Serum free light chain ratio as a biomarker for high-risk smoldering multiple myeloma

Transcription

1 Leukemia (2013) 27, & 2013 Macmillan Publishers Limited All rights reserved /13 ORIGINAL ARTICLE Serum free light chain ratio as a biomarker for high-risk smoldering multiple myeloma JT Larsen 1, SK Kumar 2, A Dispenzieri 2, RA Kyle 2, JA Katzmann 3 and SV Rajkumar 2 A markedly elevated serum free light chain (FLC) ratio may serve as a biomarker for malignant transformation in high-risk smoldering multiple myeloma (SMM) and identify patients who are at imminent risk of progression. We retrospectively studied the predictive value of the serum (FLC) assay in 586 patients with SMM diagnosed between 1970 to A serum involved/uninvolved FLC ratio X100 was used to define high-risk SMM, which included 15% (n ¼ 90) of the total cohort. Receiver operating characteristics analysis determined the optimal FLC ratio cut-point to predict progression to symptomatic multiple myeloma (MM) within 2 years of diagnosis, which resulted in a specificity of 97% and sensitivity of 16%. Fifty-six percent of patients developed progressive disease during median follow-up of 52 months, but this increased to 98% in the subgroup of patients with FLC ratio X100. The median time to progression in the FLC ratio X100 group was 15 months versus 55 months in the FLC o100 group (Po0.0001). The risk of progression to MM within the first 2 years in patients with an FLC ratio X100 was 72%; the risk of progression to MM or light chain amyloidosis in 2 years was 79%. We conclude that a high FLC ratio X100 is a predictor of imminent progression in SMM, and such patients may be considered candidates for early treatment intervention. Leukemia (2013) 27, ; doi: /leu Keywords: smoldering multiple myeloma; free light chain ratio; prognosis; biomarker INTRODUCTION Smoldering multiple myeloma (SMM) is an asymptomatic precursor disease of multiple myeloma (MM) or related plasma cell disorders such as light chain (AL) amyloidosis for which the standard of care has remained observation without therapy until symptoms develop. 1 SMM is a clinical diagnosis currently defined by the International Myeloma Working group as the presence of a serum M-protein of X3 g/dl and/or X10% bone marrow plasma cells (BMPCs) with no evidence of end-organ damage (hypercalcemia, renal insufficiency, anemia or bone lesions (CRAB)). 2 Early treatment of SMM previously has been limited because of unacceptably high rates of treatment-related toxicity, and an inability to identify high-risk SMM patients at diagnosis who will consistently progress to symptomatic MM within a short period. 3 5 The emergence of novel myeloma therapies that are highly active yet have improved tolerability has generated renewed interest in the potential for treatment of SMM to prevent later development of end-organ damage. Rather than a strategy of watchful waiting for all SMM patients until the development of CRAB features, early treatment of the highest risk patients has the potential to improve quality of life and overall survival. 6 The annual risk of progression to symptomatic disease in SMM is 10% per year for the first 5 years, 7 making SMM a more attractive target for early therapy compared with monoclonal gammopathy of undetermined significance, with the established 1% per year risk of progression. 8 However, the identification of SMM patients at diagnosis who will invariably progress to symptomatic MM in a short timeframe remains a challenge. The development of biomarkers capable of distinguishing high- and low-risk SMM patients may allow for more individualized disease management. Risk stratification schemes for SMM recently have emerged utilizing methods such as immunophenotyping of aberrant plasma cells, 9 BMPC percentage, 10 serum M-protein, as well as the serum free light chain (FLC) ratio. 11 The clinical utility of the serum FLC assay has been demonstrated previously in a range of related conditions such as monoclonal gammopathy of undetermined significance, MM and AL amyloidosis It is a sensitive method for detecting excess light chain immunoglobulins, which results in an abnormal k/l ratio. 