Prediction of response and progression in multiple myeloma with serum free light chains assay: corroboration of the serum free light chain response definitions

Original Study Prediction of Response and Progression in Multiple Myeloma With Serum Free Light Chains Assay: Corroboration of the Serum Free Light Chain Response Definitions Rami Khoriaty,1 Mohamad A. Hussein,2 Beth Faiman,3 Megan Kelly,3 Matt Kalaycio,3 Rachid Baz4 Abstract Background: The International Myeloma Working Group (IMWG) proposed response and progression criteria using serum free light chain (sFLC) testing for patients with nonsecretory multiple myeloma (MM). We attempt to validate these criteria by comparing paraprotein responses with sFLC responses in patients with secretory myeloma. Patients and Methods: Prospectively entered data for 89 patients with MM enrolled on various clinical trials at the Cleveland Clinic between April 2004 and December 2006 were reviewed. Results: By standard paraprotein criteria, 4 patients had complete remission (CR), 22 had partial remission (PR), 34 had stable disease (SD), 26 had progressive disease (PD), and 3 were inevaluable. Only 43 patients (48%) had an involved sFLC ≥ 10 mg/dL (which is considered evaluable by the IMWG), of which 14 had PR, 8 had SD, 18 had PD, and 3 were inevaluable. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for sFLC in predicting response were 81%, 83%, 64%, and 92% respectively. The sensitivity, specificity, PPV, and NPV for sFLC in predicting progression were 93%, 80%, 72%, and 95% respectively. Conclusion: sFLC reliably predicts response and progression in MM. However, half of the patients had inevaluable disease by sFLC, thus limiting the utility of sFLC testing in patients with nonmeasurable disease by electrophoretic methods. Clinical Lymphoma, Myeloma & Leukemia, Vol. 10, No. 1, E10-E13, 2010; DOI: 10.3816/CLML.2010.n.010 Keywords: Abnormal sFLC ratio, European Blood and Bone Marrow Transplantation, IMWG criteria Introduction Multiple myeloma (MM) is a malignant plasma cell disorder characterized by production of monoclonal immunoglobulin resulting in end-organ damage. The most frequently used response criteria in MM are the European Blood and Bone Marrow Transplantation (EBMT) criteria.1 Patients not meeting the paraprotein measurements criteria set by the EBMT (ie, M-protein > 1 g/dL or urinary light-chain > 200 mg/24 hours) are currently 1University of Michigan Hematology and Oncology Fellowship Program, Ann Arbor, MI 2Celgene Corporation, Summit, NJ 3Cleveland Clinic Taussig Cancer Institute, Cleveland, OH 4H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL Submitted: May 27, 2009; Revised: Aug 27, 2009; Accepted: Oct 9, 2009 Address for correspondence: Rachid Baz, MD, Department of Hematologic Malignancy, H. Lee Moffitt Cancer, Center and Research Institute, 12902 Magnolia Dr. SRB4, Tampa, FL, 33612 Fax: 813-745-3071; e-mail: rachid.baz@moffitt.org classified as having oligosecretory or nonsecretory MM. Disease monitoring for patients is difficult and precludes enrollments on clinical trials where response assessment is the primary objective. Recently, a nephelometric assay of the serum free light chains (sFLC) has been proposed to help in the assessment of patients with nonsecretory MM. The sFLC assay was able to identify a light chain imbalance in 19 of 28 patients with nonsecretory myeloma,2 and 98% of patients with primary amyloidosis.3 In addition, an abnormal sFLC ratio has been linked to a higher risk of progression of monoclonal gammopathy of undetermined significance to MM4 and of smoldering myeloma to active MM.5 Based on the aforementioned observations, sFLC criteria were incorporated into the response definition proposed by the International Myeloma Working Group (IMWG). These criteria are intended to monitor patients with oligosecretory or nonsecretory myeloma, but have not been thoroughly validated. To our knowledge, only 2 studies have evaluated the concordance between sFLC responses and EBMT responses in patients with secretory This summary may include the discussion of investigational and/or unlabeled uses of drugs and/or devices that may not be approved by the FDA. Electronic forwarding or copying is a violation of US and International Copyright Laws. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by CIG Media Group, LP, ISSN #2152-2650, provided the appropriate fee is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA. www.copyright.com 978-750-8400. E10 | Clinical Lymphoma, Myeloma & Leukemia February 2010 disease and have many limitations.6,7 The first study had a small sample size, and it was difficult to assess the specificity of sFLC in the prediction of response (only 1 patient did not respond to therapy). In addition, both studies did not review the usefulness of sFLC in the prediction of myeloma progression.6,7 Because determining responses in patients with nonsecretory myeloma can be very difficult, we thought to corroborate the responses by sFLC and EBMT criteria in patients with secretory disease, with the assumption that secretory myeloma has similar response kinetics to nonsecretory myeloma, and that our findings could be extrapolated for patients with nonsecretory myeloma. Patients and Methods Patient and Baseline Characteristics Patients with MM treated at the Cleveland Clinic Taussig Cancer Institute who had sFLC assays serially performed at 4-week intervals between April 2004 and December 2006 were selected. Most of these patients were enrolled on various clinical trials. This study included patients with relapsed or newly diagnosed disease. Baseline characteristics of the patients consisting of age, gender, creatinine, β2 microglobulin, albumin, white blood cell count, hemoglobin, platelet count, and type of the involved heavy and light chains were gathered. The sFLC were measured using a commercial kit assay (The Binding Site Ltd., Birmingham, UK). In addition, sFLC (performed at 4 week intervals) as well as serum and urine M-protein levels between April 2004 and December 2006 were documented. This study was approved by the institutional review board. Response Definitions The response to treatment of MM was assessed independently by the EBMT criteria1 and the sFLCs criteria set by the IMWG.8 Based on the sFLC criteria, patients were considered to have evaluable or nonevaluable disease if the involved sFLC was ≥ 100 mg/L or < 100 mg/L, respectively. The treatment response according to the sFLC criteria can be determined only in patients with evaluable disease as follows: partial response (PR), ≥ 50% decrease in the difference between the involved and uninvolved sFLCs; stable disease (SD), < 50% decrease and < 25% increase in the difference between the involved and uninvolved sFLCs; and progressive disease (PD), > 25% increase in the difference between the involved and uninvolved sFLCs.8 Myeloma Therapies and Survival Assessments The chemotherapeutic agents received by patients during the response period were classified into 4 categories: anthracycline/ immunomodulator-based therapy, bortezomib-based therapy, alkylator-based therapy, and others. Follow-up and survival information were obtained from the electronic database where the data were entered prospectively. Overall survival (OS) from the time of response by the sFLC criteria to the time of death or last follow-up was compared between patients with evaluable and inevaluable sFLC. Renal Adjustments Because the excretion of sFLC depends on renal function, patients’ creatinine clearances (CrCl) at the time of response by the sFLC criteria were calculated using the Cockroft-Gault formula. The mean CrCl of patients with evaluable and inevaluable disease by the sFLC criteria were compared. Mean CrCl of patients in all categories of response by the sFLC criteria were also calculated. Comparison of Responses by European Blood and Bone Marrow Transplantation and Serum Free Light Chain Criteria and Data Collection The dates of response to treatment according to EBMT and sFLCs criteria were determined and compared. Statistical Considerations Data were analyzed using the statistical software JMP 7.0 produced by SAS institute. The student t test was used to compare CrCl in patients with evaluable and inevaluable disease by the sFLC criteria. Overall survival curves were plotted by the KaplanMeier method. A P value of < .05 was considered to be statistically significant. Results Patient and Baseline Characteristics Eighty-nine patients were included in the study. The median age of patients was 61 years. Fifty-eight patients (65%) were males. Sixty-one (69%) had immunoglobulin (Ig) G heavy chains, 23 (26%) had IgA heavy chains, and 5 (6%) had light chain multiple myeloma. The involved light chain was κ in 64 patients (72%). The median serum β2 microglobulin, creatinine, and albumin levels