SEARCH

SEARCH BY CITATION

Keywords:

  • multiple myeloma;
  • smoldering;
  • prognosis;
  • chemotherapy;
  • malignant transformation

Abstract

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

Summary.  Patients with smoldering multiple myeloma (SMM) meet the diagnostic criteria of multiple myeloma (MM) but are asymptomatic. Between January 1978 and July 2001, 53 patients (median age 63 years) were diagnosed with SMM. The median serum M-protein and proportion of bone marrow plasma cells were 36 g/l and 27% respectively. Two subsets of SMM were identified: (i) evolving SMM (n = 22), characterized by a progressive increase in serum M-protein, a previously recognized monoclonal gammopathy of undetermined significance (MGUS) and a significant higher proportion of IgA type and (ii) non-evolving SMM (n = 26) with stable M-protein that abruptly increases when symptomatic MM develops. Thirty-four patients developed symptomatic MM. The median time to progression in the overall series was 3·2 years and the only feature associated with a shorter time to progression was the evolving versus non-evolving type (1·3 vs. 3·9 years respectively, P = 0·007). The pattern of progression consisted of anaemia, lytic bone lesions or both, without renal failure, hypercalcaemia or extramedullary plasmacytomas. Fifty-seven per cent of patients that required chemotherapy showed no or minimal response. The median survival from diagnosis and from progression was 8·2 and 3·5 years respectively.

The majority of patients with multiple myeloma (MM) have skeletal involvement, anaemia, renal function impairment, hypercalcaemia, recurrent bacterial infections or extramedullary plasmacytomas requiring cytotoxic therapy from the time of diagnosis, with a median survival of less than 3 years (Myeloma Trialists’ Collaborative Group, 1998). Kyle and Greipp (1980) described the so-called smoldering multiple myeloma (SMM) in six patients who fulfilled the diagnostic criteria of MM (i.e. a serum M-protein >30 g/l and a proportion of bone marrow plasma cells ≥10%), but who had no lytic lesions or symptoms because of the monoclonal gammopathy. These patients had stable disease not requiring chemotherapy for five or more years. Several series of patients with asymptomatic (Alexanian et al, 1988, Wisloff et al, 1991, Hjorth et al, 1993, Dimopoulos et al, 1993, 2000, Weber et al, 1997) or low tumour burden (Facon et al, 1995) myeloma have been reported. However, there are no reports that analysed the natural history of patients with SMM fulfilling the stringent criteria defined by Kyle and Greipp (1980).

The aims of the present study were to analyse the clinical and laboratory features, the pattern of progression, and the outcome after progression (response to therapy and survival) in 53 patients with SMM.

Patients and diagnostic criteria.

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

The records of 596 patients with MM diagnosed at our institution between 1 January 1978 and 31 July 2001 were reviewed; 53 (9%) fulfilled the criteria of SMM. The diagnosis of SMM was based on the following criteria: (i) proportion of bone marrow plasma cells ≥10%, (ii) a serum M-protein >30 g/l or light chain protein excretion in a 24-h urine specimen >1 g, (iii) serum haemoglobin >10 g/dl, and (iv) absence of lytic bone lesions, renal insufficiency, hypercalcaemia, extramedullary plasmacytomas, recurrent bacterial infections and symptoms attributable to the monoclonal gammopathy.

Criteria of progression.

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

During the asymtomatic phase, the majority of patients were followed every 3 months during the first year and then every 4 months. The development of bone pain because of lytic bone lesions or compression fractures from severe osteoporosis, symptoms of anaemia with haemoglobin level <9 g/dl, renal failure, hypercalcaemia, recurrent bacterial infections or extramedullary plasmacytomas were the criteria for progressive disease and initiation of therapy. An increase in serum M-protein was not a criterion per se for the initiation of therapy.

Criteria of response to therapy.

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

The response was assessed according to the European Group for Bone and Marrow Transplantation (EBMT), the International Bone marrow Transplantation Registry (IBMTR) and the Autologous Blood and Marrow Transplant Registry (ABMTR) (Bladéet al, 1998).

Statistical methods.

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

The χ2, Mann–Whitney's U or Student's t-test were used to assess the statistical significance of the comparison of different patient's characteristics and response to therapy. Time to progression and survival were estimated using the method of Kaplan and Meier (1958) and statistically compared by means of the log-rank test (Peto & Pike, 1973). A P-value <0·05 was considered statistically significant.