15,16 Applying the FLC assay to SMM, Dispenzieri et al. 11 determined that the prognostic effect of an abnormal k/l FLC ratio of p0.125 or X8 was independent of other risk factors on multivariate analysis, and was associated with a hazard ratio of 2.3 (95% confidence interval (CI), ) compared with FLC ratios of We hypothesize that SMM is a clinical entity comprised of both premalignant, high-risk monoclonal gammopathy of undetermined significance and early asymptomatic MM in transition to malignant disease, which may be differentiated with the use of the serum FLC ratio set to an appropriate level of sensitivity. We examined the diagnostic performance of the serum FLC ratio at the time of SMM diagnosis and its ability to predict risk of imminent progression to symptomatic MM within 2 years of diagnosis, which may be particularly relevant for use in prospective early intervention studies. PATIENTS AND METHODS Study cohort Approval for the study was obtained from the Mayo Clinic Institutional Review Board according to federal regulations, and in accordance with the Declaration of Helsinki. According to state law we excluded patients who 1 Division of Hospital Internal Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA; 2 Division of Hematology, Department of Medicine, Mayo Clinic, Rochester, MN, USA and 3 Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA. Correspondence: Dr SV Rajkumar, Division of Hematology, Department of Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. rajkumar.vincent@mayo.edu Received 15 September 2012; accepted 9 October 2012; accepted article preview online 16 October 2012; advance online publication, 27 November 2012

2 942 declined research authorization to review their medical records for research purposes. Patients were identified by searching a computerized database and reviewing the medical records of patients meeting the International Myeloma Working Group 2003 definition of SMM 2 : X10% BMPCs and/or serum M-protein X3 g/dl, plus absence of hypercalcemia (calcium mg/dl), renal insufficiency (creatinine 42 mg/dl), anemia (hemoglobin o10 g/dl) or bone lesions (lytic lesions or diffuse osteopenia) attributable to a plasma cell disorder. We performed a retrospective analysis of 586 newly diagnosed SMM patients seen at Mayo Clinic from 1970 to 2010 with available stored serum samples within 30 days of SMM diagnosis. Patients who had received prior chemotherapy or had an existing diagnosis of AL amyloidosis at the time of SMM diagnosis were excluded. Serum FLC assay The serum FLC assay (Freelite; The Binding Site, Birmingham, UK) was performed on a BNII nephelometer (Dade Behring, Deerfield, IL, USA) using stored serum. 17,18 The serum FLC assay measures absolute free k and l light chain values and by default is reported as the k/l ratio (normal range: ). 16 To simplify interpretation of this study, we reported the FLC involved/uninvolved ratio with the monoclonal light chain in the numerator. Kappa was the involved light chain ratio if the k/l ratio was 41.65, whereas lambda was the involved light chain if the k/l ratio was o0.26. The absolute difference between the involved and uninvolved light chain was determined. Statistical analysis Calculations were performed using JMP version 9.0 (SAS Institute Inc., Cary, NC, USA). Receiver operating characteristics (ROC) curves were constructed by plotting sensitivity (Y axis) versus 1-specificity (X axis) and the area under the curve was calculated. The ROC curve was used to determine the ability of the serum involved/uninvolved FLC ratio to discriminate patients who progressed to symptomatic myeloma at 12-month intervals. The sensitivity and specificity were calculated across a range of values in order to choose a cut-point with the highest possible specificity (around 95%) while also maximizing sensitivity. Positive and negative likelihood ratios, as well as positive and negative predictive values were determined. The w 2 test was used to compare nominal variables. The unpaired t-test was used to compare the median values of continuous variables between the high and low FLC ratio groups. Time to progression (TTP) was the primary end point and was measured from the date of SMM diagnosis until progression to active disease. Seventy-one patients with o24 months follow-up and without disease progression were censored from the Kaplan Meier survival analysis. Kaplan Meier analysis was performed to generate survival curves. Groups were compared with the two-tailed logrank test. All statistical tests were two-sided and