Robert A. Kyle, Mayo Clinic, 200 First Street SW, Rochester, MN 55905 USA. E-mail: firstname.lastname@example.org
Monoclonal gammopathy of undetermined significance (MGUS) is characterized by a serum monoclonal protein <30 g/l, <10% plasma cells in the bone marrow, and absence of end-organ damage (CRAB–hypercalcaemia, renal insufficiency, anaemia, or bone lesions). MGUS is present in 3% of persons >50 years and in 5% >70 years of age. The risk of progression to multiple myeloma (MM) or a related disorder is 1% per year. Patients with risk factors consisting of an abnormal serum free light chain ratio, non-immunoglobulin G (IgG) MGUS, and an elevated serum M protein ≥15 g/l had a risk of progression at 20 years of 58%, compared with 37% with two risk factors present, 21% with one risk factor present, and 5% when none of the risk factors were present. Smouldering (asymptomatic) multiple myeloma is characterized by having a serum IgG or IgA monoclonal protein of 30 g/l or higher and/or 10% or more plasma cells in the bone marrow but no evidence of end-organ damage. The cumulative probability of progression to active MM or amyloidosis was 51% at 5 years, 66% at 10 years and 73% at 15 years; the median time to progression was 4·8 years.
The monoclonal gammopathies are a group of disorders characterised by the proliferation of a single clone of plasma cells that produces a homogeneous monoclonal (M) protein. Each M protein consists of two heavy polypeptide chains of the same class: gamma (γ) constitutes immunoglobulin G (IgG), alpha (α) is found in IgA, mu (μ) is present in IgM, delta (δ) occurs in IgD, and IgE is characterized by epsilon (ε). Two light polypeptide chains [kappa (κ) and lambda (λ)] of the same type are found in each M-protein. Polyclonal immunoglobulins are produced by many clones of plasma cells and are heterogeneous with respect to heavy chain classes and include both light-chain types. A monoclonal increase in immunoglobulins results from a clonal process that is malignant or potentially malignant while a polyclonal increase in immunoglobulins is caused by a reactive or inflammatory process.
The term ‘essential hyperglobulinaemia’ was introduced by Waldenström in 1952 to describe patients who had a small serum protein electrophoretic spike but no evidence of multiple myeloma (MM), Waldenström macroglobulinaemia (WM), primary amyloidosis (AL), or related disorders (Waldenström, 1952). He stressed the constancy of the size of the protein spike and contrasted it with the increasing quantity of the protein spike in MM. Subsequently, many used the term ‘benign monoclonal gammopathy’ but this was misleading because some patients developed symptomatic MM, WM, AL, or a related monoclonal plasma cell proliferative disorder during follow-up. In 1978, the term ‘monoclonal gammopathy of undetermined significance’ (MGUS) was introduced (Kyle, 1978). MGUS is defined as a serum M protein <30 g/l; <10% plasma cells in the bone marrow, if done; little or no M protein in the urine; and absence of lytic bone lesions, anaemia, hypercalcaemia, or renal insufficiency related to the plasma cell proliferative disorder.
Recognition of a monoclonal protein
Agarose gel electrophoresis is the preferred method for detection. After recognition of a localised band or spike on electrophoresis, immunofixation must be performed to confirm the presence of an M protein and to determine its immunoglobulin heavy chain class and its light-chain type. Electrophoresis and immunofixation of serum should be performed when MM, WM, AL, or a related plasma cell disorder is suspected. Quantitation of immunoglobulins may be performed with a rate nephelometer, but measurement of the M spike is the preferred method of quantitation of the M protein. Electrophoresis and immunofixation of a 24-h urine specimen should also be carried out for all patients with MM, WM, AL, and heavy chain diseases or when one of these entities is suspected. Collection of a 24-h urine specimen is necessary because the size of the M-protein spike provides an indirect measurement of the patient’s tumour mass.
The serum free light-chain (FLC) assay is an automated nephelometric test which measures the level of free κ and free λ light chains in the serum. The normal ratio for FLC κ/λ is 0·26 to 1·65. The FLC assay is useful for following patients with plasma cell disorders who do not have a measurable M spike in the serum or urine. It is also useful for monitoring patients with non-secretory MM and is of prognostic value in MGUS and solitary plasmacytoma of bone.
In 2006, 1684 persons with a M protein in the serum or urine were identified at Mayo Clinic: MGUS, 921 (55%); MM, 276 (16·5%); AL, 194 (11·5%); lymphoproliferative disorder, 62 (4%); smouldering multiple myeloma (SMM), 57 (3%); solitary or extramedullary plasmacytoma, 35 (2%); WM, 36 (2%) and other, 103 (6%) (Fig 1).
Prevalence of MGUS
Monoclonal proteins without MM, WM, AL, or a related plasma cell disorder have been reported in approximately 1% of persons older than 50 years of age and in about 3% of those older than 70 years in Sweden (Axelsson et al, 1966), western France (Saleun et al, 1982), and in northern Minnesota (Kyle et al, 1972). MGUS occurs more often among older patients and the frequency is higher in blacks than whites. The prevalence of MGUS was 0·98% in African-Americans and 0·4% in whites in a study of 4 million males admitted to Veterans Affairs Hospitals in the United States. The age-adjusted prevalence of MGUS in the African-Americans was threefold greater than in whites (Landgren et al, 2006).
