Importance of Achieving a Complete Response in Multiple Myeloma, and the Impact of Novel Agents Asher A. Chanan-Khan and Sergio Giralt

Transcription

1 VOLUME 28 NUMBER 15 MAY JOURNAL OF CLINICAL ONCOLOGY R E V I E W A R T I C L E Importance of Achieving a Complete Response in Multiple Myeloma, and the Impact of Novel Agents Asher A. Chanan-Khan and Sergio Giralt From the Roswell Park Cancer Institute, Buffalo, NY; and the Department of Stem Cell Transplantation, University of Texas M.D. Anderson Cancer Center, Houston, TX. Submitted August 4, 2009; accepted November 16, 2009; published online ahead of print at on April 12, Supported by Millennium Pharmaceuticals. Authors disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Asher A. Chanan-Khan, MD, Department of Medicine, Roswell Park Cancer Institute, Elm & Carlton St, Buffalo, NY, 14263; asher.chanan-khan@ roswellpark.org by American Society of Clinical Oncology X/10/ /$20.00 DOI: /JCO A B S T R A C T The goal of treatment for multiple myeloma (MM) is to improve patients long-term outcomes. One important factor that has been associated with prolonged progression-free and overall survival is the quality of response to treatment, particularly achievement of a complete response (CR). There is extensive evidence from clinical studies in the transplant setting in first-line MM demonstrating that CR or maximal response post-transplant is significantly associated with prolonged progression-free and overall survival, with some studies demonstrating a similar association with postinduction response. Supportive evidence is also available from studies in the nontransplant and relapsed settings. With the introduction of bortezomib, thalidomide, and lenalidomide, higher rates of CR are being achieved in both first-line and relapsed MM compared with previous chemotherapeutic approaches, thereby potentially improving long-term outcomes. While standard CR by established response criteria has been shown to have differential prognostic impact compared with lesser responses, increasingly sensitive analytic techniques are now being explored to define more stringent degrees of CR or elimination of minimal residual disease (MRD), including multiparameter flow cytometry and polymerase chain reaction. Demonstrating eradication of MRD by these techniques has already been shown to predict for improved outcomes. Here, we review the prognostic significance of achieving CR in MM and highlight the importance of CR as an increasingly realizable goal at all stages of treatment. We discuss clinical management issues and provide recommendations relevant to practicing oncologists, such as the routine use of sensitive techniques for assessment of disease status to inform evidence-based decisions on optimal patient management. J Clin Oncol 28: by American Society of Clinical Oncology INTRODUCTION Cure represents the ultimate goal of therapy for oncologists treating patients with multiple myeloma (MM). However, in the large majority of patients cure remains elusive at present with widely used therapeutic options. 1 Although complex, intensive tandem transplant protocols such as Total Therapy 1 to 3 2 and allogeneic stem-cell transplant 3 have been suggested as potentially curative approaches, these are experimental, difficult to incorporate in general oncology practices, and associated with significant patient morbidity and mortality. 3-5 MM typically follows a relapsing course, with the clinical benefit of treatment decreasing with each subsequent line of therapy. 6 In the absence of definitive cure, the goal of treatment is to improve patients long-term outcomes, including prolonging progression-free survival (PFS) and overall survival (OS). Clinical trials have shown that numerous disease-related factors are of prognostic importance for OS, including the International Staging System (ISS), 7,8 2 - microglobulin, 7,9,10 albumin, 7,10 C-reactive protein (CRP), and lactate dehydrogenase (LDH) levels, 11,12 cytogenetic abnormalities, 10,11,13,14 plasma cell labeling index (PCLI), 9,12,13,15,16 and renal impairment. 7,16-18 Clearly, given this range of factors and the highly heterogeneous nature of the MM patient population, determining the prognosis for longterm OS is a highly complex issue for oncologists. One important factor widely associated with improved PFS and OS in MM is a patient s quality of response to treatment, and in particular the achievement of a durable complete response (CR), reflecting findings in other malignancies including chronic myeloid leukemia (CML). 19,20 CR represents elimination of detectable disease by currently available laboratory methods (Table 1) Historically, the chance of achieving CR in newly diagnosed MM patients was low, as CR was not common with conventional chemotherapy such as melphalan and prednisone (MP) or vincristine, doxorubicin, and dexamethasone (VAD) Although these regimens demonstrated antimyeloma activity in more than half of patients, treatment was generally not sufficiently intensive to eliminate all traces of by American Society of Clinical Oncology

2 Importance of CR in MM and the Impact of Novel Agents Table 1. Diagnostic and Analytical Techniques for the Detection of Myeloma Disease Burden Technique Disease Burden Assessment Useful Reading Serum/urine protein electrophoresis Detection and quantitation of monoclonal proteins/light chains in the serum/urine (sensitivity 1-2 g/l) Serum/urine immunofixation Detection of monoclonal proteins in the serum/urine (sensitivity mg/l) 21-23,25 Serum free light chain assay Quantitation of free kappa and lambda immunoglobulin light chains in the serum; ratio 22,23,26 of kappa/lambda light chains (sensitivity 1 mg/l) Bone marrow aspirate/biopsy Cytomorphologic assessment of percentage of myeloma plasma cells in the bone 21,22 marrow (sensitivity 5%) Immunohistochemistry/immunofluororescence Quantitation of myeloma plasma cells in bone marrow tissue/cell samples via antibody-antigen interaction (sensitivity ) 22,27 Immunophenotyping (multiparametric flow cytometry) Real-time quantitative polymerase chain reaction Automated cell-by-cell quantitation of myeloma plasma cells in bone marrow samples via multiple antibody-antigen interactions (sensitivity 10 4 ) Identification of residual tumor cells based on presence of patient-specific selected immunoglobulin heavy-chain genomic rearrangements (sensitivity 10 6 ) Magnetic resonance imaging Identification of focal lesions in the bone marrow (sensitivity 0.5 cm lesion size) 34 25, disease. This changed dramatically with the introduction of high-dose melphalan therapy (HDT) plus autologous stem-cell transplant (ASCT); substantially higher CR rates (up to 50%) were achieved in patients undergoing single or double transplant, and this was associated with improved outcomes, including event-free survival (EFS), PFS, and OS, in patients receiving HDT-ASCT versus nonintensive conventional chemotherapy in a number of clinical trials. 