P-values of o0.05 were considered to be significant. RESULTS Patient characteristics A total of 586 patients were available for analysis. Patient characteristics at SMM diagnosis are shown in Table 1. Median follow-up for the composite group was 52 months. In all, 109 patients had o24 months follow-up and 71 of these patients (65%) showed no evidence of developing MM during this period. Of the 71 patients who did not progress, only 1 patient had an FLC ratio of X100 (ratio 317; k mg/dl, l 312 mg/dl). The median age at SMM diagnosis was 64 years (range, years). Men represented 54% of the patient sample (n ¼ 319). The immunoglobulin (Ig) heavy chain type was 73% monoclonal IgG, 20% IgA, 1% IgM, 1% IgD and 2% had a biclonal M-protein. Light chain only disease was present in 4% of patients, however, an enrichment in Table 1. Patient characteristics at SMM diagnosis grouped according to FLC ratio X100 or o100 Variables All patients FLC ratio o100 FLC ratio X100 P (95% CI) Patients (n) (85%) 90 (15%) Age, years Sex, male (%) 319 (54%) 278 (56%) 41 (46%) 0.08 Serum monoclonal protein Median M-spike (g/dl) o3 g/dl 378 (67%) 324 (65%) 54 (60%) 0.22 X3 g/dl 186 (33%) 152 (32%) 34 (39%) 0.22 Immunoglobulin heavy chain IgG 426 (73%) 367 (74%) 59 (66%) 0.09 IgA 115 (20%) 98 (20%) 17 (19%) 0.84 IgM 3 (1%) 3 (0.6%) 0 (0%) 0.46 IgD 6 (1%) 5 (1%) 1 (1%) 0.59 Biclonal 13 (2%) 13 (3%) 0 (0%) 0.12 Light chain only 23 (4%) 10 (2%) 13 (14%) o0.0001* Involved light chain Kappa 372 (63%) 320 (65%) 52 (58%) 0.15 Lambda 214 (37%) 176 (35%) 38 (42%) 0.24 Median light chain concentration (mg/dl) Kappa 3.02 ( ) o0.0005* Lambda 1.26 ( ) * BMPCs Median BMPCs (%) o0.0001* 10 60% 557 (95%) 478 (96%) 79 (88%) 0.002* 460% 29 (5%) 18 (4%) 11 (12%) 0.002* Median difference in the involved and uninvolved FLC, mg/dl (range) 6.9 ( ) 5.2 ( ) 85 ( ) o0.0001* Abbreviations: BMPC, bone marrow plasma cell; CI, confidence interval; FLC, free light chain; SMM, smoldering multiple myeloma. *Indicates statistically significant P-values. Leukemia (2013) & 2013 Macmillan Publishers Limited

3 light chain only disease was observed in the high FLC ratio group (12% of patients; Po0.0001). The dominant light chain was k in 63% and l in 37%. The FLC ratio was abnormal (reference range o0.26 or 41.65) in 74% of patients (n ¼ 432). In all, 90 of the 586 SMM patients (15%) had a FLC ratio X100 at time of diagnosis. The median value of the involved/uninvolved FLC ratio was 12.9 (range, ). The median monoclonal protein size was 2.5 g/dl Figure 1. ROC curve demonstrating the sensitivity and specificity of the initial involved/uninvolved FLC ratio for progression to symptomatic multiple myeloma within 24 months of SMM diagnosis. The FLC ratio cut-point of X100 has a specificity of 96.7% and sensitivity of 15.8%. (range, 0 6.9). The median BMPC percentage was 20% (range, 3 95%), and was significantly higher in the FLC ratio 4100 group compared with the o100 group (30% versus 20%, Po0.0001). Outcomes By ROC analysis, the optimal involved/uninvolved FLC ratio diagnostic cut-point for differentiating SMM patients at followup through years 1 5 was 491. The ROC curve is shown in Figure 1. The sensitivity and specificity at each year is listed in Table 2. For ease of clinical application, the optimal value for the involved/uninvolved FLC ratio was rounded to X100. An FLC ratio cut-point of X100 corresponded to a sensitivity of 16% (95% CI, ), specificity of 97% ( %), positive likelihood ratio of 5.1 ( ), negative likelihood ratio of 0.87 ( ), positive predictive value of 73 ( ) and negative predictive value of 68.1 ( ). The area under the ROC curve (AUC) was During the follow-up period, 56% (n ¼ 331) of SMM patients progressed to symptomatic multiple myeloma (Table 3). Ten patients developed AL amyloidosis. Median TTP by Kaplan Meier analysis was 40 months (95% CI, 33 48) and 35% of all patients progressed within 2 years. Only 48% of the FLC o100 group developed active MM, however, 98% of the FLC X100 patients ultimately developed disease progression during the follow-up period (relative risk (RR) 2.04 (95% CI, )). The absolute difference between the involved and uninvolved light chain was significantly higher at 85 mg/dl in the FLC ratio X100 group compared with 5.2 mg/dl in the FLC o100 group (Po0.0001). The most common progression event was bone disease (43%), followed