Natural history of relapsed myeloma, refractory to immunomodulatory drugs and proteasome inhibitors: a multicenter IMWG study.

Introduction of new myeloma therapies offers new options for patients refractory to immunomodulatory drugs (IMiDs) and proteasome inhibitors (PIs). In this multicenter study, patients with relapsed multiple myeloma, who have received at least three prior lines of therapy, are refractory to both an IMiD (lenalidomide or pomalidomide) and a PI (bortezomib or carfilzomib), and have been exposed to an alkylating agent were identified. The time patients met the above criteria was defined as time zero (T0). Five hundred and forty-three patients diagnosed between 2006 and 2014 were enrolled in this study. Median age at T0 was 62 years (range 31-87); 61% were males. The median duration between diagnosis and T0 was 3.1 years. The median number of lines of therapy before T0 was 4 (range 3-13). The median overall survival (OS) from T0 for the entire cohort was 13 (95% confidence interval (CI) 11, 15) months. At least one regimen recorded after T0 in 462 (85%) patients, with a median (95% CI) progression-free survival and OS from T0 of 5 (4, 6), and 15.2 (13, 17) months, respectively. The study provides the expected outcome of relapsed multiple myeloma that is refractory to a PI and an IMiD, a benchmark for comparison of new therapies being evaluated.

Second primary malignancies in multiple myeloma: an overview and IMWG consensus.

Background: Therapeutic advancements following the introduction of autologous stem cell transplantation and 'novel' agents have significantly improved clinical outcomes for patients with multiple myeloma (MM). Increased life expectancy, however, has led to renewed concerns about the long-term risk of second primary malignancies (SPMs). This review outlines the most up-to-date knowledge of possible host-, disease-, and treatment-related risk factors for the development of SPMs in patients with MM, and provides practical recommendations to assist physicians. Design: A Panel of International Myeloma Working Group members reviewed the most relevant data published in the literature as full papers, or presented at meetings of the American Society of Clinical Oncology, American Society of Hematology, European Hematology Association, or International Myeloma Workshops, up to June 2016. Here, we present the recommendations of the Panel, based on this literature review. Results: Overall, the risk of SPMs in MM is low, multifactorial, and partially related to the length of patients' survival and MM intrinsic susceptibility. Studies suggest a significantly increased incidence of SPMs when lenalidomide is administered either following, or concurrently with, oral melphalan. Increased SPM incidence has also been reported with lenalidomide maintenance following high-dose melphalan, albeit to a lesser degree. In both cases, the risk of death from MM was significantly higher than the risk of death from SPMs, with lenalidomide possibly providing a survival benefit. No increase in SPM incidence was reported with lenalidomide plus dexamethasone (without melphalan), or with bortezomib plus oral melphalan, dexamethasone, or thalidomide. Conclusion: In general, the risk of SPMs should not alter the current therapeutic decision-making process in MM. However, regimens such as lenalidomide plus dexamethasone should be preferred to prolonged exposure to lenalidomide plus oral melphalan. SPM risk should be carefully discussed with the patient in the context of benefits and risks of different treatment options.

Racial differences in primary cytogenetic abnormalities in multiple myeloma: a multi-center study.

We examined four clinically assessed cytogenetic subtypes (t(11;14), t(4;14), monosomy 13/del13q and monosomy 17/del17p in 292 black patients with newly diagnosed multiple myeloma (MM) from four medical centers, who had fluorescent in situ hybridization testing results available in their medical records. We then compared the prevalence of these abnormalities with a previously characterized Mayo Clinic cohort of 471 patients with MM. We found a significant difference in the prevalence of the t(11;14) immunoglobulin heavy chain (IgH) translocation between blacks and whites, 6.5% versus 17.6%, respectively, P<0.0001. Blacks also had lower rates of the t(4;14) IgH translocation, (5.5% versus 10%); monosomy 13/del13q (29.1 versus 49.3%); and monosomy 17/del17p (7.9% versus 13%). Consequently, 63.4% of blacks versus 34.6% of whites did not have any of the four abnormalities that we studied, P<0.001. As almost all MM is associated with either an IgH translocation or trisomies, we hypothesize that MM in blacks is associated with either excess prevalence of either the trisomic (hyperdiploid) form of MM or an IgH translocation besides t(11;14) or t(4;14). We conclude that there are significant differences in the cytogenetic subtypes of MM that occur in blacks and whites.

A Tumor Growth Inhibition Model Based on M-Protein Levels in Subjects With Relapsed/Refractory Multiple Myeloma Following Single-Agent Carfilzomib Use.