45,49,52,54 More recently, regimens incorporating the novel agents bortezomib, thalidomide, and lenalidomide have demonstrated very high CR rates in both first-line and relapsed MM compared with previous conventional chemotherapeutic approaches 56,57 ; indeed, the introduction of these agents has already been associated with improved OS from time of diagnosis 58,59 as well as delaying the time to relapse. 56,57 CR therefore now represents an achievable and important goal of MM therapy, with the available evidence demonstrating its prognostic significance for prolonged OS. However, it is important to note that responses reported as CR in the literature are not necessarily uniform and may therefore represent different levels of disease control based on criteria used in different reports, while classifications of maximal response may also vary. The challenges from a clinical perspective are, therefore, to classify CR/maximal response appropriately, both in terms of the response itself and relapse, and to achieve this therapeutic goal and maintain the response in patients. An important question regarding the practical management of patients with MM is whether all CRs are equal in terms of depth of response and prognostic importance (ie, whether CR after HDT-ASCT is necessarily the equivalent of CR achieved with nonintensive chemotherapy or with novel-agentbased regimens). This review addresses these key issues regarding definition and achievement of CR in the management of patients with MM, and provides guidance to practicing oncologists regarding treatment approaches and goals, and assessment and monitoring of responding patients. We also review the wealth of evidence regarding the prognostic importance of achieving CR, the impact of novel therapies on CR rates, and the differential prognostic impact of depth of remission as defined by different, highly sensitive assessment techniques. DEFINITIONS OF CR/MAXIMAL RESPONSE A number of definitions of CR have been employed in clinical trials. With the introduction of HDT-ASCT and novel agents, and with the development of increasingly sensitive methods of residual disease assessment, response criteria have evolved to include more stringent definitions of CR and other responses. Table summarizes some common definitions for CR and maximal response, including those from the European Group for Blood and Marrow Transplant (EBMT) criteria 21 and the more recent International Myeloma Working Group (IMWG) uniform response criteria. 22 While the latter are substantially based on the former, there are a number of noteworthy differences regarding definition and confirmation of CR, for which validation in large ongoing studies will be important. First, the IMWG criteria include the new category of stringent CR (scr), reflecting the introduction of the free light chain assay and the use of sensitive immunohistochemistry or immunofluorescence techniques for defining a greater depth of remission than standard CR. 22 Second, these criteria feature the category of very good partial response (VGPR), which incorporates near CR (ncr) 63,64 from the modified EBMT criteria; whether ncr and other degrees of VGPR differ remains to be determined. Furthermore, responses require only one consecutive assessment for confirmation, with no requirement for a minimum 6-week interval as in the EBMT criteria. Finally, an important difference is in the different definitions of relapse from CR. In the EBMT criteria, relapse from CR requires only reappearance of M protein on immunofixation, whereas progression from other responses requires minimum threshold increases as evidence of disease recurrence. 21 This results in what has been termed the CR penalty, whereby response durations may be shorter among patients achieving CR than those with only a partial response (PR). 65 In clinical trials of regimens resulting in high rates of CR that employ the EBMT criteria, this may adversely affect time to progression (TTP) and PFS. 65,66 In the IMWG criteria, the CR penalty has been eliminated through uniform definition of progression from all types of response. 22 This may result in data from studies employing the IMWG criteria more closely reflecting the true clinical benefit of achieving CR in terms of TTP/PFS by American Society of Clinical Oncology 2613

3 Chanan-Khan and Giralt Table 2. Differences in Definitions of Highest Quality Responses and Relapse/Progressive Disease in the Most Commonly Used MM Response Criteria Criteria SWOG 60 : Remission IFM 45,61,62 :CR ncr CR VGPR CR scr Response Serum M-protein reduction Urine M-protein reduction 90 ( 0.2 g/dl) % EBMT 21,63 IMWG ( 100 mg/24 hours) Immunofixation NR NR Positive Negative Positive Negative Negative Confirmation 3 weeks 4 weeks 6 weeks Two consecutive assessments, any time Plasma cells in bone marrow NS 5% 5% 5% NS 5% IHC/IFLnegative Free light chain ratio NA NA NA NA NA NA Normal Lytic bone lesions Stable/improved NS Stable/improved Stable/improved Plasmacytomas NS NS None None 50% size reduction None None Calcium Normal NS Normal Normal Normal Normal Normal Relapse/progression Serum/urine M- protein Other Increase in tumor mass 100% over lowest level, or to greater than levels defining response Increase in lytic bone lesions; other evidence of new disease activity Reappearance of M- protein confirmed after 4 weeks Recurrence of bone marrow infiltration Increase of 25% from Reappearance on baseline, absolute immunofixation increases must be 0.5 g/dl (serum) or 200 mg/24 hours (urine) Increase of 25% from baseline in bone marrow plasma cell %, absolute % 10%; development of new/ increase in size of existing bone lesions/ plasmacytomas; development of hypercalcemia due to myeloma 5% plasma cells in bone marrow; development of new bone lesions/ plasmacytomas, increase in size of bone lesions; development of hypercalcemia due to myeloma Increase of 25% from baseline, absolute increases must be 0.5 g/dl (serum) or 200 mg/24 hours (urine); in patients without measurable disease, increase of 10 mg/dl in the difference between involved and uninvolved FLC levels Increase of 25% from baseline in bone marrow plasma cell %, absolute % 5% (CR) or 10% (VGPR); development of