by anemia (34%), renal insufficiency (18%) and hypercalcemia (5%). No significant differences were observed in the type of progression event between the FLC o100 and X100 groups. Progression data are shown in Table 3. TTP from the date of the initial FLC ratio at SMM diagnosis to active MM was compared between patients in the high (X100) and low (o100) FLC ratio groups. Median TTP to symptomatic MM in the FLC ratio X100 group was 15 months (95% CI, 9 17) versus 55 months (95% CI, 46 65) in the FLC ratio o100 group (P o0.0001; Figure 2. In the 943 Table 2. ROC analysis of the FLC involved/uninvolved ratio at interval cut-points and time points FLC ratio Progression to symptomatic myeloma over time (n, %) 12 Months 24 Months 36 Months 48 Months 60 Months (n ¼ 115, 20%) (n¼ 205, 35%) (n¼ 253, 43%) (n¼ 280, 48%) (n¼ 302, 52%) FLC 48 Specificity 69% 71% 73% 73% 75% Sensitivity 50% 44% 44% 42% 42% FLC 425 Specificity 83% 86% 88% 88% 89% Sensitivity 37% 34% 34% 31% 31% FLC 450 Specificity 90% 92% 94% 94% 95% Sensitivity 25% 23% 23% 21% 21% FLC 470 Specificity 92% 93% 96% 96% 97% Sensitivity 20% 18% 18% 17% 17% FLC 491* Specificity 95% 96% 99% 99% 99% Sensitivity 19% 17% 17% 15% 15% Abbreviations: FLC, free light chain; ROC, receiver operating characteristic. *The optimal cut-point to obtain a specificity near 95% at 24 months while still maximizing sensitivity was determined to be 491. & 2013 Macmillan Publishers Limited Leukemia (2013)

4 944 Table 3. Patient outcomes according to FLC ratio X100 or o100 Variables All patients FLC o100 FLC X100 RR (95% CI) Progression to MM, % 56% 48% 98% 2.04 ( ) Within 1 year 19% 15% 43% 1.5 ( ) Within 2 years 35% 28% 72% 2.6 ( ) Within 3 years 47% 40% 87% 4.4 ( ) Progression event, n (%) Bone disease 43% 41% 44% 1.0 ( ) Anemia (Hgb o10.0 g/dl) 34% 37% 23% 0.81 ( ) Renal insufficiency (Cr X2.0 g/dl) 18% 16% 27% 1.2 ( ) Hypercalcemia (411.5 mg/dl) 5% 6% 6% 1.0 ( ) Median TTP to active MM (months) 40 (33 48) 55 (46 65) 15 (9 17) Po Abbreviations: CI, confidence interval; Cr, serum creatinine; FLC, free light chain; Hgb, hemoglobin; MM, symptomatic myeloma; RR, risk ratio; TTP, time to progression. Table 4. Univariate analysis of bone marrow plasma cell percentage, serum monoclonal protein and FLC ratio 4100 on time to progression Prognostic variable Hazard ratio 95% CI P-value Bone marrow plasma cell, % o Serum M-spike FLC ratio X o Abbreviations: CI, confidence interval; FLC, free light chain. Bone marrow plasma cell % and serum M-spike were studied as continuous variables and the hazard ratios listed are per change in regressor over entire range. The corresponding hazard ratios per unit change in regressor are 1.03 and 1.21 respectively. Figure 2. TTP to symptomatic multiple myeloma from initial involved/uninvolved FLC ratio of X100 versus a ratio of o100. Median TTP was 15 months in the FLC ratio X100 group compared with 55 months in the FLC ratio o100 group (Po0.0001). At 24 months, 72% of patients with FLC ratio X100 had progressed to MM versus 28% of patients with FLC ratio o100. FLC ratio X100 group, progression to MM was 43% at 1 year (RR 1.5, ), 72% at 2 years (RR 2.6, ) and 87% at 3 years (RR 4.4, ). In comparison, progression to MM was 16% at 1 year (RR 1.5, ), 28% at 2 years (RR 2.6, ) and 40% at 3 years in the FLC o100 patients. In patients with FLC ratio X100 and involved FLC level X100 mg/dl, the risk of progression to symptomatic MM was 76% at 2 years. If progression to AL amyloidosis was included as an event in addition to symptomatic MM, the risk of progression in patients with an FLC ratio X100 was 79% at 2 years and 90% at 3 years. Corresponding rates for patients with both FLC ratio X100 and involved FLC level X100 mg/dl was 82% and 93%, respectively. Tables 4 and 5 provide comparison in a univariate model of BMPC percentage, serum M-spike, and serum FLC ratio X100. All three were significant prognostic factors on univariate analysis, and all remained significant independent predictors on multivariate analysis, validating each factor s ability to discriminate SMM patients at higher risk for progression. The 3 factors remained independently significant when BMPC percentage and serum M-spike were studied as categorical variables with cutpoints of 60% and 3gm/dl, respectively. DISCUSSION In this study, we evaluated the ability of the serum FLC ratio to