Change in tumor size estimated using longitudinal tumor growth inhibition (TGI) modeling is an early predictive biomarker of clinical outcomes for multiple cancer types. We present the application of TGI modeling for subjects with multiple myeloma (MM). Longitudinal time course changes in M-protein data from relapsed and/or refractory MM subjects who received single-agent carfilzomib in phase II studies (n = 456) were fit to a TGI model. The tumor growth rate estimate was similar to that of other anti-myeloma agents, indicating that the model is robust and treatment-independent. An overall survival model was subsequently developed, which showed that early change in tumor size (ECTS) at week 4, Eastern Cooperative Oncology Group performance status (ECOG PS), hemoglobin, sex, percent bone marrow cell involvement, and number of prior regimens were significant independent predictors for overall survival (P < 0.001). ECTS based on M-protein modeling could be an early biomarker for survival in MM following exposure to single-agent carfilzomib.

Carfilzomib in multiple myeloma patients with renal impairment: pharmacokinetics and safety.

This phase 2 study assessed the safety, pharmacokinetics, pharmacodynamics and efficacy of carfilzomib, a selective proteasome inhibitor, in patients with multiple myeloma and varying degrees of renal impairment, including patients on chronic hemodialysis. Patients were grouped by creatinine clearance: >80 ml/min, 50-80 ml/min, 30-49 ml/min, <30 ml/min and chronic hemodialysis. Carfilzomib was administered on days 1, 2, 8, 9, 15 and 16 in 28-day cycles: 15 mg/m(2) (Cycle 1), 20 mg/m(2) (Cycle 2) and 27 mg/m(2) (Cycles 3+). There were no differences in carfilzomib clearance or exposure among patients with normal renal function and any group with renal impairment. Grade 3/4 adverse events (AEs) included anemia (28.0%), thrombocytopenia (20.0%), lymphopenia (18.0%) and fatigue (14.0%). AEs were similar among groups. At 15 mg/m(2), proteasome inhibition up to 85% was observed and did not differ among groups. Although nearly 50% of patients were refractory to both bortezomib and lenalidomide, end of study partial response or better (overall response rate) was 25.5% with 7.9 months median duration of response. In conclusion, the pharmacokinetics and safety of carfilzomib were not influenced by the degree of baseline renal impairment, including in patients on dialysis, and carfilzomib was well tolerated and demonstrated promising efficacy.

Plasma cell leukemia: consensus statement on diagnostic requirements, response criteria and treatment recommendations by the International Myeloma Working Group.

Plasma cell leukemia (PCL) is a rare and aggressive variant of myeloma characterized by the presence of circulating plasma cells. It is classified as either primary PCL occurring at diagnosis or as secondary PCL in patients with relapsed/refractory myeloma. Primary PCL is a distinct clinic-pathological entity with different cytogenetic and molecular findings. The clinical course is aggressive with short remissions and survival duration. The diagnosis is based upon the percentage (≥ 20%) and absolute number (≥ 2 × 10(9)/l) of plasma cells in the peripheral blood. It is proposed that the thresholds for diagnosis be re-examined and consensus recommendations are made for diagnosis, as well as, response and progression criteria. Induction therapy needs to begin promptly and have high clinical activity leading to rapid disease control in an effort to minimize the risk of early death. Intensive chemotherapy regimens and bortezomib-based regimens are recommended followed by high-dose therapy with autologous stem cell transplantation if feasible. Allogeneic transplantation can be considered in younger patients. Prospective multicenter studies are required to provide revised definitions and better understanding of the pathogenesis of PCL.

Clonal cytogenetic changes and myeloma relapse after reduced intensity conditioning allogeneic transplantation.

To identify a correlation between metaphase cytogenetics and relapse after reduced intensity conditioning (RIC) allotransplant for patients with multiple myeloma, data on 60 patients (median age 52) who received grafts from a sibling (n = 49) or unrelated donor (n = 11) were analyzed. Fifty-three patients (88%) showed chromosomal abnormalities (CA) before the allotransplant, including 42 with abnormalities involving 13q (CA13). Twenty-two patients (41%) relapsed post-allotransplant at a median of 165 days. Of these, 11 patients showed abnormal cytogenetics at the time of post-allotransplant relapse at a median of 167 days. Of 54 patients who developed graft-versus-host disease, relapse occurred in 19 of 48 patients (43%) with CA present before RCI allotransplant, versus 1 of 6 without CA (17%) (P = 0.06). Loss of CA before RIC allotransplant and disease status > PR after RIC allotransplant were significantly associated with a lower risk of post-allotransplant relapse with cytogenetic abnormalities; 5.2 vs 36%, and 18 vs 53%, (both P < 0.05), respectively. The current data suggests that myeloma associated with persistent clonal cytogenetic abnormalities is an entity which most likely escapes the effects of a graft versus myeloma activity, maybe because of acquisition of resistance to immunologic manipulations.