new/increase in size of existing bone lesions/plasmacytomas; development of hypercalcemia due to myeloma Abbreviations: MM, multiple myeloma; SWOG, Southwest Oncology Group; IFM, Intergroupe Francophone du Myèlome; CR, complete response; EBMT, European Group for Blood and Marrow Transplantation; IMWG, International Myeloma Working Group; ncr, near complete response; VGPR, very good partial response; scr, stringent complete response; NR, not required; NS, not specified; NA, not applicable; IHC, immunohistochemistry; IFL, immunofluoresence; FLC, free light chain. Presence/absence of clonal cells based on free light chain ( / ) ratio; abnormal ratio is 4:1 or 1:2. CR IS ASSOCIATED WITH IMPROVED SURVIVAL IN MM Evidence From ASCT Studies There is extensive evidence from clinical studies incorporating HDT-ASCT in previously untreated MM patients that CR or maximal response post-asct significantly improves EFS, PFS, and OS. 1,4,45-47,52,55,61,67-77 Furthermore, some studies have demonstrated such an association between long-term outcomes and response after induction therapy, before ASCT, 47,72,78,79 possibly as a result of further improvements in depth of response after transplant, supporting the idea that achieving CR or maximal response is a key therapeutic goal at all stages of treatment. These data are supported by the findings of a meta-analysis of 21 studies that reported long-term outcomes by response or presented P values for correlations between response and long-term outcomes. 80 In total, data from 4,990 patients included in 10 prospective and 11 retrospective studies were analyzed using two meta-analyses based on P values reported for associations between response and outcomes, and on primary response and outcome data from individual studies. 80 Highly significant associations between maximal response and outcomes were demonstrated. Using reported P values, maximal response post-transplant was associated with prolonged OS (P.00001) and prolonged TTP/PFS/EFS (P.00001), and depth of postinduction response was also associated with OS (P.0015) and TTP/PFS/EFS (P.00001). Using primary data, similar highly significant associations were seen for respective studies. 80 Further delineating the prognostic impact of quality of response, one study has demonstrated a differential prognostic impact of CR and ncr by EBMT criteria. 76 In a series of uniformly treated patients, those achieving CR versus ncr post-asct had significantly longer EFS (P.0001) and OS (P.01); a trend toward similar differential prognosis was seen with CR versus ncr postinduction/pretransplant. 76 These results effectively highlight the importance of the greater depth of response with CR compared with ncr, and thus, by implication, CR versus VGPR. Another study has similarly identified a differential prognostic impact between CR by EBMT/IMWG criteria and incomplete CR, in which patients had 5% plasma cells in their bone marrow. 81 Patients achieving true CR had significantly longer OS from diagnosis (median 10.7 v 5.6 years, P.001) than those achieving incomplete CR, 81 indicating the importance of bone by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

4 Importance of CR in MM and the Impact of Novel Agents Table 3. Response Rates to Novel-Agent-Containing Induction Therapy, and Clinical Outcomes After HDT-ASCT in Phase III and Large Phase II Trials Study by Induction Regimen No. Postinduction (%) Post-Transplant (%) ORR CR/nCR/VGPR ORR CR/nCR/VGPR Long-Term Outcomes (%) Outcomes by Best Response IFM Vel/Dex CR CR 24-month PFS: 69 NR 15 CR/nCR 37 CR/nCR 24-month OS: VGPR 57 VGPR VAD CR 79 9 CR 24-month PFS: 60 7 CR/nCR 19 CR/nCR 24-month OS: VGPR 38 VGPR GIMEMA MMY VTD CR/nCR NR 43 CR 24-month PFS: 90 NR 62 VGPR 55 CR/nCR 24-month OS: VGPR TD CR/nCR NR 23 CR 24-month PFS: VGPR 32 CR/nCR 24-month OS: VGPR HOVON-65/GMMG-HD4 44 PAD CR/nCR CR/nCR NR NR 45 VGPR 71 VGPR VAD CR/nCR CR/nCR NR 17 VGPR 44 VGPR PETHEMA/GEM GEM05MENOS65 83 TD CR CR NR NR 12 CR/nCR 53 CR/nCR VTD CR CR NR 41 CR/nCR 64 CR/nCR VBMCP/VBAD/Vel CR CR NR 28 CR/nCR 54 CR/nCR HOVON TAD 556 NR NR CR EFS: 33 months NR 65 VGPR PFS: 33 months OS: 59 months VAD NR NR CR EFS: 22 months 54 VGPR PFS: 25 months OS: 62 months Macro et al 85 TD 100 NR 25 VGPR NR 44 VGPR NR NR VAD-like regimen 104 NR 7 VGPR NR 42 VGPR NR ECOG E1A00 86 TD CR NR NR 1-year OS: 80 NR Dexamethasone CR NR NR 1-year OS: 80 GMSG 87 T-VAD-Doxil CR NR NR 2-year PFS: 59 NR 54 VGPR 2-year OS: 77 VAD-Doxil CR NR NR 2-year PFS: VGPR 2-year OS: 65 ECOG E4A03 88 RevHD CR NR NR 3-year OS: 75 NR 51 VGPR RevLD CR NR NR 3-year OS: VGPR Total therapy 2 4,89 Including thalidomide 323 NR NR NR 62 CR EFS: 6.0 years CR significantly associated 8-year OS: 57 with EFS and OS Without thalidomide 345 NR NR NR 43 CR EFS: 4.1 years 8-year OS: 44 (continued on following page) Superior OS with sustained CR v CR relapsed in 3 year v non CR by American Society of Clinical Oncology 2615

5 Chanan-Khan and Giralt Table 3. Response Rates to Novel-Agent-Containing Induction Therapy, and Clinical Outcomes After HDT-ASCT in Phase III and Large Phase II Trials (continued) Study by Induction Regimen No. Postinduction (%) Post-Transplant (%) ORR CR/nCR/VGPR ORR CR/nCR/VGPR Long-Term Outcomes (%) Outcomes by Best Response Total therapy 3 5,91,92 VTD-PACE induction 303 NR NR NR 60 CR 2-year EFS: 85 NR 80 CR/nCR 2-year OS: 85 Palumbo et al 93 PAD-MEL 100 Rev/Pred Rev CR 60 VGPR CR 87 VGPR 2-year PFS: 78 2-year OS: 84 PFS appeared prolonged in patients achieving CR v VGPR v PR Abbreviations: CR, complete response; ECOG, Eastern Cooperative Oncology Group; EFS, event-free survival; GEM, Grupo Español de Mieloma; GIMEMA, Gruppo Italiano Malattie Ematologiche dell Adulto; GMSG, Greek Myeloma Study Group; HOVON, Hemato-Oncologie voor Volwassenen Nederland; IFM, Intergroupe Francophone du Myèlome; MEL 100, melphalan 100 mg/m 2 conditioning; ncr, near CR; NR, not reported; ORR, overall response rate; OS, overall survival; PAD, bortezomib, doxorubicin, and dexamethasone; PETHEMA, Programa para el Estudio de la Terapéutica en Hemopatía Maligna; PFS, progression-free survival; Rev, lenalidomide; RevHD, lenalidomide, high-dose dexamethasone; RevLD, lenalidomide, low-dose dexamethasone; Rev/Pred, lenalidomide and prednisone; TAD, thalidomide, doxorubicin, and dexamethasone; TD, thalidomide and dexamethasone; T-VAD-Doxil, thalidomide, vincristine, liposomal doxorubicinand dexamethasone; VAD, vincristine, doxorubicin, and dexamethasone; VBMCP/VBAD, vincristine, carmustine, melphalan, cyclophosphamide, prednisone/vincristine, carmustine, doxorubicin, and