Features at diagnosis

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

The main features of the 53 patients at diagnosis are shown in Table I. The median M-protein level and percentage of bone marrow plasma cells (BMPC) were 36 g/l and 27% respectively. The M-protein type was immunoglobulin class G (IgG) in 75% of cases. Fifteen patients (28%) had a previously recognized monoclonal gammopathy of undetermined significance (MGUS).

Table I.  Clinical and biological parameters at diagnosis according to the type of SMM.
 AllNon-evolving SMMEvolving SMM
  1. *P = 0·018.

  2. †Five patients could not be classified as evolving or non-evolving SMM as not enough data were available.

  3. ‡Normal value <2·8 mg/dL.

  4. §The bone marrow biopsy showed plasmacytic cell infiltration.

  5. ¶One patient not classified as evolving or non-evolving SMM.

Number of patients53†2622
Sex (M/F)22/317/1911/11
Age (years) (median, range)63 (35–91)66 (40–91)59 (44–83)
Haemoglobin (g/dl) (mean ± SD)12·4 ± 1·812·2 ± 1.412·2 ± 1·8
Beta-2-microglobulin (mg/l) (mean ± SD)‡  2·6 ± 1·1  2·4 ± 1·6  1·95 ± 1·6
Serum M-protein (g/l) (median, range)35·5 (30–55)35 (30·2–55)34·7 (30–49·8)
Bone marrow plasma cells (%) (median, range)26·5 (3–100)§25·5 (3–100)§26 (11–60)
M-protein type
 IgG402313
 IgA11  2*  8*
 Light chain  1  1  0
 Biclonal (IgA-λ, IgG-λ)  1  0  1
Light chain
 Kappa351815
 Lambda18  8  7
Previous MGUS (%)15¶ (28%)  1 (3·8%)13 (59%)

Two types of SMM were identified: (i) evolving SMM (n = 22) characterized by a constant increase in the serum M-protein at each visit and (ii) non-evolving SMM, with long-lasting stable serum or urinary M-protein until the time of progression. Thirteen of the 22 patients with evolving type (59%) had a previously recognized MGUS as compared with only one of the 26 patients with non-evolving SMM (P = 0·00001). The evolution of serum M-protein according to the type of SMM is illustrated in Fig 1A and B. The evolving type was defined by an increase in the M-protein level in each of the first two consecutive follow-up visits. Seventeen of the 22 patients with evolving SMM had an increase in their serum M-protein level ≥10% during the first 6 months of follow-up, while in the remaining five patients the increase in the serum M-protein ranged between 5% and 9%, and showed a progressive increase throughout the subsequent course of the disease. The patients with non-evolving disease did not show a confirmed increase in the serum M-protein at the first two consecutive follow-ups. The IgA type was more frequent in the evolving type (36% vs. 7%, P = 0·018). There were no significant differences in the other features between evolving and non-evolving SMM. Five patients could not be classified as either evolving or non-evolving because of insufficient serial M-protein measurements.

image

Figure 1. (A) Evolution of M-serum protein level in patients with evolving SMM. (B) Evolution of M-serum protein level in patients with non-evolving SMM (the patient with Bence-Jones SMM is not included).

Download figure to PowerPoint

Natural history

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

Thirty-four of the 53 patients evolved to a symptomatic MM. The main laboratory features at the time of progression and the pattern of progression are depicted in Table II. The type of M-protein did not influence the progression rate. In almost all patients the pattern of disease progression consisted of anaemia, lytic bone lesions or both. Although slightly more than 50% of the patients developed lytic bone lesions, no extensive skeletal involvement was observed. No patient developed renal failure, hypercalcaemia or extramedullary plasmacytomas at the time of progression. Sixteen of the 22 patients with evolving SMM developed overt MM, compared with 14 of the 26 with non-evolving SMM. The median time to progression in the overall series was 3·2 years. The median time to transformation for evolving and non-evolving type was 1·3 vs. 3·9 years respectively (Fig 2, P = 0·007). The actuarial probability of evolution to symptomatic MM at 2 and 5 years for the evolving or non-evolving type was 66% and 88% vs. 12% and 58% respectively.

Table II.  Patients characteristics at progression.
 GlobalNon-evolving SMMEvolving SMM
  1. *Four patients unclassified as evolving or non-evolving SMM.

  2. †Normal value <2.8 mg/dl.

