Long-term outcome of patients with mutiple myeloma-related advanced renal failure following auto-sct

Bone Marrow Transplantation (2013) 48, 1543 1547 & 2013 Macmillan Publishers Limited All rights reserved 0268-3369/13 www.nature.com/bmt ORIGINAL ARTICLE Long-term outcome of patients with mutiple myeloma-related advanced renal failure following auto-sct SV Glavey 1, MA Gertz 2, A Dispenzieri 2, S Kumar 2, F Buadi 2, M Lacy 2, SR Hayman 2, D Dingli 2, A McCurdy 2, WJ Hogan 2, DA Gastineau 2 and N Leung 2,3 Renal failure commonly complicates multiple myeloma (MM) and is associated with reduced survival. It is not clear whether auto- SCT results in improved renal function or attainment of independence from dialysis in patients with advanced renal impairment due to MM. We conducted a retrospective cohort study of all patients who underwent auto-sct for MM complicated by advanced renal failure at our institution over a 10-year period (2000 2010). We aimed to assess the association between auto-sct and renal outcome in patients with serum creatinine (SCr) over 3 mg/dl, attributable to MM, including those who were dialysis dependent. Thirty patients (2.8% of all auto-sct patients) met inclusion criteria. Fourteen of 15 patients who were dialysis dependent before auto-sct remained dialysis dependent in the long term despite hematological response (HR). Of the remaining 15 patients with SCr 43 mg/dl, an improvement in glomerular filtration rate (GFR) from 15 to 19.4 ml/min/1.73 m 2 was noted post auto-sct (P ¼ 0.035); however, neither HR post auto-sct or pre-existing renal function were independently associated with renal outcome. Auto-SCT was not associated with independence from dialysis in patients with renal failure due to MM at our institution. Although auto-sct was associated with an improvement in GFR in patients with SCr 43 mg/dl, this improvement was not related to HR. Bone Marrow Transplantation (2013) 48, 1543 1547; doi:10.1038/bmt.2013.109; published online 5 August 2013 Keywords: auto-sct; renal; dialysis INTRODUCTION Renal impairment is a common complication of multiple myeloma (MM) at presentation and is associated with a higher early mortality and reduced OS. 1,2 Blade et al. 3 reported that early recovery of renal function in MM is associated with improved survival. MM cast nephropathy is the pathology most often associated with severe renal impairment in these patients. The prevalence of renal failure at presentation of MM seen throughout the literature is of the order of 20 30%. 3,4 Approximately 8% of all patients with MM will require long-term dialysis, 3,5 8 and dialysisdependent renal failure is itself an independent poor prognostic factor for survival. 7 Auto-SCT following high-dose chemotherapy has been established as optimal therapy for patients with MM. 9,10 The introduction of novel agents to the transplant sequence has resulted in an unprecedented increase in PFS in recent years. Coupled to this success is the increased focus on individually tailored therapies for MM patients. For patients with advanced renal disease, freedom from dialysis is an important goal of therapy. Patients with milder degrees of renal impairment may avoid dialysis altogether, and the associated increased mortality, if an improvement in renal function can be attained with auto-sct. The current literature pertaining to renal recovery following auto- SCT is conflicting. Some studies have shown an improvement in renal function of up to 32 34%. 11,12 However, in patients receiving renal replacement therapy at the time of transplant, only 5 14% regain independence from dialysis following initial auto-sct for MM. 13,14 Despite a high rate of partial (PR) and complete () hematological response (HR) of MM in these studies conducted in the novel agent era, renal recovery was poor. The question of renal response following auto-sct is important for patients and clinicians, as they consider goals of therapy, particularly for dialysis-dependent patients. To address this question, we reviewed our experience in a retrospective cohort study using electronic medical records and renal biopsy data from all patients undergoing auto-sct for MM from 2000 to 2010 at Mayo Clinic, Rochester, MN, USA. MATERIALS AND METHODS We reviewed the records of all patients who underwent auto-sct for MM at Mayo Clinic, Rochester from 01/2000 to 01/2010. We selected, from the entire cohort, all patients who had advanced renal impairment