The first population-based study using sensitive laboratory techniques (agarose gel electrophoresis and immunofixation) to detect M proteins has been reported (Kyle et al, 2006). All living residents of Olmsted County, MN, as of January 1, 1995, were identified. We obtained serum that remained after the performance of routine clinical tests at Mayo Clinic or asked subjects for whom such serum was unavailable to provide a sample. Agarose gel electrophoresis was performed on all serum samples, and any serum with a discrete band of M protein or thought to have a localised band was subjected to immunofixation. Serum samples were obtained from 21 463 (77%) of the 28 038 enumerated residents aged 50 years or older. MGUS was identified in 694 (3·2%) of these persons (3·7% of men and 2·9% of women, P < 0·001) (Table I). The prevalence increased with advancing age and was almost four times as high among persons aged 80 years or older than among those 50–59 years of age. The prevalence of MGUS was 8·9% in men older than 85 years, while in women it was 7·0% (total 7·5%). There was no significant difference in the size of the M protein among the age groups.
Table I. Prevalence of MGUS according to age group and sex among residents of Olmsted County, Minnesota.
*The percentage was calculated as the number of patients with MGUS divided by the number who were tested.
†Prevalence was age-adjusted to the 2000 US total population as follows: men, 4·0% [95% confidence interval (CI), 3·5–4·4]; women, 2·7% (95% CI, 2·4–3·0); and total, 3·2% (95% CI, 3·0–3·5).
‡Prevalence was age- and sex-adjusted to the 2000 US total population.
343/11 994 (2·9)†
694/21 463 (3·2)†,‡
The M protein concentration was modest with 63% having a monoclonal spike <10 g/l. The monoclonal spike was 10–14·9 g/l in 16·6%, 15–19·9 g/l in 15·4% and 20 g/l or greater in only 4·5%. The M protein was too small to measure in 13·1%. The median size was 5 g/l and 7 g/l if the unmeasurable M proteins were excluded. The isotype of the M protein was IgG in 69% of the 694 patients with MGUS, IgM in 17%, IgA in 11% and biclonal in 3%. The light-chain type was κ in 62% and λ in 38%. The concentration of uninvolved immunoglobulins was reduced in 28%. One of the two immunoglobulins was reduced in 22%, while both uninvolved immunoglobulins were decreased in 6%. Urine from 79 patients with MGUS was tested. Immunofixation revealed a monoclonal κ light chain in 16·5% and λ in 5·0%.
Long-term natural history of MGUS
The MGUS produces no symptoms and is found during laboratory testing of an apparently normal patient or during evaluation of an unrelated disorder. MGUS is a common finding in the medical practice of all physicians. It is important for both the patient and the physician to know whether the M protein will remain stable and benign or progress to MM or a related disorder.
Mayo Clinic referral patients
We reviewed the medical records of 241 patients with an M protein but without evidence of MM, WM, or AL who were examined at Mayo Clinic in Rochester, Minnesota, between January 1, 1956 and December 31, 1970 to determine the long-term outcome of patients with MGUS (Kyle et al, 2004). After 3579 person-years (median 13·7 years, range 0–39 years), patients were classified into one of four groups: group 1: patients still living without an increase in serum M protein; group 2: patients in whom the M protein had increased to 30 g/l or higher or in whom bone marrow plasma cells increased to ≥10% but who had not required therapy for their plasma cell disorder; group 3: patients who died of unrelated causes and group 4: patients in whom MM, AL, WM, or a related lymphoplasma cell proliferative disorder had developed (Table II).
Table II. Course of 241 patients with MGUS.*
No. patients at follow-up (%)†
*The patient groups are described in the ‘Patients and Methods’ section in Kyle et al, 2004. MGUS, monoclonal gammopathy of undetermined significance.
†Person-years follow-up = 3579 (median 13·7 year per patient; range, 0–39 years). Permission requested from Kyle et al (2004).
Living patients with no substantial increase of monoclonal protein
Monoclonal protein value ≥30 g/l but no myeloma or related disorder
Died of unrelated causes
Developed multiple myeloma, macroglobulinaemia, amyloidosis, or related disorder
The median age was 64 years when MGUS was recognised. Only 4% were younger than 40 years of age, while one-third were 70 years or older. There were 140 men (58%) and 101 women (42%). IgG accounted for 73·5%, IgA 10·5%, IgM 14% and biclonal 2%. Reduction of uninvolved immunoglobulins was noted in 38%. The number of living patients with a stable M-protein value (group 1) consisted of 14 (6%) patients. The median duration of follow-up in this subgroup was 33 years. An M-protein value of 30 g/l or higher occurred in 25 patients (10%) (group 2), but they did not require chemotherapy for MM or WM. Fifty-seven per cent died without evidence of symptomatic MM, WM, AL or other plasma cell proliferative disorders (group 3). Cardiac disease, cerebrovascular disease and non-plasma cell malignancies accounted for the majority of deaths.
A malignant lymphoproliferative disorder developed in 64 patients (27%) (group 4). The actuarial rate of progression was 17% at 10 years, 34% at 20 years and 39% at 25 years, approximately 1·5% per year (Fig 2). Forty-four (69%) of the 64 patients who progressed developed MM. The interval from recognition of MGUS to progression was 10·4 years (range 1–32 years) (Table III). It should be noted that MM was diagnosed in 10 patients after the M protein had remained stable for 20 years. AL developed in eight patients while WM occurred in seven patients and a lymphoproliferative disorder consisting of malignant lymphoma (three), chronic lymphocytic leukaemia (one) and an atypical malignant lymphoproliferative disorder (one) was observed in five patients.