Number of patients34*1416
Hemoglobin (g/dl) (mean ± SD)10·6 ± 2·4  9·9 ± 3·210·5 ± 1·9
Beta-2-microglobulin (mg/l) (mean ± SD)†  4·3 ± 1·4  4·4 ± 1·2  3·9 ± 1·1
Serum M-protein (g/l) (median, range)52·1 (31–87)52 (31–87)52·9 (34–71)
Bone marrow plasma cells (%) (median, range)40 (24–100)36 (24–70)47 (27–100)
Pattern of progression:
 Anaemia11  5  6
 Lytic lesions  6  2  4
 Anaemia and lytic lesions12  6  5
 Others (bone pain, unknown)  5 (3, 2)  1  1
image

Figure 2. Time to progression according to SMM type (evolving, non-evolving).

Download figure to PowerPoint

The only feature significantly associated with a shorter time to disease progression was the type of SMM (evolving versus non-evolving, P = 0·007). Patients with a lower Hb level (<12 g/l) and a higher serum M-protein (>35 g/l) showed a trend towards a greater probability of disease progression (P = 0·08 for both variables). The proportion of BMPC as well as the type of heavy and light chain had no impact on disease progression.

Response to therapy and survival

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

Thirty-three of the 34 patients who evolved to a symptomatic MM were given cytotoxic therapy: 16 received a single alkylating agent (melphalan, n = 15; cyclophosphamide, n = 1) plus prednisone and 17 were given combination chemotherapy [VBMCP/VBAD (vincristine, carmustine (BCNU), melphalan, cyclophosphamide, prednisone/vincristine, BCNU, Adriamycin, Dexamethasone), n = 11; VCMP (vincristine, cyclophosphamide, melphalan, prednisone), n = 1; VCMP/VBAP (vincristine, BCNU, Adriamycin, prednisone), n = 5]. One patient refused treatment with chemotherapy and died from pneumonia shortly after progression while other patient has not yet been evaluated for response. In addition, in two patients could not be evaluated for response because of lack of data. Among the 30 evaluable patients, the overall response rate was 63% (partial response: 43%, minimal response: 20%). Eleven patients (37%) did not show any response (17% no change, 20% progression). The partial response rate was 53% in patients with evolving SMM versus 21% for patients with non-evolving disease (P = 0·08). Seven patients were subsequently given high-dose therapy/autologous stem cell support (six as early intensification and one as salvage because of refractory disease). This latter patient achieved a transient complete response followed by an aggressive relapse with extramedullary involvement. Among the remaining six patients, two achieved a partial response, three a minimal response and one showed no change after high dose therapy.

The median survival of the 53 patients from the diagnosis of SMM was 8·3 years (Fig 3). The survival from diagnosis was not significantly different in patients with evolving or non-evolving SMM type (median, 6·3 vs. 8·8 years; P = 0·25). The median survival from disease progression in the overall series was 3·5 years, with no significant differences between the evolving and non-evolving types (Fig 4; median, 4·2 vs. 3·3 years; P = 0·59).

image

Figure 3. Overall survival from diagnosis for all the SMM patients.

Download figure to PowerPoint

image

Figure 4. Survival from progression to MM in both evolving and non-evolving SMM patients (P = NS).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References

Most patients with MM have symptomatic disease and require chemotherapy from the time of diagnosis. Kyle and Greipp (1980) coined the term SMM for a subset of patients with a high serum M-protein and a significant proportion of bone marrow plasma cells, with no lytic lesions. The first six patients with recognized SMM remained in a stable condition with no chemotherapy requirement for many years. Since then, several series of patients with asymptomatic disease, referred to as asymptomatic or low tumour mass myeloma, have been reported (Alexanian et al, 1988, Wisloff et al, 1991, Hjorth et al, 1993, Dimopoulos et al, 1993, 2000, Facon et al, 1995, Weber et al, 1997) with particular emphasis on predictive factors for disease progression. However, the diagnostic criteria in the above studies were different from those of the original report (Kyle & Greipp, 1980). Thus, the MD Anderson Cancer Center group required a serum M-protein >20 g/l in one study (Alexanian et al, 1988) and >25 g/l in a more recent series (Weber et al, 1997) while the requirement by Wisloff et al (1991) was >15 g/l for IgA and >30 g/l for IgG. The series by Facon et al (1995) was composed of asymptomatic patients with >15% BMPC and an M-protein in serum or urine, regardless of the amount. In contrast with the original report, most of the above series included some patients with lytic bone lesions (Alexanian et al, 1988, Wisloff et al, 1991, Hjorth et al, 1993, Dimopoulos et al, 1993, Facon et al, 1995). As the presence of lytic lesions is consistently associated with a short time to progression (Alexanian et al, 1988, Wisloff et al, 1991, Dimopoulos et al, 2000), patients with lytic bone disease should be considered as having indolent rather than smoldering disease. In the current study, only those patients that fulfilled the stringent criteria originally reported (Kyle & Greipp, 1980) have been included.