defined as a serum creatinine (SCr) 43 mg/dl, or who were dialysis dependent at the time of transplantation. Although other studies have selected a cutoff of 2 mg/dl, we chose this higher level of SCr to select patients who were close to needing dialysis at the time of auto-sct, for whom attainment of a renal response was critical. All patients with end-stage kidney disease or chronic kidney disease not related to the development of MM were excluded. Patients who had developed acute renal failure in the days before auto-sct were also excluded. Clinical information for all patients was captured prospectively in a continuously updated database. Follow-up data were available until the time of death or the time of analysis on all patients. No patients were lost to follow-up. All patients provided written informed consent prospectively for use of their medical records and access to renal biopsy data. Approval was obtained for the study from the Mayo Clinic Institutional Review Board in accordance with federal regulations and the Declaration of Helsinki. Baseline demographics were recorded for all patients. The SCr level used in our analysis was the value measured immediately before auto-sct. We chose this value in order to isolate the effect of auto-sct on renal function from that of upfront therapy. In addition, creatinine clearance or iothalamate clearance (ml/min/1.73 m 2 ) 1 Department of Hematology, Galway University Hospital, Galway, Ireland; 2 Division of Nephrology, Mayo Clinic, Rochester, MN, USA and 3 Division of Hematology, Mayo Clinic, Rochester, MN, USA. Correspondence: Dr SV Glavey, Department of Hematology, Galway University Hospital, Newcastle Road, Galway, Ireland. E-mail: siobhan.glavey@nuigalway.ie Received 18 March 2013; revised 11 June 2013; accepted 13 June 2013; published online 5 August 2013

1544 was measured in all patients who were not anuric immediately before auto-sct. The cause of kidney disease was recorded as found at renal biopsy report. The stage of MM at the time of auto-sct was defined using Durie Salmon criteria. 3 Post auto-sct SCr was measured between 1 and 3 months to assess the immediate effect of auto-sct on renal function. Long-term renal and hematological outcome data were also collected for all patients until time of analysis or death. Estimated glomerular filtration rate (egfr) was calculated using the modification of diet in renal disease (MDRD) equation. 15 The International Myeloma Working Group Uniform Response Criteria for Multiple Myeloma were used to assess response to auto-sct in all patients. 5 All interval complications were noted, including acute kidney injury, sepsis, mucositis, diarrhea and the need for dialysis in the post transplant follow-up period. Statistical methods The minimally important difference in outcome following auto-sct considered to be clinically significant was avoidance of dialysis for patients with advanced renal impairment and freedom from dialysis for those already requiring it before auto-sct. Descriptive statistics were carried out using SPSS (SPSS Inc., Chicago, IL, USA). For non-parametric analysis of paired samples, we used the Wilcoxon signed-rank test and for unpaired sample data, we used the Mann Whitney U-test. Parametric analysis for paired samples was carried out using the paired samples t-test. Univariate analysis was carried out to determine any association between explanatory variables and renal function following auto-sct. Survival probability curves were constructed using the Kaplan Meier method, and survival was compared using the log-rank test. RESULTS Baseline characteristics Out of 1074 patients, 30 patients (2.8%) had a SCr level over 3 mg/dl or were dialysis dependent before auto-sct for MM. Characteristics of the group are shown in Table 1. The median age was 61 years (range 37 72), 18 patients (60%) were male. The median egfr of the entire group before SCT was 12 ml/min/ 1.73 m 2 and, following auto-sct, the median egfr was higher at 13.1 ml/min/1.73 m 2. Before undergoing auto-sct, eight (26.6%) patients had achieved a PR and a further five patients (16.7%) had achieved a very good partial response, 11 (36.6%) patients had stable disease and four (13.3%) patients had relapsed MM. One patient who was dialysis dependent had achieved a. Nineteen (63.3%) patients were Durie Salmon stage 3(b). Twenty-nine patients (96.6%) received conditioning with melphalan 140 mg/m 2 with one patient receiving