dexamethasone; Vel, bortezomib; Vel/Dex, bortezomib and dexamethasone; VGPR, very good partial response; VTD, bortezomib, thalidomide, and dexamethasone. Updated data presented at Annual Meeting of the American Society of Hematology, December 6-9, 2008, San Francisco, CA. Updated data presented at Annual Meeting of the European Hematology Association, June 4-7, 2009, Berlin, Germany. marrow analysis in assessing response and its association with longterm outcomes. The high rates of CR being achieved with novel-agent-based induction regimens may prove important in improving outcomes in MM. Table summarizes data from phase III and large phase II studies of novel-agent-based regimens as induction before HDT- ASCT. Preliminary findings from some studies suggest that PFS and OS advantages may be emerging, associated with higher rates of CR/ maximal response; however, it is important to note that prolonged follow-up is required in many studies to confirm the long-term impact of the improved responses achieved with these novel-agentbased regimens. The importance of CR on long-term outcomes has already been demonstrated by the Arkansas group in the Total Therapy 2 study, in which CR was significantly associated with prolonged EFS and OS. 4 Data from Total Therapy 2, incorporating thalidomide, 90 and from Total Therapy 3, 10 incorporating bortezomib and thalidomide, are also demonstrating the prognostic significance of durability of CR, rather than just achievement of CR per se, on long-term outcomes. Continued CR clearly represents an essential condition for potential cure, 1,2 as in CML, 19,20 and has been associated with significantly prolonged OS. In Total Therapy 2, patients who achieved sustained CR for 3 years after the start of treatment had significantly superior survival compared with those who either never achieved CR or achieved CR but relapsed within 3 years of the start of therapy. 90 Meanwhile, significantly improved CR durability has been reported in Total Therapy 3 versus Total Therapy 2 (with or without thalidomide), 2,10 together with improved EFS and indications of improved OS, associated with the addition of bortezomib. 10 Time to CR and CR rates were similar between studies; however, the 2-year sustained CR rate was 91% in Total Therapy 3 versus 81% (P.005) in Total Therapy 2 plus thalidomide, which translated into improved EFS (2-year EFS: 84% v 77%, P.008) and a trend toward improved OS (2-year OS: 87% v 83%, P.12). 10 Thus, maintaining CR status appears an important goal for oncologists in first-line MM treatment. This appears particularly true for patients with high-risk cytogenetic abnormalities, as demonstrated in Total Therapy 2. 89,90 Evidence From Nontransplant Studies Achievement of CR/maximal response is also associated with prolonged long-term outcomes in patients not undergoing high-dose therapy, albeit in a limited number of studies. 1,12,94-98 This smaller evidence base in this setting versus the transplant population is likely a result of the fact that rates of CR or CR/VGPR were until recently typically low with first-line standards of care in these patients, preventing meaningful analysis of differential outcomes according to response. An important study regarding the impact of achieving CR in this setting was the ECOG E9486 trial of vincristine, carmustine, melphalan, cyclophosphamide, and prednisone (VBMCP) with or without interferon alfa-2 or high-dose cyclophosphamide. 12 After median follow-up of 13.1 years, median OS in patients achieving CR versus PR was 5.1 years versus 3.3 years from time of first response (P.0001); a 2-year landmark analysis, conducted to overcome the inherent bias arising from the additional time required to achieve CR compared with PR, also demonstrated significantly longer OS after 2 years in patients achieving CR versus PR (median 3.6 v 2.4 years; P.006). 12 Interestingly, however, multivariate analysis identified PCLI, percentage of plasma cells by immunofluorescence, sex, plasmablastic morphology, and CRP, but not CR, as independent predictors of postresponse survival, 12 illustrating the complexity of OS prognosis. Nevertheless, a recent analysis of data from this and other phase III ECOG trials in the nontransplant setting identified response duration after initial therapy as a major predictor of OS. 99 A retrospective analysis also demonstrated significantly longer OS in nontransplant patients achieving CR versus PR versus no response to initial dexamethasone-based therapy. 1 The addition of bortezomib, thalidomide, and lenalidomide to first-line therapies for nontransplant MM patients has resulted in high rates of CR and VGPR, associated with improved outcomes in some phase III studies. Table summarizes data from key studies of novel-agent-based regimens in this setting. A recent analysis of the phase III Velcade as Initial Standard Therapy in Multiple Myeloma; Assessment with Melphalan and Prednisone (VISTA) trial comparing bortezomib plus MP (VMP) with MP demonstrated that achievement of CR versus PR with VMP by EBMT criteria was associated with by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

6 Importance of CR in MM and the Impact of Novel Agents Table 4. Clinical Outcomes After Novel-Agent-Containing First-Line Therapy in Patients With MM Not Undergoing HDT-ASCT: Phase III Studies and Phase II Trials Reporting Outcomes by Response Study by Regimen No. ORR (%) CR/nCR/VGPR Rates Long-Term Outcomes Outcomes by Best Response VISTA 100 VMP CR DOR: 19.9 months, DOR (CR): 24.0 months, TTP: 24.0 months, TNT: 28.1 months, TFI: 16.6 months, 3-year OS: 72% MP CR DOR: 13.1 months, DOR (CR): 12.8 months, TTP: 16.6 months, TNT: 19.2 months, TFI: 8.4 months, 3-year OS: 59% PETHEMA/GEM GEM05MAS VMP CR, 36 CR/nCR 2-year TTP: 81%, 2-year OS: 92% NR VTP CR, 37 CR/nCR 2-year TTP: 83%, 2-year OS: 94% GIMEMA phase III 102 VMPT CR, 51 VGPR 3-year PFS: 71%, 3-year TNT: 80%, 3-year OS: 90% VMP CR, 42 VGPR 3-year PFS: 56%, 3-year TNT: 78%, 3-year OS: 89% IFM MPT CR, 47 VGPR PFS: 27.5 months, OS: 51.6 months NR MP CR, 7 VGPR PFS: 17.8 months, OS: 33.2 months VAD MEL CR, 43 VGPR PFS: 19.4 months, OS: 38.3 months GISMM2001-A 98 MPT CR, 29 VGPR TTP: 24.7 months, PFS: 21.8 months, OS: 45.0 months MP CR, 11 VGPR TTP: 15.0 months, PFS: 14.5 months, OS: 47.6 months IFM01/ MPT CR, 21 VGPR PFS: 24.1 months, OS: 44.0 months NR MP CR, 7 VGPR PFS: 18.5 months, OS: 29.1 months HOVON MPT VGPR 2-year EFS: 36%, 2-year PFS: 33%, NR 2-year/4-year OS: 67%/36% MP VGPR 2-year EFS: 12%, 2-year PFS: 19%, 2-year/4-year OS: 60%/25% NMSG 106 MPT CR, 23 VGPR TTP: 22 months, PFS: NR months, OS: 