In the present series two types of SMM were recognized for the first time: evolving and non-evolving. Both types have similar laboratory features at presentation, but patients with evolving SMM showed a constant and progressive increase in the serum M-protein level until symptomatic myeloma developed while patients with non-evolving SMM were characterized by a stable M-protein until the onset of symptomatic disease. Of interest, the evolving type could be identified after the first two follow-up measurements. In fact, most patients with evolving SMM had an increase ≥10% during the first 6 months of follow-up. It is also of note that more than 50% of our patients with evolving SMM had a previous long-lasting MGUS, also with an evolving pattern, while a previous MGUS was rarely seen in patients with non-evolving SMM. It could be speculated that, all patients with an evolving pattern have had a previous MGUS, with evolving SMM being a transient step between MGUS and symptomatic MM, with a short time to transformation. In contrast, non-evolving SMM is a more stable condition with a longer time to transformation that is rarely preceded by MGUS. In fact, the clinical and biological behaviour of the latter condition would be comparable with MGUS, but with a higher plasma cell burden and, in consequence, with a higher probability of evolution to symptomatic MM (Kyle et al, 2002, Montoto et al, 2002). It is likely that evolving and non-evolving conditions have different pathogenetic mechanisms.

Concerning the pattern of progression to symptomatic myeloma, most of our patients progressed clinically, with increasing anaemia and/or skeletal involvement consisting of lytic lesions that were generally moderate in size, or diffuse osteoporosis. Of note, none of our patients developed renal function impairment, hypercalcaemia or extramedullary plasmacytomas at the time of progression. This is in contrast with the findings of Wisloff et al (1991) who found that six and three of their 71 patients developed renal insufficiency or neurological symptoms respectively, at the time of progression. In the remaining series, the pattern of progression consisted of increasing bone disease, increasing M-protein or development of anaemia, with only occasional cases developing renal failure or hypercalcaemia (Alexanian et al, 1988, Hjorth et al, 1993, Dimopoulos et al, 1993, 2000, Facon et al, 1995, Weber et al, 1997). The fact that SMM patients do not develop renal insufficiency during the asymptomatic phase or at progression indicates the lack of nephrotoxic potential of their light chains. Furthermore, the absence of extensive skeletal involvement, hypercalcaemia or extramedullary plasmacytomas is consistent with a low proliferative activity in SMM, even at the time of progression. All the above reassures the practicing physician of the relative safety of the watch-and-wait approach in the management of SMM. This is particularly important in non-evolving SMM patients as more than 40% of them will not have developed symptomatic myeloma at the 5-year follow-up.

Despite the stringent criteria employed in our study, the time to progression was similar (median of about 3 years) to that observed in other studies that included patients with a lower tumour burden (Alexanian et al, 1988, Wisloff et al, 1991, Hjorth et al, 1993, Dimopoulos et al, 1993, 2000, Facon et al, 1995, Weber et al, 1997). In our series, the only significant predictor for an earlier progression was the type of SMM (evolving versus non-evolving). In the MD Anderson series (Alexanian et al, 1988, Weber et al, 1997), an M-protein >30 g/l, IgA type and light chain proteinuria >50 mg/24 h were significantly associated with a shorter time to progression, whereas in the series reported by Facon et al (1995) the significant factors were an M-protein >30 g/l, Hb level <12 g/dl, and a proportion of BMPC >25%. In our series, the M-protein level, which was higher than in the previous studies, and the Hb level showed only a trend towards an earlier progression.

The reported response rate to chemotherapy ranges between 52% and 64% (Alexanian et al, 1988, Hjorth et al, 1993, Dimopoulos et al, 1993, Weber et al, 1997). However, no details on the degree of response were given in these published series. Although the overall response rate was 63% in our study, it must be emphasized that only 43% of the patients achieved a partial response, whereas 57% of the patients had minimal or no response to the initial chemotherapy. Patients with evolving SMM had a trend towards a higher partial response rate than those with non-evolving disease (53% vs. 21%). However, the difference was not statistically significant, probably because of the low number of patients included in each group. It is noteworthy that four of the seven patients who underwent autologous transplantation only achieved minimal response (three cases) or no change (one case) with high-dose therapy. Our results indicate that the response to therapy in SMM is characterized by a moderate tumour reduction or no significant change. However, despite the relative poor response to treatment the median survival after progression was 3·5 years, which is slightly longer that that reported in the general myeloma population. This again supports the indolent nature of the disease, even after progression, with an overall median survival longer than 8 years from diagnosis.