Table III. Development of multiple myeloma or related disorder in 64 patients with MGUS.
MGUS, monoclonal gammopathy of undetermined significance.
Follow-up of 1384 MGUS patients from southeastern Minnesota
A population-based study was done to confirm the findings of the original Mayo Clinic study, which consisted mainly of patients referred to a tertiary care centre (Kyle et al, 2002). A total of 1384 persons who resided in the 11 counties of southeastern Minnesota were identified as having MGUS, which was defined as a serum M protein of 30 g/l or less, fewer than 10% plasma cells in the bone marrow if the test was done, little or no M protein in the urine, and the absence of lytic bone lesions, anaemia, hypercalcaemia, or renal insufficiency related to the plasma cell proliferative disorder. These patients were evaluated at Mayo Clinic from 1960 to 1994. Seven hundred and fifty-three (54%) were men and 631 (46%) were women. The median age at the recognition of MGUS was 72 years. Two per cent were younger than 40 years of age at diagnosis, while 59% were 70 years of age or older. The size of the serum M protein at diagnosis ranged from unmeasurable (visible as a small band on electrophoresis but not quantifiable by densitometry) to 30·0 g/l. Seventy per cent of the M proteins were IgG, 12% IgA, 15% IgM and 3% were biclonal. The light chain was κ in 61% and λ in 39%. The concentrations of uninvolved immunoglobulins were reduced in 38% of 840 patients whose immunoglobulins were measured. Electrophoresis, immunoelectrophoresis and immunofixation were performed on urine from 418 of the patients with MGUS. Twenty-one per cent had a monoclonal κ light chain and 10% had a λ light chain, while 69% were negative for a monoclonal light chain. The urinary M protein was small with only 17% having a value >150 mg/24 h. The median percentage of bone marrow plasma cells was 3% (range 0–10%). Anaemia or renal insufficiency, when present, were due to causes other than the plasma cell proliferative process.
The 1384 patients were followed for 11 009 person-years (median 15·4 years; range 0–35 years) during which 963 (70%) died. During follow-up, MM, lymphoma with an IgM monoclonal protein, AL, WM, chronic lymphocytic leukaemia (CLL), or plasmacytoma developed in 115 patients (8%) (Table IV). The cumulative probability of progression to one of those disorders was 10% at 10 years, 21% at 20 years and 26% at 25 years (Fig 3). The overall risk of progression was about 1% per year. It must be emphasized that patients were at risk for progression even after 25 years or more of stable MGUS. The rates of death due to other diseases, such as cardiovascular or cerebrovascular diseases and non-plasma cell malignancies, were 53% at 10 years, 72% at 20 years and 76% at 25 years, when compared with 6% at 10 years, 10% at 20 years and 11% at 25 years for death due to plasma cell malignancies (Fig 4).
Table VI. Risk factors for progression among 276 Mayo clinic patients with smouldering multiple myeloma, 1970–1995.
Median time to progression, months
Rate of progression, %
*Patients were divided into three prognostic groups: group 1 (bone marrow plasma cells, ≥10%; monoclonal protein level, ≥3 g per deciliter), group 2 (plasma cells, ≥10%; monoclonal protein level, <3 g per deciliter), and group 3 (plasma cells, <10%; monoclonal protein level, ≥3 g per deciliter. NA denotes not applicable
†The multivariate P values for the type of heavy chain represent the test for an additional significant contribution to the predictive ability of a model already containing the prognostic group; the multivariate P values for the remaining variables represent the test for an additional significant contribution when added individually to a model already containing prognostic groups and the type of heavy chain.
‡One patient had progression, but the date of progression was not available; therefore, this patient was excluded from the analysis of progression.
Serum heavy chain
Bone marrow plasma cells, %
Serum monoclonal protein, g/l
Urinary light chain
κ or λ
Type of urinary light chain
Reduction of uninvolved immunoglobulins
Pattern of bone marrow plasma cell involvement
Interfat space or sheets
Singly distributed or small clusters
The number of patients with progression to a plasma cell or related disorder (115 patients) was more than seven times that expected on the basis of incidence rates for those conditions in the general population (Table IV). The risk of disease was increased by a factor of 25 for MM, 46 for WM and 8·4 for AL. The risk of lymphoma was only modestly increased with a relative risk (RR) of 2·4, while the risk of CLL was not increased (RR = 0·9).
Table IV. Risk of progression among 1384 residents of southeastern Minnesota in whom monoclonal gammopathy of undetermined significance was diagnosed in 1960 through 1994.
*Expected numbers of cases were derived from the age- and sex-matched white population of the Surveillance, Epidemiology, and End Results program in Iowa, except for primary amyloidosis for which the data are from Kyle et al.
†All 19 patients had serum IgM monoclonal protein. If the 30 patients with IgM, IgA, or IgG monoclonal protein and lymphoma were included, the relative risk would be 3·9 (95% CI, 2·6–5·5).
‡All three patients had serum IgM monoclonal protein. If all six patients with IgM, IgA, or IgG monoclonal protein and chronic lymphocytic leukaemia were included, the relative risk would be 1·7 (95% CI, 0·6–3·7).