melphalan 120 mg/m 2. All patients had either one or two chemotherapy regimens before auto-sct, with one patient having three prior regimens. The cause of renal disease was cast nephropathy seen on renal biopsy in 18/30 patients (60%). Direct plasma cell infiltration was the cause of renal disease in one patient. Acute tubular necrosis was noted on biopsy in seven patients either alone or in association with cast nephropathy. One patient had renal failure attributed to platinum toxicity, and three patients related to non-steroidal use in combination with cast nephropathy. At the time of follow-up, 16 patients were alive. Of the 30 selected patients, 26 (86.7%) had a HR according to International Myeloma Working Group criteria in response to auto-sct, with achieved for 12 patients (40%). One patient in the low-clearance group died in the stemcell mobilization period because of sepsis. There were no early deaths in the group of patients who were dialysis dependent before auto-sct. Dialysis-dependent group Fifteen patients were dialysis dependent at the time of transplantation. All of these patients, except one, were undergoing dialysis three times per week. The median egfr of the eleven patients who had their creatinine clearance measured before auto-sct was 8 ml/min/1.73 m 2 (range 7 14). Six patients had their creatinine clearance measured following auto-sct with a median of Table 1. Patients characteristics Entire cohort Low-clearance group Dialysis-dependent group N 30 15 15 Median time from Dx to Auto-SCT (days) 209 172 237 Age 61 (37 72) 61 (37 71) 60 (43 72) Sex (male) 18 (60%) 11 (73.3%) 7 (46.6%) Pre auto-sct egfr ml/min/1.73 m 2 12 (7.5 16.4) 15.0 (11.7 16.9) 8 (7 14) Post auto-sct egfr ml/min/1.73 m 2 13.1 (10.1 20.7) 19.4 (13.1 24.4) 11 (8 14) Durie Salmon stage 3(b) 19 (63.3%) 10 (66.6%) 9 (60%) Melphalan dosage 140 mg/m 2 29 14 15 120 mg/m 2 1 1 Number of prior chemotherapy regimens 1 18 10 8 2 11 5 6 3 1 1 Renal pathology Cast nephropathy 18 8 10 Direct plasma cell infiltration 1 1 Acute tubular necrosis 7 4 3 Platinum toxicity 1 1 NSAIDS 3 3 Median follow-up (months) 40.7 Pre auto-sct hematological status /VGPR/pPR (n) 1/5/23 0/3/11 1/2/12 Post auto-sct hematological status /VGPR/pPR (n) 12/10/8 5/5/4 7/5/3 Abbreviations: ¼ complete hematological response; Dx ¼ diagnosis; egfr ¼ estimated glomerular filtration rate; NSAIDS ¼ non-steroidal anti-inflammatory drugs; PR¼ partial hematological response; VGPR ¼ very good partial response. Values median and range for GFR and age, others n(%). Bone Marrow Transplantation (2013) 1543 1547 & 2013 Macmillan Publishers Limited

11 ml/min/1.73 m 2 (range 8 14). One of these patients had an improvement in creatinine clearance from 12 to 18 ml/min of egfr, but attempts to stop dialysis were unsuccessful. The difference in egfr of 3 ml/min/1.73 m 2 was not statistically significant (P ¼ 0.06). The patients who did not have a SCr clearance measured were either anuric or were not clinically expected to have improved renal function based on dialysis parameters. Low-clearance group Fifteen patients were found to have low creatinine clearance; creatinine 43 g/dl, but were not dialysis dependent at the time of transplantation. Baseline characteristics for this group are shown in Table 1 and were comparable to that of the dialysis-dependent group. The median age was 61 years (range 37 71). The median egfr before harvest was 15 ml/min/1.72 m 2 (range 11.7 16.9) and post transplant median egfr was19.4 ml/min/1.73 m 2 (range 13.1 24.4). The improvement in egfr was statistically significant (P ¼ 0.035) (Figure 1). Only one patient required hemodialysis during the peri-transplant period (day 0 100) This patient s renal function returned to baseline 17 days post auto-sct and dialysis was not required in the long term. Predictors of renal recovery When classifying the entire group of 30 patients based on response status, the median egfr before auto-sct for those who egfr(ml/min/1.73m 2 ) 35 30 25 20 15 10 5 egfr pre auto-sct egfr post auto-sct Low clearance group Figure 1. Boxplot demonstrating median egfr in the low-clearance group pre and post auto-sct. Pre auto-sct ¼ 15.0 ml/min/1.73 m 2. Post auto-sct ¼ 19.4 ml/min/1.73 m 2 (P ¼ 0.035). 