Autologous stem cell transplantation followed by consolidation chemotherapy for relapsed or refractory Hodgkin's lymphoma.

Relapse remains a major cause of treatment failure after autotransplantation (auto-PBSCT) for Hodgkin's disease (HD). The administration of non-crossresistant therapies during the post-transplant period may delay or prevent relapse. We prospectively studied the role of consolidation chemotherapy (CC) after auto-PBSCT in 37 patients with relapsed or refractory HD. Patients received high-dose gemcitabine-BCNU-melphalan and auto-PBSCT followed by involved-field radiation and up to four cycles of the DCEP-G regimen, which consisted of dexamethasone, cyclophosphamide, etoposide, cisplatin, gemcitabine given at 3 and 9 months post transplant alternating with a second regimen (DPP) of dexamethasone, cisplatin, paclitaxel at 6 and 12 months post transplant. The probabilities of event-free survival (EFS) and overall survival (OS) at 2.5 years were 59% (95% CI=42-76%) and 86% (95% CI=71-99%), respectively. In all, 17 patients received 54 courses of CC and 15 were surviving event free (2.5 years, EFS=87%). There were no treatment-related deaths during or after the CC phase. Post-transplant CC is feasible and well tolerated. The impact of this approach on EFS should be evaluated in a larger, randomized study.

Amifostine and autologous hematopoietic stem cell support of escalating-dose melphalan: a phase I study.

This study was conducted to define a new maximum tolerated dose and the dose-limiting toxicity (DLT) of melphalan and autologous hematopoietic stem cell transplantation (AHSCT) when used with the cytoprotective agent amifostine. Fifty-eight patients with various types of malignancy who were ineligible for higher-priority AHSCT protocols were entered on a phase I study of escalating doses of melphalan beginning at 220 mg/m(2) and advancing by 20 mg/m(2) increments in planned cohorts of 4 to 8 patients until severe regimen-related toxicity (RRT) was encountered. In all patients, amifostine 740 mg/m(2) was given on 2 occasions before the first melphalan dose (ie, 24 hours before and again 15 minutes before). AHSCT was given 24 hours after the first melphalan dose. Melphalan was given in doses up to and including 300 mg/m(2). Hematologic depression was profound, although it was rapidly and equally reversible at all melphalan doses. Although mucosal RRT was substantial, it was not the DLT, and some patients given the highest melphalan doses (ie, 300 mg/m(2)) did not develop mucosal RRT. The DLT was not clearly defined. Cardiac toxicity in the form of atrial fibrillation occurred in 3 of 36 patients treated with melphalan doses >/=280 mg/m(2) and was deemed fatal in 1 patient given melphalan 300 mg/m(2). (Another patient with a known cardiomyopathy was given melphalan 220 mg/m(2) and died as a result of heart failure but did not have atrial fibrillation.) Another patient given melphalan 300 mg/m(2) died of hepatic necrosis. The maximum tolerated dose of melphalan in this setting was thus considered to be 280 mg/m(2), and 27 patients were given this dose without severe RRT. Moreover, 38 patients were evaluable for delayed toxicity related to RRT; none was noted. Tumor responses have been noted at all melphalan doses and in all diagnostic groups, and 21 patients are alive at median day +1121 (range, day +136 to day +1923), including 16 without evidence of disease progression at median day +1075 (range, day +509 to day +1638). Amifostine and AHSCT permit the safe use of melphalan 280 mg/m(2), an apparent increase over the dose of melphalan that can be safely administered with AHSCT but without amifostine. Further studies are needed not only to confirm these findings, but also to define the antitumor efficacy of this regimen. Finally, it may be possible to evaluate additional methods of further dose escalation of melphalan in this setting.

Activity of single-agent melphalan 220-300 mg/m2 with amifostine cytoprotection and autologous hematopoietic stem cell support in non-Hodgkin and Hodgkin lymphoma.