Chronic lymphocytic leukaemia
The M protein disappeared without a known cause in 2% of MGUS patients. The characteristics of the 75 patients with MM were comparable to those of the 1027 patients with newly diagnosed MM referred to Mayo Clinic from 1985 to 1988, except that the southeastern Minnesota patients were older (median 72 years vs. 66 years) and less likely to be male (45% vs. 60%) (Kyle et al, 2003a).
Follow-up in other series
The actuarial probability of development of malignant plasma cell disease in 128 patients with MGUS from Spain was 8·5% at 5 years and 19·2% at 10 years (Blade et al, 1992). Malignant transformation was the cause of death in 97 patients in a cohort of 1324 patients with MGUS in North Jutland, Denmark, compared with 4·9 deaths expected (Gregersen et al, 2001). Sixty-four new cases of malignancy (five expected, RR 12·9) were found among 1229 patients with MGUS from the Danish Cancer Center (Gregersen et al, 2000). The risk for development of MM was 34·3-fold, WM 63·8-fold and non-Hodgkin lymphoma (NHL) 5·9-fold. In a cohort of 504 patients with MGUS from Iceland, a plasma cell malignancy developed in 51 (10%) after a median follow-up of 6 years (Ogmundsdottir et al, 2002). In summary, most studies confirm that the risk of progression from MGUS to MM or a related disorder is about 1% per year. These studies also point out that the risk does not disappear, even after long-term follow-up.
Causes of progression of MGUS
Genetic changes, cytokines related to myeloma bone disease, marrow angiogenesis and infectious agents may all play a role in the progression of MGUS. Unfortunately, the specific role of these alterations is not known (Blade, 2006; Kyle & Rajkumar, 2006).
Cytogenetic changes are common in MM and MGUS. Approximately 60% of patients with MM have IgH (14q32) translocations (Avet-Loiseau et al, 1998). These same translocations are also present in MGUS. For example, 27 (46%) of 59 patients with MGUS studied with fluorescence in situ hybridization (FISH) revealed t(11;14)(q13;q32) in 25%, t(4;14)(p16·3;q32) in 9% and t(14;16)(q32;q23) in 5% (Avet-Loiseau et al, 1999a). IgH translocations may lead to the dysregulation of oncogenes such as CCND1 (11q13), MAF (16q23), FGFR3/MMSET (fibroblastic growth factor receptor 3/MM SET domain) (4p16) and CCND3 (6p21). However, these changes may be involved with the initiation of the MGUS clone rather than progression of MGUS to MM. Hyperdiploidy was reported in 40% of 28 patients with SMM or MGUS (Chng et al, 2005). This percentage is similar to that of hyperdiploid MM reported in the literature and suggests that hyperdiploid MM originates early during disease evolution. Apparently, all MGUS is associated with evidence of genomic instability that is manifested as primary IgH translocations in approximately one-half and hyperdiploidy in most of the remaining patients.
Deletions of chromosome 13 are found in similar frequencies in both MM and MGUS (Avet-Loiseau et al, 1999b). Deletion of chromosome 13 detected by conventional cytogenetics is associated with an adverse prognostic effect in MM. No obvious clinical or biological correlations have been associated with chromosomal abnormalities in MGUS. In fact, the findings in MGUS and MM are similar. It is unlikely that chromosome 13 deletion detected by FISH plays a significant role in the evolution of MGUS to MM (Fonseca et al, 2002).
KRAS and NRAS mutations have been noted in 5% of MGUS in contrast to 31% of MM, but this may or may not play a role in the progression of MGUS to MM (Rasmussen et al, 2005). Epigenetic changes, such as aberrant methylation of the 5′ gene promoter regions of tumour suppression genes, have been observed in MGUS, although they are lower in frequency compared with myeloma (Takahashi et al, 2004).
Myeloma bone disease associated cytokines
Clinically, lytic bone lesions, osteoporosis, hypercalcaemia and pathological fractures differentiate MM from MGUS. However, we found a 2·7-fold increase in axial fractures but no increase in limb fractures in 488 Olmsted County residents with MGUS (Melton et al, 2004).
Osteoclastic activation and inhibition of osteoblast differentiation are responsible for bone lesions with progression of MGUS to MM. Osteoclast activation is caused by overexpression of various cytokines, such as receptor activator of nuclear factor κ-β ligand (RANKL) and macrophage inflammatory protein 1-α (MIP-1α) (Roodman, 2002). RANKL is modulated by a decoy receptor, osteoprotegerin. Thus, myeloma bone disease may occur from excess RANKL or reduced levels of osteoprotegerin (Croucher et al, 2001). Interleukin (IL)-1β is produced by plasma cells in almost all cases of MM and has strong osteoclast-activating factor activity (Lust & Donovan, 1998). Tumour necrosis factor-α (TNFα) and IL-6 may also play a role.
Vacca et al (1994) first reported that angiogenesis was increased in MM. The median microvessel density (vessels per high-power field) in the bone marrow of 400 patients with plasma cell disorders was 1·3 in 42 normal controls, 1·7 in AL, three in MGUS, four in SMM, 11 in MM and 20 in relapsed MM (Rajkumar et al, 2002a). There may also be a loss of an endogenous angiogenesis inhibitor that may be involved in the increasing angiogenesis that occurs with disease progression. In one study, 63% of MGUS sera inhibited angiogenesis, whereas 43% of SMM and 4% of MM samples did so (P < 0·0001) (Kumar et al, 2004).