1545 did not go on to achieve a was 13.2 ml/min/1.73 m 2. Those who did achieve a (n ¼ 12) following auto-sct had a higher median pre-transplant egfr of 17.5 ml/min/1.73 m 2. Following auto-sct, the median egfr for the patients who did not achieve a was 13.5 ml/min/1.73 m 2 and for those who did achieve a, the median egfr was 23.9 ml/min/1.73 m 2, Figure 2b. This improvement in egfr based on response status was not statistically significant for either group (P ¼ 0.52). Importantly, in univariate logistic regression analysis, reponse status was not independently associated with an improvement in egfr regardless of whether patients were dialysis dependent before auto-sct. Furthermore, considering the entire group of 30 patients, on univariate analysis, improvement in egfr was not independently associated with age at auto-sct, renal function before auto-sct or time from MM diagnosis to transplant (Table 2). Long-term follow-up Although initial assessment of renal response was made in the first 3 months following auto-sct, all patients were followed for renal and hematological outcomes over the 10-year study period. During long-term follow-up of the low-clearance group, four patients progressed to end-stage kidney disease following auto- SCT and became dialysis dependent. One of these patients had achieved a following auto-sct, the remaining three patients achieved at least a PR. Of the 15 dialysis-dependent patients, one patient became dialysis independent following auto-sct. This patient s renal function had been improving before auto-sct with an iothalomate clearance of 19 ml/min of egfr. This patient became independent of hemodialysis 17 days after transplantation. This observation was not coincident, at that time, with HR or reduction in plasma or urine paraprotein levels, and iothalamate clearance was unchanged after transplant at 20 ml/min of egfr; however, this patient did go on to achieve a in the long term. In terms of long-term survival, as expected, the OS of patients who were dialysis dependent at the time of auto-sct was shorter than that of patients with low creatinine clearance at 29.3 versus 64.7 months, respectively; log-rank P ¼ 0.06 (Figure 3). Treatment-related toxicity The treatment-related complications in the immediate post transplant period are outlined in Table 3. As previously noted, there was one death in the stem-cell mobilization period. The median time to neutrophil engraftment was 13 days and the median time to 50 000 platelets/ml platelets was 18 days. egfr pre auto-sct (ml/min/1.73m 2 ) 20.00 30.00 18.00 16.00 14.00 12.00 8.00 6.00 No egfr post auto-sct (ml/min/1.73m 2 ) 25.00 20.00 15.00 No Difference in egfr(ml/min/1.73m 2 ) 25.00 20.00 15.00 5.00 0.00-5.00 No Figure 2. Median GFR (a) pre and (b) post auto-sct according to response status. The median egfr before auto-sct for those who did not go on to achieve a was 13.2 ml/min/1.73 m 2 vs 17.5 ml/min/1.73 m 2 for those who did. (b) Post auto-sct, the median egfr for the patients who did not achieve a was 13.5 ml/min/1.73 m 2 vs 23.9 ml/min/1.73 m 2 for those who did. (c) The improvement in egfr based on response status was not statistically significant for either group (P ¼ 0.52). & 2013 Macmillan Publishers Limited Bone Marrow Transplantation (2013) 1543 1547

1546 Table 2. Predictors of renal recovery following auto-sct Univariate predictor variable b (95% CI) P-value Age 0.21 0.869 egfr pre auto-sct 0.368 0.187 Time to auto-sct 0.004 0.608 Abbreviations: CI ¼ class interval; egfr ¼ estimated glomerular filtration rate. Proportions surviving 1.0 0.9 0.8 0.7 Low CrCl 0.6 0.5 0.4 Dialysis 0.3 0.2 0.1 0.0 0 10 20 30 40 50 60 70 80 Months since ASCT Figure 3. Kaplan Meier survival probability curve demonstrating the difference in survival for the dialysis-dependent group before auto-sct vs low-clearance group 29.3 vs 64.7 months, respectively. (Log rank P ¼ 0.06). Table 3. Treatment-related toxicity Treatment-related toxicity n (%) Mucositis 11 (36.6%) Line sepsis 7 (23%) Bacteremia 5 (16.6%) Engraftment syndrome 1 (3.3%) Acute kidney injury requiring dialysis 1 (3.3%) DISCUSSION The superiority of high-dose chemotherapy followed by auto-sct in MM has been established with a higher rate of and EFS. 10,12 Renal failure is a common complication of MM at presentation and is seen in B20% of overall cases. 