High-dose chemotherapy using melphalan (HDMEL) is an important component of many conditioning regimens that are given before autologous hematopoietic stem cell transplantation (AHSCT). In contrast to the situation in myeloma, and to a lesser degree acute leukemia, only a very limited published experience exists with the use of HDMEL conditioning as a single agent in doses requiring AHSCT for lymphoma, both Hodgkin lymphoma (HL) and especially non-Hodgkin lymphoma (NHL). Thus, we report results of treating 26 lymphoma patients (22 with NHL and four with HL) with HDMEL 220-300 mg/m(2) plus amifostine (AF) cytoprotection and AHSCT as part of a phase I-II trial. Median age was 51 years (range 24-62 years); NHL histology was varied, but was aggressive (including transformed from indolent) in 19 patients, indolent in two patients and mantle cell in one. All 26 patients had been extensively treated; 11 were refractory to the immediate prior therapy on protocol entry and two had undergone prior AHSCT. All were deemed ineligible for other, 'first-line' AHSCT regimens. Of these 26 patients, 22 survived to initial tumor evaluation on D +100. At this time, 13 were in complete remission, including four patients who were in second CR before HDMEL+AF+AHSCT. Responses occurred at all HDMEL doses. Currently, seven patients are alive, including five without progression, with a median follow-up in these latter patients of D +1163 (range D +824 to D +1630); one of these patients had a nonmyeloablative allograft as consolidation on D +106. Conversely, 14 patients relapsed or progressed, including five who had previously achieved CR with the AHSCT procedure. Two patients, both with HL, remain alive after progression; one is in CR following salvage radiotherapy. Six patients died due to nonrelapse causes, including two NHL patients who died while in CR. We conclude that HDMEL+AF+AHSCT has significant single-agent activity in relapsed or refractory NHL and HL. This experience may be used as a starting point for subsequent dose escalation of HDMEL (probably with AF) in established combination regimens.

Molecular remission of CML after autotransplantation followed by adoptive transfer of costimulated autologous T cells.

Four patients with chronic myelogenous leukemia (CML) that was refractory to interferon alpha (two patients) or imatinib mesylate (two patients), and who lacked donors for allogeneic stem cell transplantation, received autotransplants followed by infusions of ex vivo costimulated autologous T cells. At day +30 (about 14 days after T-cell infusion), the mean CD4+ cell count was 481 cells/microl (range 270-834) and the mean CD8+ count was 516 cells/microl (range 173-1261). One patient had a relative lymphocytosis at 3.5 months after T-cell infusion, with CD4 and CD8 levels of 750 and 1985 cells/microl, respectively. All the four patients had complete cytogenetic remissions early after transplantation, three of whom also became PCR negative for the bcr/abl fusion mRNA. One patient, who had experienced progressive CML while on interferon alpha therapy, became PCR- post transplant, and remained in a molecular CR at 3.0 years of follow-up. All the four patients survived at 6, 9, 40, and 44 months post transplant; the patient who remained PCR+ had a cytogenetic and hematologic relapse of CML, but entered a molecular remission on imatinib. Autotransplantation followed by costimulated autologous T cells is feasible for patients with chronic phase CML, who lack allogeneic donors and can be associated with molecular remissions.

Transplantation as salvage therapy for high-risk patients with myeloma in relapse.

Patients with myeloma relapsing after tandem transplant have a poor survival and treatment options are limited. The role of additional salvage transplant procedures for these patients is unknown. To evaluate the benefit and identify prognostic factors, the outcome of 76 consecutive patients with recurrent myeloma after tandem transplant receiving salvage transplants (ST) was analyzed. Prior to ST, 23 patients (30%) had shown chemosensitive response to preceding salvage chemotherapy: two complete remissions (CR); eight near CRs (nCR: only immunofixation positive); 13 partial remissions (PR >or=75% reduction in M protein). Fifty received an autologous transplant, 22 a sibling-matched allogeneic transplant, and four a matched-unrelated allogeneic transplant. Overall response after ST was 59%: eight CRs (11%); 14 nCRs (18%); 23 PRs (30%). Overall survival (OS) at 2 years was 19%; 2 year event-free survival rate (EFS) 7%. On univariate analysis for survival, only pre-transplant chemosensitive relapse (P < 0.05), serum albumin >3 g/dl (P = 0.001), normal LDH (P = 0.04), and long interval between the second transplant and relapse/progression were significant beneficial factors. In a Cox proportional hazard model, chemosensitive relapse, and albumin >3 g/dl were significant for better OS: hazard ratio (HR) 1.4, 1.7, respectively, while normal LDH, and absence of CA13 were significant for better EFS: HR 1.8, 1.7, respectively. Patients with albumin >3 g/dl who had chemosensitive disease before ST (n = 16) had a median survival of 16 months, compared to 7 months (n = 34) and 2 months (n = 26) for patients with only one (n = 34) or no favorable prognostic factors (n = 28), respectively (P < 0.001). Their survival at 2 years post-ST was 43%, 17% and 11%, respectively. Our study suggests further transplantation should only be considered in the setting of a clinical trial in patients with favorable prognostic factors.