Helicobacter pylori infection
In one report, 68% of patients with MGUS had Helicobacter pylori infection. Eradication of the infection led to resolution of the monoclonal gammopathy in 11 of 39 patients (Malik et al, 2002). In another report, 30% of 93 patients with MGUS who were residents of Olmsted County, Minnesota, had positive serological results for H. pylori, as did 32% of 98 control patients from the same population. In addition, 33% of 154 patients from Mayo Clinic with MGUS were positive for H. pylori as were 33% without MGUS. Thus, the role of H. pylori infection in MGUS is controversial (Rajkumar et al, 2002b).
Predictors of risk of progression in MGUS
Prediction of MGUS patients who remain stable and those in whom progression develops is very difficult at the time of diagnosis (Kyle, 1993). However, the size of the M protein, type of M protein, number of bone marrow plasma cells and the serum FLC ratio are all helpful in identifying those patients who are at higher risk of progression.
Size of monoclonal protein
The size of the M protein at the time of diagnosis of MGUS was the most important predictor of progression in 1384 patients with MGUS (Kyle et al, 2002). The risk of progression to MM or a related disorder 20 years after recognition of MGUS was 14% for patients with an initial M protein value of 5 g/l or less, 25% for 15 g/l, 41% for 20 g/l, 49% for 25 g/l and 64% for 30 g/l. The risk of progression with an M protein value of 15 g/l was almost double that of a patient with an M protein value of 5 g/l, while the risk of progression with an M protein of 25 g/l was 4·6 times that of a patient with a 5 g/l spike Rosinol et al (2007) reported that a progressive increase in size of the M protein during the first year of follow-up was the most important risk factor for progression.
Type of immunoglobulin
Patients with IgM or IgA M protein have an increased risk of progression compared to those with an IgG protein (P = 0·001) (Kyle et al, 2002). Blade et al, (1992) also reported that patients with an IgA MGUS had a greater probability for a progression to MM.
Bone marrow plasma cell involvement
In a series of 1104 patients with MGUS, more than 5% bone marrow plasma cells, presence of Bence Jones proteinuria, polyclonal immunoglobulin reduction, and elevated erythrocyte sedimentation rate were independent factors influencing progression of MGUS (Cesana et al, 2002). In another series, the transformation rate was 6·8% when the bone marrow plasma cells were <10% and 37% in patients in whom the plasma cell level was 10–30% (Baldini et al, 1996).
Serum free light chain ratio
Thirty-three per cent of 1148 patients with MGUS from southeastern Minnesota had an abnormal FLC ratio. The risk of progression was significantly higher than that in patients with a normal FLC ratio (hazard ratio 3·5; P < 0·0001) and was independent of the size and type of serum M protein (Rajkumar et al, 2005).
Age, sex, presence of hepatosplenomegaly, levels of haemoglobin, serum creatinine, and serum albumin; presence, type and amount of urinary light chain; number of bone marrow plasma cells; and reduction of uninvolved immunoglobulins were not predictors for progression (Kyle et al, 2002).
Risk-stratification model for MGUS
A risk-stratification model for progression of MGUS was developed. Patients with risk factors consisting of an abnormal serum FLC, non-IgG MGUS and an elevated serum M protein value ≥15 g/l had a risk of progression at 20 years of 58% compared with 37% with two risk factors present, 21% with one risk factor present, and 5% when none of the risk factors were present.
At the time of presentation, the differentiation of a patient with MGUS from one with MM may be difficult and is based on the clinical and laboratory findings. A bone marrow aspirate and biopsy should be performed if myeloma is suspected and in all patients with an M protein value of 15 g/l or more; patients with an IgA or an IgM M protein and abnormal FLC ratio; the presence of unexplained anaemia or elevation of creatinine or calcium levels; or the presence of bone lesions. The size of the serum M protein is of help because higher levels are associated with a greater likelihood of MM or WM. Reduction of uninvolved immunoglobulins or the presence of an M protein in the urine (Bence Jones proteinuria) may also be present in MGUS and are of little help in differentiation. Patients with more than 10% bone marrow plasma cells or an M spike >30 g/l represent SMM, which will be discussed later. Patients with SMM have a higher risk of progression than patients with MGUS. The presence of osteolytic lesions suggests MM, but metastatic carcinoma must be excluded. An elevated plasma cell labelling index usually indicates symptomatic MM, but one-third of patients with symptomatic MM have a normal labelling index. Furthermore, the test is not widely available (Greipp et al, 1987). The presence of circulating plasma cells in the peripheral blood is suggestive of symptomatic MM (Kumar et al, 2005). Conventional cytogenetic studies are not useful because an abnormal karyotype is rare in MGUS as a result of the low proliferative rate and the small number of plasma cells. FISH is not helpful because abnormalities are found in both MGUS and MM.
In summary, the differentiation of symptomatic MM requiring therapy from MGUS or SMM depends mainly on the presence or absence of end-organ damage (CRAB: hypercalcaemia, renal insufficiency, anaemia, bone lesions) that is due to the underlying plasma cell proliferative disorder. MGUS and SMM are distinguished from each other by the size of the serum M protein and the number of bone marrow plasma cells.