12,16 When deciding on treatment options for these patients, the clinical and prognostic significance of various degrees of impairment of renal function require close attention. For those who have reached ESRD, regaining independence from renal replacement therapy is an extremely important goal. We aimed to address whether auto-sct for MM was associated with an improvement in renal function or freedom from dialysis for patients with severe renal impairment secondary to this disease. We found that auto-sct is safe and efficacious in this challenging group of patients. High rates of HR are achieved and OS is quite respectable, especially in the low CrCl, dialysisindependent group. Unfortunately, there was little evidence of recovery of renal function in either group of patients. In our group of 30 patients with advanced renal impairment, only one patient was able to regain independence from renal replacement therapy after auto-sct. The OS of the dialysis-dependent group was disappointing with a median of o3 years, this might be in part owing to the fact that the majority of these patients were transplanted before the modern era of proteasome inhibitors and immune modulator drugs. Our rate of 40% is similar to that seen in the literature for patients undergoing auto-sct with MM and renal impairment. 1,13 In contrast, it was quite encouraging that among the patients with low creatinine clearance, who were not dialysis dependent, none of them required long-term dialysis post auto- SCT. Only one patient required dialysis in the immediate post transplant period. This possibly reflects the fact that these patients are identified as being at risk, and therefore nephrotoxic insults were minimized and efforts to reverse renal impairment were emphasized. Unfortunately, over time 26.6% of the low CrCl patients progressed to ESRD, which was related in all cases to progression of light chain cast nephropathy in the context of progressive disease. Strengths of this study include the fact that our study population has a high case load of MM with a wide distribution of disease spectrum, making these results generalizable to other transplant populations. Our large cohort over 10 years is suitable to investigate the independent effect of auto-sct on renal function, as the variability in treatment protocols and physician practices is minimal. The study of renal outcomes in MM is difficult because of the heterogeneous nature of MM-associated renal disease. This study contributes greatly to the paucity of data pertaining to the renal outcomes in MM patients post transplantation, particularly those with cast nephropathy and end-stage kidney disease who were well represented in this cohort. Our study has certain limitations. The first is the use of SCr X3 mg/dl as our cutoff, which may have led to the selection of patients with more advanced kidney disease, which is less likely to be reversible and to respond to treatment. However, we were specifically interested in this group of patients who make up the majority of treatment-resistant MM-related renal disease and pose a specific challenge for clinicians. SCr assays are known to be potentially affected by paraproteins that can spuriously elevate the SCr, which means that egfr calculated using SCr could potentially underestimate actual egfr in those with a large paraprotein. Therefore, we measured creatinine clearance or iothalamate clearance when possible, which are more reliable markers of egfr. In conclusion, we identified all patients who had advanced renal impairment or who were dialysis dependent at the time of auto- SCT for MM at our institution over a 10-year period. Previous studies have reported that auto-sct has a favorable impact on renal outcome in MM. According to our data, there is no clinically significant benefit in terms of renal outcome with auto-sct in MM. There is also no evidence from this study that auto-sct retards the progression of severe renal impairment in MM with a 26.6% progression to ESRD in the long term despite high levels of HR. We have found that auto-sct does not lead to deterioration in renal function in these patients and should be considered safe and efficacious in patients with advanced renal impairment. This highlights the importance of early intervention in these patients using effective therapies such as bortezomib in order to reverse renal damage before auto-sct is reached. 17,18 Auto-SCT is not precluded in patients who are dialysis dependent or have advanced renal impairment secondary to MM, however, it is important that these patients are carefully counseled. We propose caution in advising patients before auto- SCT that there is likely to be an improvement in renal function. Outlining realistic goals with patients is important with particular reference to dialysis, which has major quality of life implications for patients. Further studies are needed to help identify risk factors for progression of renal disease in these patients and to optimize patient selection for auto-sct. CONFLICT OF INTEREST The authors declare no conflict of interest. Bone Marrow Transplantation (2013) 1543 1547 & 2013 Macmillan Publishers Limited