reliably identify high-risk SMM that will result in early progression Table 5. Multivariate analysis of bone marrow plasma cell percentage, serum monoclonal protein level and FLC ratio X100 on time to progression Prognostic variable Hazard ratio 95% CI P-value Bone marrow plasma cell, % Serum M-spike FLC ratio X o Abbreviations: CI, confidence interval; FLC, free light chain. Bone marrow plasma cell % and serum M-spike were studied as continuous variables and the hazard ratios listed are per change in regressor over entire range. The corresponding hazard ratios per unit change in regressor are 1.02 and 1.08 respectively. to active MM necessitating treatment. Our data strongly support the conclusion that a serum involved/uninvolved FLC ratio X100 (or if k/l ratio is used, X100 or p0.01) is a highly specific independent biomarker with the ability to identify SMM patients at significantly increased risk of developing end-organ damage because of MM within 2 years. A serum involved/uninvolved FLC ratio of X100 was present in 15% of the total cohort (n ¼ 90). By ROC analysis, a cut-point of X100 was 97% specific for disease progression within 2 years. On survival analysis, the 2-year rate of progression from SMM to MM was slightly lower, but still substantial at 72% versus 28% in the FLC X100 and o100 groups, respectively. The incidence of bone disease, anemia, renal insufficiency or hypercalcemia did not significantly differ between the FLC ratio groups. The FLC ratio X100 group had an increased number of light chain only patients and involvement of BMPC compared with the lower FLC ratio group, both of which are associated with increased disease aggressiveness in MM. 19 The Leukemia (2013) & 2013 Macmillan Publishers Limited

5 predictive value increases when AL amyloidosis is included as an endpoint in addition to MM, and when an involved FLC level of X100 mg/dl is added as a requirement to the high-risk definition. Our findings provide further support to an accumulating body of evidence that elevated FLC levels are a risk factor for progression not only in MM, but in myeloma precursor disease such as SMM as well. The Mayo Clinic SMM risk stratification model proposed by Dispenzieri et al. 11 uses the criteria of serum FLC ratio o0.125 or 48, BMPCs X10% and serum M-protein X3 g/dl, and 1 point is assigned for each risk factor. TTP decreased with each additional risk factor, from 10 years with 1 risk factor, 5.1 years with 2 risk factors, to 1.9 years with 3 risk factors. This simple risk model provides useful prognostic information, however, a single reliable marker of disease progression has remained elusive. 20 In this study, we used more stringent criteria for an elevated FLC ratio in order to maximize the test specificity and positive predictive value to discriminate the highest risk SMM patients from the remainder of the group. The development of individualized treatment strategies will necessitate better understanding of underlying biological behavior, and an elevated FLC ratio X100 is shown here to be a high-quality surrogate marker for malignant transformation. To date, the International Myeloma Working Group 2010 recommendations advise against treatment of SMM outside the context of a clinical trial. 1 The recommended clinical monitoring for these patients has been serial laboratory follow-up every 3 months in the first year, then every 4 6 months thereafter, with the initiation of treatment on symptomatic progression. 7 Significant interest in developing an effective treatment regimen for SMM has led to multiple prospective treatment trials, however, improvement in overall survival has been difficult to establish. Initial studies using alkylator-based therapy concluded early treatment should be deferred in asymptomatic patients because of the lack of improvement in overall survival, increased treatment-related toxicities and concern for earlier development of resistant subclones. 