35 months MP CR, 7 VGPR TTP: 18 months, PFS: months, OS: 35 months Ludwig et al 107 TD CR, 26 VGPR TTP: 21.2 months, PFS: 16.7 NR months, OS: 41.5 months MP CR, 13 VGPR TTP: 29.1 months, PFS: 20.7 months, OS: 49.4 months ECOG MM TD CR, 44 VGPR TTP: 22.6 months, PFS: 14.9 NR months Dexamethasone CR, 16 VGPR TTP: 6.5 months, PFS: 6.5 months SWOG S RevHD CR 1-year PFS: 77%, 1-year OS: 93% NR HD CR 1-year PFS: 55%, 1-year OS: 91% GIMEMA 97,110 MPR 53/21 (No.) at MTD 81 at MTD 24 CR, 48 VGPR at MTD 1-year EFS: 92%, 1-year OS: 100%, TTP/PFS: 28.5 months at MTD 2-year OS: 91% at MTD Offidani et al 95 ThaDD CR, 58 VGPR 3-year TTP: 60%, 3-year EFS: 57%, 3-year OS: 74% (continued on following page) CR v PR v no response: longer TTP, TNT, TFI, OS CR v VGPR: longer TTP, TNT, TFI 96 PFS prolonged in patients achieving CR v PR (P.005) and in patients achieving VGPR v PR (P.0003) VGPR v VGPR correlated with better PFS (HR 0.64, P.02) in overall population VGPR v VGPR: 1-year EFS 100% v 88% VGPR v VGPR: 3-year TTP 78% v 40%, 3-year EFS 78% v 37%, 3-year OS 84% v 61% by American Society of Clinical Oncology 2617

7 Chanan-Khan and Giralt Table 4. Clinical Outcomes After Novel-Agent-Containing First-Line Therapy in Patients With MM Not Undergoing HDT-ASCT: Phase III Studies and Phase II Trials Reporting Outcomes by Response (continued) Study by Regimen No. ORR (%) CR/nCR/VGPR Rates Long-Term Outcomes Outcomes by Best Response Hussein et al 94 DVD-T CR, 49 VGPR PFS: 28.2 months CR/VGPR v PR/no response: longer PFS (P.01) and OS (P.03) Abbreviations: CR, complete response; DOR, duration of response; DVD-T, liposomal doxorubicin, vincristine, dexamethasone, thalidomide; ECOG, Eastern Cooperative Oncology Group; EFS, event-free survival; GEM, Grupo Español de Mieloma; GIMEMA, Gruppo Italiano Malattie Ematologiche dell Adulto; GISMM, Gruppo Italiano per lo Studio del Mieloma Multiplo; HOVON, Hemato-Oncologie voor Volwassenen Nederland; IFM, Intergroupe Francophone du Myèlome; MEL100, melphalan 100 mg/m 2 autologous stem cell transplant; MP, melphalan, prednisone; MPR, melphalan, prednisone, lenalidomide; MPT, melphalan, prednisone, thalidomide; MTD, maximum tolerated dose; ncr, near-cr; NMSG, Nordic Myeloma Study Group; NR, not reported; ORR, overall response rate; OS, overall survival; PETHEMA, Programa para el Estudio de la Terapéutica en Hemopatía Maligna; PFS, progression-free survival; PR, partial response; RevHD, lenalidomide, high-dose dexamethasone; SWOG, Southwest Oncology Group; TD, thalidomide, dexamethasone; TFI, treatment-free interval; ThaDD, thalidomide, liposomal doxorubicin, dexamethasone; TNT, time to next therapy; TTP, time to progression; VAD, vincristine, doxorubicin, dexamethasone; VGPR, very good partial response; VMP, bortezomib, melphalan, prednisone; VMPT, VMP, thalidomide; VTP, bortezomib, thalidomide, prednisone. Updated data presented at Annual Meeting of the American Society of Hematology, December 6-9, 2008, San Francisco, CA. Updated data presented at Annual Meeting of the European Hematology Association, June 4-7, 2009, Berlin, Germany. significantly longer TTP (P.004), time to next therapy (TTNT; P.014), and treatment-free interval (TFI, P.004), and longer OS, with TTPsimilarlyprolongedamongpatientsachievingCRversusVGPR(P.019) by IMWG criteria. 96 In addition, in a phase III study of thalidomide plus MP (MPT) versus MP, patients achieving VGPR or better had significantly longer PFS (P.02) but not OS compared with those achieving less than VGPR, 98 whereas in a phase I/II study of lenalidomide plus MP, EFS appeared longer in patients achieving VGPR or better. 97 Furthermore, two phase II studies of thalidomide-based regimens have demonstrated the prognostic importance of achieving CR/VGPR. In a study of liposomal doxorubicin, vincristine, dexamethasone, and thalidomide, PFS (P.01) and OS (P.03) were significantly longer in newly diagnosed patients achieving CR/VGPR versus PR/no response, after adjusting for various confounding factors. 94 A study of thalidomide, dexamethasone, and liposomal doxorubicin demonstrated significantly higher 3-year TTP (78% v 40%, P.031), EFS (78% v 37%, P.021), and OS (84% v 61%, P.053) rates among patients achieving VGPR versus less than VGPR. 95 Evidence From Relapsed/Refractory Studies The importance of achieving CR/maximal response, and the impact of novel agents on response rates, 111 has also been demonstrated in the relapsed setting. Table shows efficacy results from phase III studies of novel agents in patients with relapsed MM and from phase II studies in which outcomes were reported according to quality of response. Among patients who received bortezomib in the phase III Assessment of Proteasome Inhibition for Extending Remissions (APEX) trial, 113 TTP, OS, TTNT, and TFI were evaluated in patients achieving CR, ncr/vgpr, PR, minimal response (MR), and no response (NR). Median TTNT and TFI were significantly longer in patients achieving CR (27.1 and 24.1 months, respectively), compared with patients achieving ncr/vgpr (13.6 and 6.9 months), PR (14.0 and 6.4 months), MR (8.7 and 3.8 months), or NR (6.2 and 2.3 months). Figure 1 shows the Kaplan-Meier plot of median OS stratified by quality of response; the curves indicate differences in long-term outcome between response categories. 113 Similarly, a pooled analysis of outcomes among patients achieving CR/VGPR or PR in the MM-009 and MM-010 phase III studies of lenalidomide and dexamethasone versus dexamethasone demonstrated a differential prognostic impact. 116 Duration of response and TTP were both significantly longer with CR/VGPR versus PR, although OS appeared similar. However, OS was prolonged in patients achieving CR compared with PR. 116 In addition, reflecting findings from the first-line setting, another pooled analysis of MM-009 and MM-010 demonstrated that prolonged maintenance of response with lenalidomide and dexamethasone was associated with longer OS. 119 The prognostic importance of CR/VGPR versus PR or less than PR has also been shown in analyses of two phase II studies, of bortezomib, liposomal doxorubicin, and dexamethasone, 118 and liposomal doxorubicin, vincristine, dexamethasone, and thalidomide, 94 in patients with relapsed MM (Table 5). ARGUMENTS AGAINST THE IMPORTANCE OF CR Not all studies have shown that CR/maximal response is prognostic for OS or that there is an association between higher CR rates and improved outcomes. 