REFERENCES 1 Gertz MA, Lacy MQ, Dispenzieri A, Hayman SR, Kumar S, Leung N et al. Impact of age and serum creatinine value on outcome after autologous blood stem cell transplantation for patients with multiple myeloma. Bone Marrow Transplant 2007; 39: 605 611. 2 San Miguel JF, Lahuerta JJ, Garcia-Sanz R, Alegre A, Blade J, Martinez R et al. Are myeloma patients with renal failure candidates for autologous stem cell transplantation? Hematol J 2000; 1: 28 36. 3 Blade J, Fernandez-Llama P, Bosch F, Montoliu J, Lens XM, Montoto S et al. Renal failure in multiple myeloma: presenting features and predictors of outcome in 94 patients from a single institution. Arch Intern Med 1998; 158: 1889 1893. 4 Knudsen LM, Hippe E, Hjorth M, Holmberg E, Westin J. Renal function in newly diagnosed multiple myeloma--a demographic study of 1353 patients. The Nordic Myeloma Study Group. Eur J Haematol 1994; 53: 207 212. 5 Matsue K, Fujiwara H, Iwama K, Kimura S, Yamakura M, Takeuchi M. Reversal of dialysis-dependent renal failure in patients with advanced multiple myeloma: single institutional experiences over 8 years. Ann Hematol 2010; 89: 291 297. 6 Dimopoulos MA, Richardson PG, Schlag R, Khuageva NK, Shpilberg O, Kastritis E et al. VMP (Bortezomib, Melphalan, and Prednisone) is active and well tolerated in newly diagnosed patients with multiple myeloma with moderately impaired renal function, and results in reversal of renal impairment: cohort analysis of the phase III VISTA study. J Clin Oncol 2009; 27: 6086 6093. 7 Knudsen LM, Hjorth M, Hippe E. Renal failure in multiple myeloma: reversibility and impact on the prognosis. Nordic Myeloma Study Group. Eur J Haematol 2000; 65: 175 181. 8 Torra R, Blade J, Cases A, Lopez-Pedret J, Montserrat E, Rozman C et al. Patients with multiple myeloma requiring long-term dialysis: presenting features, response to therapy, and outcome in a series of 20 cases. Br J Haematol 1995; 91: 854 859. 9 Attal M, Harousseau JL, Stoppa AM, Sotto JJ, Fuzibet JG, Rossi JF et al. A prospective, randomized trial of autologous bone marrow transplantation and chemotherapy in multiple myeloma. Intergroupe Francais du Myelome. New Engl J Med 1996; 335: 91 97. 10 Child JA, Morgan GJ, Davies FE, Owen RG, Bell SE, Hawkins K et al. High-dose chemotherapy with hematopoietic stem-cell rescue for multiple myeloma. New Engl J Med 2003; 348: 1875 1883. 11 Parikh GC, Amjad AI, Saliba RM, Kazmi SM, Khan ZU, Lahoti A et al. Autologous hematopoietic stem cell transplantation may reverse renal failure in patients with multiple myeloma. Biol Blood Marrow Transplant 2009; 15: 812 816. 12 Knudsen LM, Nielsen B, Gimsing P, Geisler C. Autologous stem cell transplantation in multiple myeloma: outcome in patients with renal failure. Eur J Haematol 2005; 75: 27 33. 13 Badros A, Barlogie B, Siegel E, Roberts J, Langmaid C, Zangari M et al. Results of autologous stem cell transplant in multiple myeloma patients with renal failure. Br J Haematol 2001; 114: 822 829. 14 Lee CK, Zangari M, Barlogie B, Fassas A, van Rhee F, Thertulien R et al. Dialysisdependent renal failure in patients with myeloma can be reversed by high-dose myeloablative therapy and autotransplant. Bone Marrow Transplant 2004; 33: 823 828. 15 Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999; 130: 461. 16 Kyle RA, Gertz MA, Witzig TE, Lust JA, Lacy MQ, Dispenzieri A et al. Review of 1027 patients with newly diagnosed multiple myeloma. Mayo Clin Proc 2003; 78: 21 33. 17 Ludwig H, Adam Z, Hajek R, Greil R, Tothova E, Keil F et al. Light chain-induced acute renal failure can be reversed by bortezomib-doxorubicin-dexamethasone in multiple myeloma: results of a phase II study. J Clin Oncol 2010; 28: 4635 4641. 18 Sonneveld P, Schmidt-Wolf IG, van der Holt B, El Jarari L, Bertsch U, Salwender H et al. Bortezomib induction and maintenance treatment in patients with newly diagnosed multiple myeloma: results of the randomized phase III HOVON-65/ GMMG-HD4 trial. J Clin Oncol 2012; 30: 2946 2955. 1547 & 2013 Macmillan Publishers Limited Bone Marrow Transplantation (2013) 1543 1547