at baseline were 3.6 mg/L (range, 1.4-47.7 mg/L), 1.1 mg/dL (range, 0.6-7.7 mg/dL), and 3.7 g/L (range, 2.4-4.6 g/L), respectively. Table 1 lists the demographic, laboratory, and treatment characteristics of patients. Disease Progression and Response by European Blood and Bone Marrow Transplantation and Serum Free Light Chain Criteria The distribution of responses according to EBMT and sFLC criteria is shown in Table 2. We note that only 43 patients (48%) had an involved sFLC ≥ 100 mg/L, and thus were evaluated for response by the sFLC criteria. Of these patients, 7 were newly diagnosed, and 36 had relapsed disease. According to EBMT criteria, 4 patients (5%) had CR, 22 patients (25%) had PR, 34 patients (38%) had SD, 26 patients (29%) had PD, and 3 patients (3%) were inevaluable (Table 2). On the other hand, only 43 patients (48%) had an involved sFLC ≥ 100 mg/L. Of these patients with evaluable sFLC, 14 (32%) had PR, 8 (19%) had SD, 18 (42%) had PD, and 3 (7%) were inevaluable (Table 2). The ability of sFLC to predict response and progression of MM is shown in Table 3. Regarding prediction of response, the sensitivity, specificity, positive predictive value, and negative predictive value of sFLC testing were 81%, 83%, 64%, and 92%, respectively. The sensitivity, specificity, positive predictive value, and negative predictive value of sFLC in detecting progression of MM compared with the IMWG criteria were 93%, 80%, 72%, and 95%, respectively. We then divided the patients into 3 groups. Groups 1, 2, and 3 consisted of patients with measurable baseline serum M-protein only (n = 5), measurable baseline urine light chains only (n = 10), and both measurable serum M-protein and urine light chains at Clinical Lymphoma, Myeloma & Leukemia February 2010 | E11 sFLC Predicts Response and Progression of MM Table 1 Patient Characteristics (N = 89) Table 3 Characteristics of sFLC Testing for the Prediction of Response/Progression Characteristic Value 61 (41-87) Median Age, Years (Range) Predictive Measure Sensitivity, % (CI) Specificity, % (CI) PPV, % (CI) NPV, % (CI) Response 81 (51-94) 83 (65-92) 64 (38-83) 92 (68-98) Progression 93 (68-98) 80 (62-91) 72 (49-87) 95 (78-99) 58 (65) Men, n (%) Myeloma Therapy Received, n (%) Anthracycline/immunomodulatora 59 (66) Bortezomib basedb 4 (5) Alkylator based 9 (10) Otherc 17 (19) Heavy-Chain, n (%) IgG 61 (69) IgA 23 (26) Abbreviations: CI = confidence interval; NPV = negative predictive value; PPV = positive predictive value; sFLC = serum free light chain Figure 1 Overall Survival of 89 Patients With Evaluable or Inevaluable sFLC 100 5 (6) None, Light-chain MM sFLC < 100 mg/L sFLC ≥ 100 mg/L P = .0325 80 64 (72) κ Laboratory Measure Baseline β2 microglobulin (mg/L): median (range) 3.6 (1.4-47.7) Baseline serum creatinine (mg/dL): median (range) 1.1 (0.6-7.7) Baseline albumin (g/L): median (range) 3.7 (2.4-4.6) Baseline WBC (× 103/µL): median (range) 4.15 (0.81-17.88) 11.9 (7.1-15.8) Baseline hemoglobin (g/dL): median (range) Baseline platelet count (× 103/µL): median (range) aAnthracycline/immunomodulator: 184 (5-387) pegylated liposomal doxorubicin/vincristine/dexamethasone/ thalidomide or lenalidomide. bBortezomib based: bortezomib/vorinostat. cOther: arsenic trioxide + thalidomide + dexamethasone + ascorbic acid, organic arsenic, SGN-40. Abbreviations: MM = multiple myeloma; WBC = white blood cell count Table 2 Response by European Blood and Bone Marrow Transplantation and sFLC Criteriaa Response Measure Patients With Involved sFLC ≥ 100 mg/L, n = 43 (48%) All Patients, N = 89 (100%) sFLC response, n (%) PR 14 (32) SD 8 (19) PD 18 (42) Inevaluable NA 3 (7) EBMT response, n (%) CR 0 (0) 4 (5) PR 11 (26) 22 (25) SD 18 (42) 34 (38) PD 14 (32) 26 (29) 0 (0) 3 (3) Inevaluable asFLC response can only be evaluated in patients with involved sFLC ≥ 100 mg/L. Abbreviations: CR = complete remission; NA = not applicable; PD = progressive disease; PR = partial remission; SD = stable disease; sFLC = serum free light chain baseline (n = 28), respectively. The sensitivity and specificity of sFLC to detect response or progression of disease could not be calculated in group 1 because all patients in group 1 had SD by E12 | Clinical Lymphoma, Myeloma & Leukemia February 2010 Survival, % Light-Chain, n (%) 60 40 20 0 0 10 20 30 40 Overall Survival, Months Kaplan-Meier curve showing that patients with evaluable sFLC (involved sFLC ≥ 100 mg/L) have a