Serum protein electrophoresis should be repeated after 3–6 months to exclude MM; if the results are stable and the patient has no clinical features of MM, WM, or AL and a serum M protein value <15 g/l, IgG type, and normal FLC ratio, serum protein electrophoresis may be repeated every 2–3 years. In this setting, a bone marrow examination or skeletal radiography are rarely necessary. In a MGUS patient with a serum M protein value >15 g/l, IgA or IgM protein type, or an abnormal FLC ratio, a bone marrow aspirate and biopsy should be performed. A bone marrow examination should also be carried out in patients who have unexplained anaemia, renal insufficiency, hypercalcaemia, or bone lesions. If possible, cytogenetic studies (both conventional and FISH), determination of the plasma cell labelling index, and a search for circulating plasma cells in the peripheral blood detected by flow cytometry should be performed (Nowakowski et al, 2005). Patients with an IgM monoclonal protein should have a computed tomographic scan of the abdomen because asymptomatic retroperitoneal lymph nodes may be present. Levels of lactate dehydrogenase, β-2 microglobulin and C-reactive protein should be determined if there is evidence of MM or WM. If the results of these tests are satisfactory, serum protein electrophoresis and a complete blood count should be repeated at 6-month intervals for a year and then annually. If there is any change in their clinical condition, the patient must contact their physician immediately.
Variants of MGUS
The IgM MGUS is defined as a serum IgM monoclonal protein <30 g/l, bone marrow lymphoplasmacytic infiltration <10%, and no evidence of anaemia, constitutional symptoms, hyperviscosity, lymphadenopathy, or hepatosplenomegaly (Rajkumar et al, 2006). In a group of 430 patients with an IgM monoclonal protein seen at Mayo Clinic between 1956 and 1978, MGUS was found in 56% (Kyle & Garton, 1987). In another report, Gobbi et al (2005) found an IgM MGUS in approximately 20% of patients with MGUS and in 30% of patients who had an IgM paraprotein.
The IgM MGUS was diagnosed in 213 Mayo Clinic patients who were residents of the 11 counties of southeastern Minnesota (Kyle et al, 2003b). During long-term follow-up, 29 (14%) of these 213 patients developed NHL (n = 17), WM (n = 6), CLL (n = 3) and AL (n = 3) with relative risks of 15-, 262-, 6- and 16-fold respectively. The overall risk of progression was approximately 1·5% per year. The level of serum M protein and the serum albumin value at diagnosis were independent predictors of progression. In another report, 14 (10%) of 138 patients with IgM MGUS progressed after having remained stable for 12 months (Morra et al, 2005). In another series of 83 patients with an IgM-related disorder (type I cryoglobulinaemia in 19, type 2 cryoglobulinaemia in 56, peripheral neuropathy in five and idiopathic thrombocytopenia in three), overt WM or a related disorder developed in eight of the 83 patients (Cesana et al, 2005).
Baldini et al (2005) reported that 15 of 217 patients with IgM MGUS and 45 of 201 patients with indolent WM progressed to symptomatic WM. The variables related to progression were the initial M protein value, haemoglobin value and gender in both groups.
Although the presence of an IgD monoclonal protein almost always indicates MM, AL, or plasma cell leukaemia, there have been rare patients with IgD MGUS (Blade & Kyle, 1994).
Biclonal gammopathies have two different M proteins and occur in 3–6% of patients with monoclonal gammopathies. They may result from the proliferation of two different clones of plasma cells or by production of two M proteins by a single clone of plasma cells. In a series of 57 patients with biclonal gammopathy, 37 had biclonal gammopathy of undetermined significance (Kyle et al, 1981). Two localised bands were found in only 18 patients with electrophoresis on cellulose acetate; while in the remainder, the second M protein was not recognised until immunoelectrophoresis or immunofixation was performed. The clinical findings of biclonal gammopathies are similar to those of patients with monoclonal gammopathies.
In a review of 24 patients with triclonal gammopathy, 16 were associated with a malignant lymphoproliferative disorder, five occurred in non-haematological disease and three were of undetermined significance (Grosbois et al, 1997). In another report, three separate populations of M protein-producing cells were identified in a patient with NHL (Tirelli et al, 2003).
Idiopathic Bence Jones proteinuria
Seven patients have been described with Bence Jones proteinuria (monoclonal light chain >1 g/24 h), but in whom no M protein was found in the serum and who had no evidence of MM or a related disorder (Kyle & Greipp, 1982). MM developed in two, SMM in one, AL in one and two patients died of unrelated causes. One of these patients had excreted up to 1·8 g/24 h of κ light chain for 37 years without evidence of MM or AL at autopsy.
Association of monoclonal gammopathy with other disorders
Certain diseases are associated with MGUS, as would be expected in an older population. The association of two diseases depends on the frequency with which each occurs independently. An apparent association may occur because of differences in the referral practice or in other selected patient groups. Valid epidemiological and statistical methods are essential in evaluating these associations. The need for appropriate control populations cannot be overemphasized. Recent reviews contain more detail concerning the association of M proteins with other diseases (Kyle & Rajkumar, 2004).