53, There are a number of possible reasons why these apparently counterintuitive findings may arise. CR may have greater or lesser prognostic significance depending on the characteristics of the population being analyzed. For example, CR was not associated with prolonged OS in an analysis of patients with a history of monoclonal gammopathy of undetermined significance (MGUS) or smoldering MM before active MM, possibly as a result of a more indolent disease course in these patients obscuring the benefits of achieving CR versus PR. 124 Conversely, CR was shown to have greater prognostic importance for OS in patients with high-risk versus standard-risk MM defined by gene-expression profiling. 125 Small sample size, whether a result of limited total patient numbers or a result of low CR rates in studies of older conventional therapies, may also be an important reason for a lack of association between response and outcome. In an analysis of long-term data from a Group-Myelome Autogreffe study, 53 median OS was 59 months among 34 patients achieving CR or minimal residual disease (MRD) post-transplant, compared with 40.5 months among 37 patients achieving lesser responses; however, the OS curves were not significantly different (P.22), possibly as a result of the small patient numbers. This may particularly be the case if analyses are conducted after limited follow-up. Indeed, maturity of data may be an important factor in whether a correlation is seen; studies with only limited follow-up, and consequently immature OS data, may not show a by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

8 Importance of CR in MM and the Impact of Novel Agents Table 5. Response Rates and Long-Term Outcomes With Novel Agent Containing Therapy in Patients With Relapsed/Refractory MM: Phase III Studies and Phase II Trials Reporting Outcomes by Response APEX 64,112 Study by Regimen No. ORR (%) CR/nCR/VGPR Rates (%) Long-Term Outcomes Outcomes by Best Response Single-agent bortezomib CR/nCR DOR: 7.8 months, TTP: 6.2 months, OS: 29.8 months High-dose dexamethasone CR/nCR DOR: 5.6 months, TTP: 3.5 months, OS: 23.7 months DOXIL-MMY Bortezomib liposomal doxorubicin CR/nCR, 27 VGPR Single-agent bortezomib CR/nCR, 19 VGPR DOR: 10.2 months, TTP: 9.3 months, PFS: 9.0 months, 15- month OS: 76% DOR: 7.0 months, TTP: 6.5 months, PFS: 6.5 months, 15-month OS: 65% TTP: CR/VGPR/PR MR no response OS: CR/VGPR/PR MR no response TTAT and TFI: CR VGPR/PR MR no response 113 NR MM Lenalidomide dexamethasone CR/nCR TTP: 11.1 months, OS: 29.6 months CR/VGPR v PR: prolonged DOR (HR 2.56, Dexamethasone CR/nCR TTP: 4.7 months, OS: 20.2 months P.001) and TTP (HR 2.43, P.001), but similar OS (HR 1.29, P.294) CR v PR: MM prolonged OS (HR 2.03, P.038) 116 Lenalidomide dexamethasone CR/nCR TTP: 11.3 months, OS: not reached Dexamethasone CR/nCR TTP: 4.7 months, OS: 20.6 months Palumbo et al 118 Bortezomib liposomal doxorubicin dexamethasone Hussein et al 94 Liposomal doxorubicin, vincristine, dexamethasone, thalidomide CR, 25 VGPR CR, 45 VGPR 1-year EFS: 34%, 1-year OS: 66% PFS: 15.5 months, OS: 39.9 months NR CR/VGPR v PR: 1-year EFS 83% v 16% (HR 4.90, P.02); 1-year OS 90% v 63% (HR 7.09, P.06) CR/VGPR v PR/no response: longer PFS (P.01) and OS (P.03) Abbreviations: ORR, overall response rate; CR, complete response; ncr, near-complete response; VGPR, very good partial response; DOR, duration of response; TTP, time to progression; OS, overall survival; MR, minimal response; TTAT, time to alternative therapy; TFI, treatment-free interval; NR; no response; PFS, progression-free survival; HR, hazard ratio; EFS, event-free survival. significant correlation between quality of response and survival, which may only emerge after prolonged follow-up. In the Southwest Oncology Group (SWOG) S0232 trial of lenalidomide and dexamethasone versus dexamethasone alone in patients with newly diagnosed MM, the combination arm resulted in a significantly higher CR rate (22% v 4% with dexamethasone; P.001), and while this was associated with a significant improvement in 1-year PFS rate (77% v 55%; P.002), no difference was seen in 1-year OS rate (93% v 91%). 109 Another important reason for the absence of an OS benefit associated with a higher CR rate in studies in the first-line setting may be that the comparison between regimens is confounded by subsequent therapies used. This may particularly be an issue in studies of novel agent based regimens versus previous standards of care, in which substantial numbers of patients from the comparator arm either cross over to receive the experimental treatment or receive highly active novel agent based therapies after relapse. Palumbo et al 98 suggested that this was the reason for no OS advantage being seen with MPT versus MP as first-line therapy in elderly patients with MM, despite better responses and prolonged PFS with MPT. By contrast, in the VISTA trial the OS advantage seen with VMP versus MP was not affected by subsequent therapy use. 100 An additional factor requiring consideration, particularly with regard to older studies, may be that the definition of CR or maximal response is not stringent enough to result in sufficient prognostic Proportion of Patients CR (n = 27) VGPR (n = 31) PR (n = 77) MR (n = 21) NR (n = 159) All bortezomib (n = 315) * * Time (months) * * * * ** Fig 1. Overall survival among all response-evaluable patients treated with bortezomib in the APEX trial (n 315) and according to quality of response to bortezomib. Reproduced with permission from Niesvizky et al. 113 The relationship between quality of response and clinical benefit for patients treated on the bortezomib arm of the international, randomized, phase III APEX trial in relapsed multiple myeloma. CR, complete response; VGPR, very good partial response; PR, partial response; MR, minimal response; NR, no response by American Society of Clinical Oncology 2619