statistically significant worse survival than patients with inevaluable sFLC (log-rank; P = .0325). Abbreviation: sFLC = serum free light chain IMWG. The sensitivity and specificity of sFLC to detect response of MM were 100% (95% confidence interval (CI), 5%-100%) and 78% (95% CI, 40%-96%), respectively, in group 2, and 80% (95% CI, 44%-96%) and 94% (95% CI, 71%-100%), respectively, in group 3. On the other hand, the sensitivity and specificity of sFLC to detect MM progression were 100% (95% CI, 40%-100%) and 83% (95% CI, 36%-99%), respectively, in group 2, and 90% (95% CI, 54%-99%), and 83% (95% CI, 58%-96%) in group 3. Our practice has been to measure sFLC and M-protein in the serum and urine approximately every 4 weeks. In our study, the majority (27 patients) had responses detected by sFLC and M-protein criteria at the same time, 8 patients had responses detected earlier by the sFLC criteria, and 5 patients had responses detected earlier by the M-protein criteria. The mean time to a sFLC response was 17 days less than an EBMT response. Creatinine Clearance and Serum Free Light-Chain Analysis Two patients were on dialysis, both of whom had evaluable sFLC levels. Their CrCl was assumed to be zero. The mean CrCl of the patients at the time of response by sFLC criteria was 89 mL/min (median, 90 mL/min; range, 0-179 mL/min). Mean CrCl of patients with and without evaluable sFLC were 87 and 90 mL/min, respectively (P = .68). Of the patients with evaluable sFLC, the mean CrCl of patients who had a response, SD, PD, and were inevaluable were 87, 87, 87, and 89 mL/min, respectively. Rami Khoriaty et al Effect of Serum Free Light Chain on Survival As of October 2008, 50 patients (56%) died. The OS curves plotted by the Kaplan-Meier method showed that patients with evaluable sFLC had a statistically significant worse survival than patients with inevaluable sFLC (log rank; P = .0325; Figure 1). Discussion To the best of our knowledge, this is only the second study to evaluate the usefulness of sFLC in detecting response and the only study to evaluate the test’s usefulness in evaluating progressive disease. This study shows that sFLC can reliably predict response and progression of MM with particularly elevated negative predictive values of 92% and 95%, respectively. However, about half of the patients had inevaluable disease by the sFLC criteria. For those patients, sFLC cannot be used to monitor the course of MM and the effect of therapy. The benefits of monitoring MM with sFLC rather than intact immunoglobulins are multifold. First, the shorter half-life of sFLC (in the order of hours) compared with that of intact immunglobulins (2-25 days) makes the assessment of MM with sFLC after 1 or 2 weeks of chemotherapy feasible, which can guide early decisions in the course of treatment. In our study, because sFLC and M-proteins were measured at the same 4-week intervals, we could not assess with high accuracy how much earlier the responses are detected by sFLC criteria. The majority of the patients in our study (67.5%) had detectable responses by sFLC and M-protein criteria at the same time. However, overall the mean time to a sFLC response was 17 days earlier than an EBMT response. Second, sFLC correlates better than intact monoclonal IgG with malignant plasma cells burden.9 Third, sFLC can detect the relapses that occur shortly after response to treatment,10 which might not be detected by intact immunoglobulin concentrations as those might still be decreasing after an earlier response because of their long half-lives. In addition, relapse can happen with exclusive production of sFLC as a shift from intact immunoglobulin secretion, a process termed free light chain escape.11 Free light chain levels are affected by renal function: the worse the renal function, the less they are excreted in the urine.12 In our study, CrCl was not different between patients with evaluable and inevaluable sFLC, which suggests that renal function does not account for the differences in the sFLC between the 2 groups of patients. This analysis suggests that patients with evaluable sFLC had worse OS than patients with inevaluable sFLC (P = .0325; Figure 1). The prognostic value of sFLC for survival has been studied by different groups, and the results are not completely consistent. Some studies showed a correlation between sFLC and survival whereas others have not.9,13-15 However, the majority of the