Malignant lymphoma is associated with monoclonal gammopathies. In a group of 640 patients with diffuse NHL or CLL, an M protein was found in 44 (7%), but only four of 292 patients with nodular lymphoma and one of 218 patients with Hodgkin lymphoma had an M protein (Alexanian, 1975). In a cohort of 430 patients with an IgM monoclonal protein seen at Mayo Clinic, the following diseases were found: MGUS (56%), WM (17%), malignant lymphoproliferative disease (14%), NHL (7%), CLL (5%) and AL (1%) (Kyle & Garton, 1987). In another series of 382 patients with a lymphoid neoplasm and an IgM monoclonal gammopathy, the following diagnoses were made: WM (59%), CLL (20%), marginal zone lymphoma (7%), follicular lymphoma (5%), mantle cell lymphoma (3%), diffuse B-cell lymphoma (2%) and miscellaneous (4%) (Lin et al, 2005). In another report, seven (27%) of 26 patients with extranodal marginal zone lymphoma had an M protein (Asatiani et al, 2004).
The IgM accounted for 38%, IgG 51%, IgA 1%, and light chain only 10% in a group of 100 patients with CLL and an M protein in the serum or urine (Noel & Kyle, 1987). M proteins have also been recognized in hairy cell leukaemia, chronic myelocytic leukaemia, chronic neutrophilic leukaemia, acute leukaemia, Sézary syndrome and mycosis fungoides.
Acquired von Willebrand disease has been reported with MGUS (Lamboley et al, 2002). On the other hand, thromboembolic events were reported in 19 (6·1%) of 310 patients with MGUS (Sallah et al, 2004). Monoclonal gammopathies have also been reported with myelodysplastic syndrome, idiopathic myelofibrosis, polycythemia vera, paroxysmal nocturnal haemoglobinuria, Gaucher disease, pernicious anaemia and pure red cell aplasia.
Connective tissue disorders
Rheumatoid arthritis, lupus erythematosus, scleroderma and polymyalgia rheumatica have been reported with M proteins.
In a cohort of 279 patients with a sensorimotor peripheral neuropathy of unknown cause, 16 (6%) had MGUS (Kelly et al, 1981). The incidence of MGUS in sensorimotor peripheral neuropathy is variable and depends upon patient selection bias, the vigor with which an M protein is sought, and whether the diagnosis of peripheral neuropathy is made on clinical or electrophysiological grounds.
IgM is the most common M protein associated with peripheral neuropathy. In approximately one-half of patients with an IgM monoclonal gammopathy and peripheral neuropathy, the IgM protein binds to myelin-associated glycoprotein (MAG). The role of antibodies and peripheral neuropathy has been reviewed (Quarles & Weiss, 1999). In our series of 65 patients with MGUS and sensorimotor peripheral neuropathy, 31 had IgM, 24 had IgG and 10 had IgA (Gosselin et al, 1991). Neither the size of the M protein nor anti-MAG activity influenced the type and severity of the neuropathy. MGUS neuropathies differ from those associated with AL in the following fashion: (i) the lower extremities are more often involved in MGUS, while both upper and lower extremities may be involved in AL; (ii) the course of neuropathy in AL is almost always slow progression; and (iii) autonomic features, such as hypotension, anhidrosis, bowel changes, etc. and heart or liver failure, often occur with AL.
Therapy of peripheral neuropathy and monoclonal gammopathy is challenging. Plasmaphaeresis has been of benefit in some patients, while chlorambucil has been helpful in those with IgM MGUS. Melphalan and prednisone may be helpful for IgG and IgA gammopathies. Fludarabine and rituximab have both been reported as producing some benefit, but this is quite modest. Intravenous immunoglobulin infusions have been of minimal benefit. Therapy of neuropathies associated with monoclonal gammopathies has recently been reviewed (Nobile-Orazio, 2005). Monoclonal gammopathy and motor neuron disease, myasthenia gravis, ataxia telangiectasia and nemaline myopathy have all been reported.
POEMS syndrome (osteosclerotic myeloma)
POEMS syndrome is characterized by polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, and skin changes. It is often defined by the presence of a monoclonal plasma cell disorder, peripheral neuropathy, and at least one of the following seven features: osteosclerotic bone lesions, Castleman’s disease, organomegaly, endocrinopathy (excluding diabetes mellitus or hypothyroidism), oedema, typical skin changes and papilledema (Dispenzieri et al, 2003; Rajkumar et al, 2006). The features should have a temporal relationship to each other and no other attributable cause. The presence of single or multiple osteosclerotic lesions are important features. In fact, the absence of osteosclerotic lesions should make the diagnosis suspect. Hypertrichosis, hyperpigmentation, gynecomastia and testicular atrophy may be present. Thrombocytosis or erythrocytosis may occur. Almost all patients have a M protein of the λ light chain type. The serum M protein is modest in size and Bence Jones proteinuria, renal insufficiency, hypercalcaemia and skeletal fractures are rarely seen. The bone marrow usually contains fewer than 5% plasma cells. Elevated levels of IL-1β, TNF-α, IL-6 and vascular endothelial growth factor are frequently present. Radiation therapy in tumouricidal doses if single or multiple sclerotic lesions are found in a limited area is the treatment of choice. If the lesions are widespread, systemic therapy similar to that for myeloma, such as autologous stem cell transplantation or alkylating agent therapy, is indicated. The median duration of survival was 13·8 years in a 99 patient cohort (Dispenzieri et al, 2003).
Lichen myxoedematosus (scleromyxoedema, papular mucinosis) is a rare dermatologic condition usually associated with a cathodal IgG λ protein. Necrobiotic xanthogranuloma and pyoderma gangrenosum are frequently associated with an M protein. Schnitzler syndrome is characterized by the presence of chronic urticaria and an IgM monoclonal protein. Plane xanthomatosis or subcorneal pustular dermatoses have been associated with a monoclonal gammopathy. A review of dermatologic disorders associated with an M protein has been published (Daoud et al, 1999).