9 Chanan-Khan and Giralt differentiation compared with lesser categories of response. In a combined analysis of results from four SWOG trials involving 1,555 patients, outcomes were compared among patients achieving remission by SWOG criteria ( 75% reduction in tumor mass, previously identified as CR), PR (50% to 74% reduction), and no response. 60 No differences were seen in median OS by response achieved at 6-month and 12-month landmarks, but such response classifications may have obscured the OS benefit in any patients achieving true CR by the more stringent EBMT or IMWG criteria. Nevertheless, this analysis did indicate that absence of progression by the 6-/12-month landmarks was associated with prolonged OS, 60 supporting the suggestion that durability of response is of key prognostic importance. As discussed in the following section, determining CR by more sensitive methods of analysis or more stringent criteria will distinguish patients with a very high degree of response from those with more substantial residual tumor burden. It is likely that the more stringent the criteria or sensitive the analytic techniques used, the stronger the prognostic value of CR. BEYOND CR: IMPROVED OUTCOMES WITH GREATER DEPTH OF RESPONSE While standard CR by the established EBMT or IMWG criteria has been shown to have differential prognostic impact compared with lesser degrees of response, it does not necessarily represent a patient s maximal response. More sensitive techniques than immunofixation are being explored to define increasingly stringent degrees of response or elimination of MRD, including, as previously noted, the use of immunohistochemistry or immunofluorescence in the definition of scr in the IMWG criteria. 22 Furthermore, MRD may be evaluated, and thus CR more stringently defined, using multiparametric flow cytometry 28,29,31,126 and polymerase chain reaction (PCR) 31-33, ; demonstrating eradication of MRD by these techniques has been shown to be prognostic for improved outcomes. Notably, in a recent analysis of patients uniformly treated in the GEM2000 protocol, MRD eradication by multiparameter flow cytometry was shown to be prognostic for prolonged PFS and OS and to have differential prognostic significance for OS among patients with standard immunofixation-negative CR. 28 Median PFS was 71 months versus 37 months (P.001) among patients who were MRD negative versus MRD positive by multiparameter flow cytometry, and median OS was not reached versus 89 months (P.002). In addition, patients who were MRD negative but remained immunofixation positive nevertheless had significantly longer PFS compared with patients who were MRD positive and immunofixation negative, demonstrating the clinical importance of MRD evaluation by multiparameter flow cytometry. 28 Similarly, another analysis of patients treated in GEM2000 showed that patients who were MRD negative by fluorescent PCR had better PFS than those remaining MRD positive (68% v 28%; P.001), with similar findings even in patients achieving immunofixation-negative CR (75% v 25%; P.002). 130 In addition, in a study of patients who had received an allogeneic transplant, the cumulative rate of relapse at 5 years was 0%, 33%, and 100% among patients who were persistently negative, mixed negative/positive, and consistently positive, respectively, by PCR. 128 Likewise, in a study of bortezomib, thalidomide, and dexamethasone as consolidation after HDT-ASCT in 40 patients, no clinical relapses were seen among six PCR-negative patients after a median follow-up of 26 months, with only one patient experiencing a molecular relapse at 24 months. By contrast, among patients not achieving molecular remission by PCR, eight had relapsed in a median of 12 months. 131 The findings indicate that molecular remission by PCR is key to long-term disease-free survival and, potentially, OS. Bone marrow magnetic resonance imaging (MRI) has been demonstrated to be an effective technique for routine determination of disease burden in patients with MM, 132 and use of MRI to detect focal lesions harboring viable monoclonal plasma cells may also improve the prognostic significance of CR. 133 Eradication of myeloma cells in focal lesions has been shown to offer superior OS, 34 and routine use of MRI has been recommended for documenting a superior CR compared with that defined by standard metastatic bone surveys. 34,132 DISCUSSION Prognosis in MM is a highly complex issue, with OS being influenced by multiple factors. Nevertheless, the extensive data reviewed herein provides strong evidence that achieving CR is an important prognostic factor at all stages of treatment, including both before and after HDT- ASCT, with first-line treatment in the nontransplant setting, and in the relapsed setting. Data from phase III and other studies are demonstrating that treatment regimens incorporating the novel agents bortezomib, thalidomide, and lenalidomide are associated with very high rates of CR compared with previous standard regimens, and the introduction of these agents has already been associated with improved outcomes among patients with MM in general. 58,59 Given the wealth of evidence from clinical studies, as reviewed here, CR is now recognized as a surrogate for OS, and as a clinically relevant end point in MM studies. 134 Therefore, it is reasonable to consider CR as an important therapeutic goal in MM therapy at all stages of treatment. Thus, it is important that patients are regularly observed and the required tests, such as immunofixation and bone marrow assessment, routinely conducted to confirm CR, as is practiced in the treatment of patients with CML. 19,20 As demonstrated by data from clinical studies, immunofixation status confers important differential prognostic information, 76 as does meeting the criteria for CR by bone marrow examination. 81 In patients remaining immunofixation positive or having 5% plasma cells in their bone marrow, continuing/additional treatment with the aim of eliminating these residual signs of disease may be important for improving long-term outcomes, as may the use of maintenance therapy to prolong the durability of CR. It is also important to note that not all CRs, as defined by current standard response criteria, are necessarily equal. The IMWG criteria already differentiate scr from standard CR using the free light chain assay and analysis by immunofluorescence or immunohistochemistry. In addition, emerging data from exploratory studies using highly sensitive techniques to evaluate MRD and define CR more stringently, such as multiparameter flow cytometry and PCR, suggest that the more stringent the definition used for CR, the greater the prognostic significance of achieving that degree of response. This parallels experience in CML, in which hematologic, cytogenetic, and molecular CRs result in increasingly better prognosis. 19,20 Indeed, based on the findings of studies reviewed herein, further refinement of MM response by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