21 Single-arm trials with preemptive thalidomide 3 and thalidomide plus pamidronate 4 both achieved at least a minor response in 66% and 63% of patients, respectively, but were unable to establish a clear advantage to early therapy. Numerous other trials investigating the role of bisphosphonate monotherapy, 5 anakinra (interleukin-1 receptor antagonist), 22 and curcumin 23 have not demonstrated a clear reduction in the risk of progression. The increased recognition of the biological overlap between monoclonal gammopathy of undetermined significance, SMM and the heterogeneity within groups 24 has led to the design of trials focused on high-risk SMM. In an ongoing phase III randomized controlled trial, Mateos et al. 25 are studying lenalidomide dexamethasone combination treatment as induction treatment in high-risk SMM. High risk was defined by both BMPC 410% and M-spike 43 g/dl, or if only one of these criteria were present, immunoparesis and aberrant plasma cells in the bone marrow by immunophenotyping. The interim results show an overall response rate of 81% by intention to treat analysis. Median TTP was 23 months in the delayed treatment arm but had not been reached in the lenalidomide dexamethasone group. The estimated 3-year overall survival in the treatment arm was 98% versus 82% for the delayed treatment group (P ¼ 0.05). Demonstrating the clinical benefit of early treatment strategies in SMM has been difficult previously, however, with improved identification of highrisk patients and development of less toxic therapies, renewed interest in this area is developing. As the development of biomarkers and genetic profiling advances, further refinements in the clinicopathologic categorization of SMM will likely become necessary. For example, an upper limit of BMPC involvement of o60% was recently proposed for SMM. 10 In a retrospective review of 655 patients with newly diagnosed SMM with available bone marrow samples for review, 21 patients (3.2%) had BMPC involvement X60%. Ninety-five percent of these patients progressed to active myeloma within 2 years, with a mean TTP of 7 months (95% CI, ). These patients are now considered as having MM, even in the absence of end-organ damage. The primary limitation of our study is the retrospective nature of its design. In addition, it was not statistically powered for further subgroup analysis by cytogenetics or other risk factors. An inherent advantage and disadvantage was the long period of patient eligibility spanning from 1970 to 2010, which allowed a larger study population but may have also introduced an increased number of confounders because of changes in imaging, physician practice styles and the less rigorous clinical documentation in previous decades. Despite these limitations, our findings show that patients with markedly high FLC ratio (X100) are at high risk of progression to MM or related disorder within 2 years of diagnosis and hence may be considered candidates for intervention, especially as the mode of progression in this subset is likely to be renal failure in a substantial proportion of patients. CONFLICT OF INTEREST The authors declare no conflict of interest. ACKNOWLEDGEMENTS This work was supported in part by the Jabbs Foundation (Birmingham, UK); National Cancer Institute grants CA168762, CA , CA 62242, CA , CA 83724, and the Henry J Predolin Foundation, USA. AUTHOR CONTRIBUTIONS JTL, SKK and SVR designed the research, analyzed the data, wrote and edited the manuscript. RAK, JAK and AD participated in data interpretation, reviewed the manuscript and provided critical comments. All authors reviewed and approved the final manuscript. REFERENCES 1 Kyle RA, Durie BG, Rajkumar SV, Landgren O, Blade J, Merlini G et al. Monoclonal gammopathy of undetermined significance (MGUS) and smoldering (asymptomatic) multiple myeloma: IMWG consensus perspectives risk factors for progression and guidelines for monitoring and management. Leukemia 2010; 24: Group TIMW. Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the International Myeloma Working Group. Br J Haematol 2003; 121: Rajkumar SV, Gertz MA, Lacy MQ, Dispenzieri A, Fonseca R, Geyer SM et al. Thalidomide as initial therapy for early-stage myeloma. Leukemia 2003; 17: Barlogie B, van Rhee F, Shaughnessy Jr. JD, Epstein J, Yaccoby S, Pineda-Roman M et al. Seven-year median time to progression with thalidomide for smoldering myeloma: partial response identifies subset requiring earlier salvage therapy for symptomatic disease. Blood 2008; 112: Musto P, Petrucci MT, Bringhen S, Guglielmelli T, Caravita T, Bongarzoni V et al. A multicenter, randomized clinical trial comparing zoledronic acid versus observation in patients with asymptomatic myeloma. Cancer 2008; 113: Rajkumar SV, Merlini G, San Miguel JF. Redefining