evidence is in favor of the presence of a correlation between sFLC and survival, and sFLC have been proposed to be included in the international staging system of MM.15 This study has a number of potential limitations that are inherent to retrospective reports. The patients were treated with different regimens and it is possible, although unlikely, that a therapeutic agent could interfere with the sFLC test; however, the limited number of patients treated with each agent makes this assumption difficult to disprove. On the other hand, the fact that the patients were receiving various chemotherapeutic regimens could be considered a strength of this study because it makes the conclusions more generalizable. Finally, it is difficult to draw firm conclusions regarding the effect of sFLC results on OS in this study because of the limited number of events and follow-up. Conclusion In summary, sFLC assay can be useful for monitoring patients with MM when the involved sFLC is evaluable (ie, > 100 mg/L) and can reliably predict response and progression of MM. Further studies are needed to assess the optimal frequency of measurement and to delineate the prognostic properties of sFLC. Disclosures Beth Faiman has served as a member of a Speaker’s Bureau for Celgene Corporation and Millennium Pharmaceuticals, Inc. Mohamad A. Husssein has been an employee of Celgene Corporation. The remaining authors report no relevant potential conflicts of interest. References 1. Blade J, Samson D, Reece D, et al. Criteria for evaluating disease response and progression in patients with multiple myeloma treated by high-dose therapy and haemopoietic stem cell transplantation. Myeloma Subcommittee of the EBMT. European Group for Blood and Marrow Transplant. Br J Haematol 1998; 102:1115-23. 2. Drayson M, Tang LX, Drew R, et al. Serum free light-chain measurements for identifying and monitoring patients with nonsecretory multiple myeloma. Blood 2001; 97:2900-2. 3. Lachmann HJ, Gallimore R, Gillmore JD, 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; 122:78-84. 4. Rajkumar SV, Kyle RA, Therneau TM, et al. Serum free light chain ratio is an independent risk factor for progression in monoclonal gammopathy of undetermined significance. Blood 2005; 106:812-7. 5. Dispenzieri A, Kyle RA, Katzmann JA, et al. Immunoglobulin free light chain ratio is an independent risk factor for progression of smoldering (asymptomatic) multiple myeloma. Blood 2008; 111:785-7. 6. Kumar S, Gertz MA, Hayman SR, et al. Use of the serum free light chain assay in assessment of response to therapy in multiple myeloma: validation of recently proposed response criteria in a prospective clinical trial of lenalidomide plus dexamethasone for newly diagnosed multiple myeloma. Blood 2005; 106: (abstract 3479). 7. Dispenzieri A, Zhang L, Katzmann JA, et al. Appraisal of immunoglobulin free light chain as a marker of response. Blood 2008; 111:4908-15. 8. Durie BG, Harousseau JL, Miguel JS, et al. International uniform response criteria for multiple myeloma. Leukemia 2006; 20:1467-73. 9. Mead GP, Carr-Smith HD, Drayson MT, et al. Serum free light chains for monitoring multiple myeloma. Br J Haematol 2004; 126:348-54. 10. Kühnemund A, Liebisch P, Bauchmuller K, et al. Secondary light-chain multiple myeloma with decreasing IgA paraprotein levels correlating with renal insufficiency and progressive disease: clinical course of two patients and review of the literature. Onkologie 2005; 28:165. 11. Dawson MA, Patil S, Spencer A. Extramedullary relapse of multiple myeloma associated with a shift in secretion from intact immunoglobulin to light chains. Haematologica 2007; 92:143-4. 12. Alyanakian MA, Abbas A, Delarue R, et al. Free immunoglobulin light-chain serum levels in the follow-up of patients with monoclonal gammopathies: correlation with 24-hr urinary light-chain excretion. Am J Hematol 2004; 75:2468. 13. Kyrtsonis MC, Vassilakopoulos TP, Kafasi N, et al. Prognostic value of serum free light chain ratio at diagnosis in multiple myeloma. Br J Haematol 2007; 137:2403. 14. van Rhee F, Bolejack V, Hollmig K, 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; 110:827-32. 15. Snozek CL, Katzmann JA, Kyle RA, et al. Prognostic value of the serum free light chain ratio in newly diagnosed myeloma: proposed incorporation into the international staging system. Leukemia 2008; 22:1933-7. Clinical Lymphoma, Myeloma & Leukemia February 2010 | E13