. There are 2 case reports of cytogenetics in patients with nonsecretory myeloma, but these need to be considered in light of more sensitive current diagnostic methods for detecting monoclonal proteins.18,19 At this time there is no reason to suspect that the chromosomal abnormalities would be any different than those of other PCM cases.
In most cases of nonsecretory PCM, electron microscopy shows the typical ultrastructural features of plasma cells.1 However, there are rare cases of nonsecretory PCM in which the plasma cells were highly undifferentiated or had granular cytoplasm packed with mitochondria.20
The absence of a monoclonal protein in the serum and urine of patients with nonsecretory PCM may be a result of (1) an inability to excrete immunoglobulin, (2) low synthetic capacity for immunoglobulin, (3) increased intracellular degradation, or (4) rapid extracellular degradation of abnormal immunoglobulins. The molecular basis for a case of nonsecretory myeloma (with a normal serum FLC assay) was recently reported.21 This individual had abnormal κ light chains that resulted from a frameshift mutation that led to the absence of cysteines required for disulfide bonds. The misfolded κ light chains were presumably retained within the plasma cell cytosol and then underwent proteolysis.
The Figure shows an approach to the diagnosis of nonsecretory PCM. If a patient has lytic bone lesions and conventional serum and urine studies show no evidence of a monoclonal protein, a directed biopsy may aid diagnosis more so than random posterior iliac crest bone marrow biopsies. If hypogammaglobulinemia is present, the serum FLC assay should be performed expeditiously.
In summary, patients with nonsecretory PCM almost invariably have lytic bone lesions and/or hypogammaglobulinemia. Recent data suggest that approximately three fourths of these cases do have a serum monoclonal FLC. The definition of nonsecretory PCM should be modified to exclude those cases with a clonal serum FLC.
The author has no relevant financial interest in the products or companies described in this article.
Reprints: Gene R. Shaw, MD, Marshfield Laboratories, 1000 North Oak Ave, Marshfield, WI 54449 (email@example.com)