myeloma. Nat Rev Clin Oncol 2012; 9: Kyle RA, Remstein ED, Therneau TM, Dispenzieri A, Kurtin PJ, Hodnefield JM et al. Clinical course and prognosis of smoldering (asymptomatic) multiple myeloma. N Engl J Med 2007; 356: Kyle RA, Therneau TM, Rajkumar SV, Offord JR, Larson DR, Plevak MF et al. A longterm study of prognosis in monoclonal gammopathy of undetermined significance. N Engl J Med 2002; 346: Perez-Persona E, Vidriales MB, Mateo G, Garcia-Sanz R, Mateos MV, de Coca AG et al. New criteria to identify risk of progression in monoclonal gammopathy of uncertain significance and smoldering multiple myeloma based on multiparameter flow cytometry analysis of bone marrow plasma cells. Blood 2007; 110: Rajkumar SV, Larson D, Kyle RA. Diagnosis of smoldering multiple myeloma. N Engl J Med 2011; 365: & 2013 Macmillan Publishers Limited Leukemia (2013)

6 Dispenzieri A, Kyle RA, Katzmann JA, Therneau TM, Larson D, Benson J et al. Immunoglobulin free light chain ratio is an independent risk factor for progression of smoldering (asymptomatic) multiple myeloma. Blood 2008; 111: Rajkumar SV, Kyle RA, Therneau TM, Clark RJ, Bradwell AR, Melton III LJ et al. Presence of monoclonal free light chains in the serum predicts risk of progression in monoclonal gammopathy of undetermined significance. Br J Haematol 2004; 127: Snozek CL, Katzmann JA, Kyle RA, Dispenzieri A, Larson DR, Therneau TM et al. Prognostic value of the serum free light chain ratio in newly diagnosed myeloma: proposed incorporation into the international staging system. Leukemia 2008; 22: Lachmann HJ, Gallimore R, Gillmore JD, Carr-Smith HD, Bradwell AR, Pepys MB et al. Outcome in systemic AL amyloidosis in relation to changes in concentration of circulating free immunoglobulin light chains following chemotherapy. Br J Haematol 2003; 122: Katzmann JA, Abraham RS, Dispenzieri A, Lust JA, Kyle RA. Diagnostic performance of quantitative kappa and lambda free light chain assays in clinical practice. Clin Chem 2005; 51: Katzmann JA, Clark RJ, Abraham RS, Bryant S, Lymp JF, Bradwell AR et al. Serum reference intervals and diagnostic ranges for free kappa and free lambda immunoglobulin light chains: relative sensitivity for detection of monoclonal light chains. Clin Chem 2002; 48: Bradwell AR, Carr-Smith HD, Mead GP, Harvey TC, Drayson MT. Serum test for assessment of patients with Bence Jones myeloma. Lancet 2003; 361: Drayson M, Tang LX, Drew R, Mead GP, Carr-Smith H, Bradwell AR. Serum free light-chain measurements for identifying and monitoring patients with nonsecretory multiple myeloma. Blood 2001; 97: van Rhee F, Bolejack V, Hollmig K, Pineda-Roman M, Anaissie E, Epstein J et al. High serum-free light chain levels and their rapid reduction in response to therapy define an aggressive multiple myeloma subtype with poor prognosis. Blood 2007; 110: Waxman AJ, Kuehl M, Balakumaran A, Weiss B, Landgren O. Smoldering (asymptomatic) multiple myeloma: revisiting the clinical dilemma and looking into the future. Clin Lymphoma Myeloma Leuk 2010; 10: Alexanian R, Barlogie B, Dixon D. Prognosis of asymptomatic multiple myeloma. Arch Intern Med 1988; 148: Lust JA, Lacy MQ, Zeldenrust SR, Dispenzieri A, Gertz MA, Witzig TE et al. Induction of a chronic disease state in patients with smoldering or indolent multiple myeloma by targeting interleukin 1{beta}-induced interleukin 6 production and the myeloma proliferative component. Mayo Clin Proc 2009; 84: Golombick T, Diamond TH, Manoharan A, Ramakrishna R. Monoclonal gammopathy of undetermined significance, smoldering multiple myeloma, and curcumin: a randomized, double-blind placebo-controlled cross-over 4 g study and an open-label 8 g extension study. Am J Hematol 2012; 87: Landgren O, Kyle RA, Pfeiffer RM, Katzmann JA, Caporaso NE, Hayes RB et al. Monoclonal gammopathy of undetermined significance (MGUS) consistently precedes multiple myeloma: a prospective study. Blood 2009; 113: Mateos M, Lopez-Corral L, Hernandez M, Giraldo P, De la Rubia J, De Arriba F et al. Smoldering multiple myeloma (SMM) at high-risk of progression to symptomatic disease: a phase III, randomized, multicenter trial based on lenalidomide-dexamethasone (Len-Dex) as induction therapy followed by maintenance therapy with Len alone vs no treatment. Blood (ASH Annual Meeting Abstracts) 2011; 118: Leukemia (2013) & 2013 Macmillan Publishers Limited