The association of hyperparathyroidism and MGUS is controversial. We reported that nine (1%) of 911 patients with hyperparathyroidism at Mayo Clinic had MGUS, which is similar to that in a normal population (Mundis & Kyle, 1982). On the other hand, 20 of 101 patients with hyperparathyroidism had an M protein compared with only two of 127 controls (Arnulf et al, 2002).
Monoclonal proteins are frequently seen after kidney, liver (Badley et al, 1996), heart, or autologous bone marrow transplantation (Zent et al, 1998). In a report of five patients with MGUS undergoing transplantation, SMM developed in two and one other had an increase in the serum M protein (Rostaing et al, 1994).
Capillary leak syndrome (Droder et al, 1992) and acquired C1q inhibitor deficiency (Pascual et al, 1987) have been reported with monoclonal gammopathies. Idiopathic segmental glomerulosclerosis may be associated with MGUS (Dingli et al, 2005). MGUS has been reported following silicone breast implants, but the frequency does not appear to be increased (Karlson et al, 2001). M proteins may be bound to calcium, copper, transferrin, or serum phosphorus. M proteins may also be associated with antibody activity (Merlini et al, 1986).
We reported a cohort of 276 patients with SMM (Kyle et al, 2007). The median age at diagnosis was 64 years (range 26–90 years). Three per cent were younger than 40 years of age. Men accounted for 62% and women 38%. The serum M protein level at diagnosis ranged from 5 to 54 g/l (median 29). Eleven per cent had an M protein ≥40 g/l, while 37% were 30–39 g/l and 52% were <30 g/l. Of the 276 M proteins, 74% were IgG, 22·5% were IgA, 0·5% were IgD and 3% were biclonal. The light chain type was kappa in 67% and lambda in 33%. The level of uninvolved immunoglobulins were reduced in 83% of 230 patients in whom the immunoglobulin levels were measured. Ninety-two (36%) of 259 patients had a monoclonal κ light chain in the urine, 43 (17%) had a λ light chain and 123 (47%) had no monoclonal light chain. Only four patients had more than 1·0 g of light chains per 24 h. The most common plasma cell category was 15–19%. Cyclin D1 was expressed by the plasma cells in 18% of the bone marrow samples.
During 2131 cumulative person-years of follow-up (range 0–29 years; median 6·1 years), 85% of the patients with SMM died (median follow-up of those still living was 11·6 years). During the period of follow-up, active MM developed in 158 patients (57%) who had a median survival after diagnosis of 3·4 years; amyloidosis developed in five (2%) (Table V). The cumulative probability of progression to active MM or AL was 51% at 5 years, 66% at 10 years and 73% at 15 years; the median time to progression was 4·8 years (Fig 5). The overall risk of progression was 10% per year for the first 5 years, approximately 3% per year for the next 5 years and 1% per year for the last 10 years.
Table V. Relative risk of progression in 276 Mayo Clinic patients with smouldering multiple myeloma, by prognostic group.*
BMPC, bone marrow plasma cells; CI, confidence interval; M, monoclonal protein; RR, relative risk.
‡One patient had progression, but the date of progression was not available; therefore, this patient was excluded from the analysis of progression.
§Expected rate of amyloidosis based on data from Olmsted County, Minnesota (Kyle et al, 1992).
Group 1: serum M ≥ 30 g/l and BMPC ≥ 10%
Group 2: serum M < 30 g/l and BMPC ≥ 10%
Group 3: serum M ≥ 30 g/l and BMPC < 10%
Group 1: serum M ≥ 30 g/l and BMPC ≥ 10%
Group 2: serum M < 30 g/l and BMPC ≥ 10%
Group 3: serum M ≥ 30 g/l and BMPC < 10%
Rates of death due to other diseases including cardiovascular and cerebrovascular disease and non-plasma cell cancers were 18% at 5 years, 26% at 10 years, 30% at 15 years and 35% at 20 years. The overall survival of patients was 60% at 5 years, 34% at 10 years and 20% at 15 years (median 6·3 years).
The study cohort was stratified into three groups at initial diagnosis: group 1: bone marrow plasma cells of ≥10% and a serum M protein level of 30 g/l or more; group 2: bone marrow plasma cells of 10% or more and a serum M protein level <30 g/l and group 3: bone marrow plasma cells of <10% and a serum M protein level of 30 g/l or more.
Risk factors for progression
Significant baseline risk factors for progression of SMM to active MM or AL in the univariate and multivariate analyses are shown in Table VI.
On multivariate analysis, the serum M protein level and the number of plasma cells in the bone marrow emerged as significant independent risk factors for progression. The cumulative probability of progression at 15 years was 87% in the 106 patients in group 1 (plasma cells ≥10% and serum M protein ≥ 30 g/l); 70% in the 142 patients in group 2 (plasma cells ≥10% and serum M protein <30 g/l); and 39% in the 27 patients in group 3 (bone marrow plasma cells of <10% and a serum M protein of ≥30 g/l) (Fig 3). The median time to progression was 2 years in group 1, 8 years in group 2 and 19 years in group 3 (P < 0·001). The type of serum heavy chain added significantly to the multivariate model containing the three prognostic risk groups (Table VI).