10 Importance of CR in MM and the Impact of Novel Agents criteria with revised definitions of CR, reflecting the prognostic importance of these greater depths of response, may be warranted. Although preliminary, these observations are potentially useful in clinical practice as these more sensitive techniques, such as multiparameter flow cytometry, are relatively easy to perform. Clinicians can use them to monitor depth of remission in their patients, as is routine in CML, 19,20 and thereby inform evidence-based decisions regarding ongoing or subsequent therapy. Use of such assessments would help inform decisions regarding whether patients should undergo first or second transplantation as part of first-line therapy. It has been suggested that the high rates of CR with novel agent based induction therapy may obviate the need for HDT-ASCT in a proportion of patients in the first-line setting, as in some analyses, consolidation with transplant after achieving CR with induction therapy does not appear to offer an OS benefit. 1 However, prospective studies are warranted using sensitive techniques to determine whether HDT-ASCT results in further elimination of MRD, and thus improves prognosis, in patients achieving CR to novel-agent based induction regimens. Consolidation with HDT-ASCT may be warranted for patients achieving CR to induction but not MRD-negative status by multiparameter flow cytometry or PCR, with the aim of deepening and prolonging their CR. Similarly, use of such sensitive monitoring techniques may help inform decisions about the use/continuation of maintenance therapy; as noted, one study has already demonstrated the prognostic significance of conversion to MRD-negative status by PCR using bortezomib, thalidomide, and dexamethasone maintenance. 131 Use of HDT-ASCT and/or maintenance for sustaining CR may be of particular importance in patients with high-risk cytogenetic abnormalities, as discussed earlier. Finally, the issue of patient management on relapse from a CR defined using sensitive assessment techniques must be considered. Oncologists should also be aware of what constitutes relapse from CR and whether changes in status are of practical relevance regarding ongoing management, both clinically and in consideration of patients perceptions regarding the success of their treatment; for example, a change from immunofixation-negative to immunofixation-positive status or from MRD-negative to MRD-positive status by PCR does not necessarily warrant starting a patient on further treatment or changing the current treatment. However, it may indicate the need for more intensive follow-up to determine whether the patient is beginning to experience true clinical relapse or whether the findings represent temporary biochemical or molecular changes in disease status. The definition of progression from CR in the IMWG criteria 22 corresponds more closely with a clinical need for subsequent treatment. In conclusion, CR is an important and clinically meaningful goal in MM therapy, demonstrating a survival benefit at all stages of treatment. Encouragingly, CR as a treatment goal is increasingly achievable after the introduction of HDT-ASCT and the recent development of potent regimens incorporating the novel agents. Careful monitoring of patients and assessment of depth of response should be routinely considered, as defined in the published guidelines for evaluating response in MM. 21,22 Furthermore, there is the need for further development and validation of more sensitive techniques as routine methods for assessment of disease status, given the demonstrated differential prognostic significance of these analyses. Such monitoring will inform decisions on optimal patient management, hopefully thereby ultimately resulting in improved outcomes. AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a U are those for which no compensation was received; those relationships marked with a C were compensated. For a detailed description of the disclosure categories, or for more information about ASCO s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: None Consultant or Advisory Role: Asher A. Chanan-Khan, Millennium Pharmaceuticals (C), Celgene (C), ImmunoGen (C) Stock Ownership: None Honoraria: Asher A. Chanan-Khan, Millennium Pharmaceuticals, Celgene, ImmunoGen; Sergio Giralt, Celgene, Millennium Pharmaceuticals, Genzyme Research Funding: None Expert Testimony: None Other Remuneration: None AUTHOR CONTRIBUTIONS Conception and design: Asher A. Chanan-Khan, Sergio Giralt Administrative support: Asher A. Chanan-Khan Collection and assembly of data: Sergio Giralt Manuscript writing: Asher A. Chanan-Khan, Sergio Giralt Final approval of manuscript: Asher A. Chanan-Khan, Sergio Giralt REFERENCES 1. Wang M, Delasalle K, Feng L, et al: CR represents an early index of potential long survival in multiple myeloma. Bone Marrow Transplant /bmt [epub ahead of print, March 2010] 2. Barlogie B, Anaissie EJ, van Rhee F, et al: Total therapy (TT) for myeloma (MM) 10% cure rate with TT1 suggested by 10yr continuous complete remission (CCR): Bortezomib in TT3 overcomes poor-risk associated with T(4;14) and DelTP53 in TT2. J Clin Oncol 26:458s, 2008 (abstr 8516) 3. Bensinger WI: The current status of reduced-intensity allogeneic hematopoietic stem cell transplantation for multiple myeloma. Leukemia 20: , Barlogie B, Tricot G, Anaissie E, et al: Thalidomide and hematopoietic-cell transplantation for multiple myeloma. N Engl J Med 354: , Barlogie B, Anaissie E, van RF, et al: Incorporating bortezomib into upfront treatment for multiple myeloma: Early results of total therapy 3. Br J Haematol 138: , Kumar SK, Therneau TM, Gertz MA, et al: Clinical course of patients with relapsed multiple myeloma. Mayo Clin Proc 79: , Greipp PR, San Miguel J, Durie BG, et al: International staging system for multiple myeloma. J Clin Oncol 23: , Ailawadhi S, Sher T, Patel M, et al: International Staging System (ISS) is superior to Durie- Salmon (DS) staging in predicting overall mortality in multiple myeloma (MM). Blood 112:948a, 2008 (abstr 2745) 9. Greipp PR, Lust JA, O Fallon WM, et al: Plasma cell labeling index and beta 2-microglobulin predict survival independent of thymidine kinase and C-reactive protein in multiple myeloma. Blood 81: , Pineda-Roman M, Zangari M, Haessler J, et al: Sustained complete remissions in multiple myeloma linked to bortezomib in total therapy 3: Comparison with total therapy 2. Br J Haematol 140: , Barlogie B, Tricot GJ, van RF, et al: Longterm outcome results of the first tandem autotransplant trial for multiple myeloma. Br J Haematol 135: , Kyle RA, Leong T, Li S, et al: Complete response in multiple myeloma: Clinical trial E9486, an Eastern Cooperative Oncology Group study not involving stem cell transplantation. Cancer 106: , Stewart AK, Bergsagel PL, Greipp PR, et al: A practical guide to defining high-risk myeloma for clinical trials, patient counseling and choice of therapy. Leukemia 21: , by American Society of Clinical Oncology 2621