In summary, two subsets of SMM were identified: (i) evolving disease with a progressive increase in the M-protein level, a previously recognized MGUS, a higher frequency of IgA type and short time to progression and (ii) patients with non-evolving disease with no previous MGUS and a longer time to transformation. In both types the pattern of progression mainly consisted of anaemia with or without bone lytic lesions with no renal failure, hypercalcaemia or extramedullary plasmacytomas at the time of progression. Thus, patients with SMM can be safely observed for long periods of time. In addition, the response to therapy after progression is poor, particularly in the non-evolving type.

References

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Patients and diagnostic criteria.
  5. Criteria of progression.
  6. Criteria of response to therapy.
  7. Statistical methods.
  8. Results
  9. Features at diagnosis
  10. Natural history
  11. Response to therapy and survival
  12. Discussion
  13. Acknowledgments
  14. References
  • Alexanian, R., Barlogie, B. & Dixon, D. (1988) Prognosis of asymptomatic multiple myeloma. Archives Internal Medicine, 148, 19631965.
  • Bladé, J., Samson, D., Reece, D., Apperley, J., Bjorkstrand, B., Gahrton, G., Gertz, M., Giralt, S., Jagannath, S. & Vesole, D. (1998) Criteria for evaluating disease response and progression in patients with multiple myeloma treated by high dose therapy and hematopoietic stem cell transplantation. British Journal of Haematology, 102, 11151123.
  • Dimopoulos, M., Moulopoulos, L., Smith, T., Delasalle, K. & Alexanian, R. (1993) Risk of disease progression in asymptomatic multipe myeloma. American Journal of Medicine, 94, 5761.
  • Dimopoulos, M., Monopoulos, L., Maniatis, A. & Alexanian, R. (2000) Solitary plasmacytoma of bone and asymptomatic multiple myeloma. Blood, 96, 20372044.
  • Facon, T., Menard, J.F., Michaux, J.L., Euller-Ziegher, L., Bernard, J.F., Grosbois, B., Daragon, A., Azais, I., Courouble, Y., Kaplan, G., Laporte, J.P., De Gramont, A., Duclos, B., Leonard, A., Mineur, P., Delannoy, A., Jouet, J.P., Bauters, F. & Monconduit, M. (1995) Prognostic factors in low tumor mass asymptomatic multiple myeloma: a report on 91 patients. American Journal of Hematology, 48, 7175.
  • Hjorth, M., Hellquist, L., Homberg, E., Magnusson, B., Rödjer, S. & Westin, J. (1993) Initial versus deferred Melphalan-prednisone therapy for asymptomatic multiple myeloma stage I – A randomized study. European Journal of Haematology, 50, 95102.
  • Kaplan, E.L. & Meier, P. (1958) Non-parametric estimation from incomplete observations. Journal American Statistically Association, 53, 457481.
  • Kyle, R.A. & Greipp, P.R. (1980) Smoldering multiple myeloma. New England Journal of Medicine, 302, 13471349.
  • Kyle, R., Thernau, T., Rajkumar, V., Offord, J.R., Larson, D.R., Plevak, M.F. & Melton, L.J. III (2002) A long-term study of prognosis in monoclonal gammopathy of undetermined significance. New England Journal of Medicine, 346, 564569.
  • Montoto, S., Bladé, J. & Montserrat, E. (2002) Monoclonal gammopathy of undetermined significance. New England Journal of Medicine, 346, 20872088 (Letter to the Editor).
  • Myeloma Trialists’ Collaborative Group (1998) Combination chemotherapy versus melphalan plus prednisone as treatment for multiple myeloma: an overview of 6·633 patients from 27 randomized trials. Journal Clinical of Oncology, 16, 38323842.
  • Peto, R. & Pike, M.C. (1973) Conservation of the aproximation (O-E) /E in the logrank test for survival data or tumor incidence data. Biometrics, 29, 579584.
  • Weber, D., Dimopoulos, M., Monopoulos, L., Delasalle, K., Smith, T. & Alexanian, R. (1997) Prognostic features of asymptomatic multiple myeloma. British Journal of Haematology, 97, 810814.
  • Wisloff, F., Andersen, P. & Andersson, T.R., (1991) Incidence and follow-up of asymptomatic multiple myeloma. European Journal of Haematology, 47, 338341.