Clinical activity of arsenic trioxide for the treatment of multiple myeloma.

Arsenic has been used since ancient times as a therapeutic agent. However, until recently its use in modern medicine has been restricted to the treatment of a limited number of parasitic infections. In the early 1990s, reports from China described impressive results with arsenic trioxide in patients with de novo, relapsed, and refractory acute promyelocytic leukemia (APL). Other investigators subsequently confirmed these results leading to approval of its use for relapsed or refractory APL in the United States. Investigations of this agent have demonstrated that its efficacy in APL and preclinical tumor models is dependent upon a number of mechanisms, including induction of apoptosis, effects on cellular differentiation, cell cycling, and tumor angiogenesis. Subsequent preclinical studies showed significant activity of arsenic trioxide in multiple myeloma (MM). Based on this, in a phase II trial, we have evaluated the activity of arsenic trioxide in 14 patients with relapsed MM, refractory to conventional salvage therapy. With the dose and schedule used, treatment with arsenic trioxide produced responses in three patients and prolonged stable disease in a fourth patient, with the longest response lasting 6 weeks. Although treatment was reasonably well tolerated, in these patients with extensive prior therapy, 11 developed cytopenia, five associated with infectious complications and three developed deep vein thromboses. The results of this small trial support further investigation of this novel drug for the treatment of patients with relapsed or refractory MM.

Separating graft-versus-tumor from graft-versus-host reactions.

The full therapeutic potential of allogeneic stem cell transplantation, through its immunologically mediated graft-versus-tumor effect, in patients with hematologic malignancies is greatly compromised by the occurrence of graft-versus-host disease. Unfortunately, the use of non-selective immunosuppressive agents to reduce the incidence and severity of graft-versus-host disease is associated with severe immune compromise of the host and most likely a greater relapse risk of the underlying malignancy. Many attempts have been made to clinically separate these two effects. A critical overview of the published experience is the focus of this report. As the effector cells responsible for the two reactions are largely unknown, the limited success of the various approaches used is not surprising. A more thorough understanding of the antigenic stimuli involved in the initiation of the two reactions and of the molecular pathways through which the cytotoxic effects of T-cells are mediated is essential for abrogating graft-versus-host disease while preserving the graft-versus-tumor effect.

Activated protein C resistance in the absence of factor V Leiden mutation is a common finding in multiple myeloma and is associated with an increased risk of thrombotic complications.

Thromboembolism is not uncommon in multiple myeloma (MM) patients on treatment, but its pathogenesis remains poorly understood. We report the results of a prospective randomized trial of 62 newly diagnosed MM patients tested at baseline for hypercoagulability and treated with intensive chemotherapy with or without thalidomide in a randomized fashion. During the induction phase, 12 patients (19%) developed evidence of deep venous thrombosis (DVT), which was significantly more common in the thalidomide arm (36%) than in the control group (3%) (P = 0.001). Fourteen patients (23%) were found to have a baseline-reduced response to activated protein C (APC) in the absence of factor V Leiden mutation. Using a Kaplan-Meier analysis, a significantly higher proportion of patients with APC resistance developed DVT (5/14 versus 7/38; P = 0.04) irrespective of thalidomide administration. The risk of DVT was highest (50%) in patients with APC resistance on thalidomide. None of the patients with normal APC response and not receiving thalidomide developed DVT. In conclusion, in this series, acquired APC resistance was present in almost one-quarter of newly diagnosed myeloma patients and significantly increased the risk of DVT.

Autotransplantation for advanced lymphoma and Hodgkin's disease followed by post-transplant rituxan/GM-CSF or radiotherapy and consolidation chemotherapy.

Disease relapse occurs in 50% or more of patients who are autografted for relapsed or refractory lymphoma (NHL) or Hodgkin's disease (HD). The administration of non-cross-resistant therapies during the post-transplant phase could possibly control residual disease and delay or prevent its progression. To test this approach, 55 patients with relapsed/refractory or high-risk NHL or relapsed/refractory HD were enrolled in the following protocol: stem cell mobilization: cyclophosphamide (4.5 g/m(2)) + etoposide (2.0 g/m(2)) followed by GM-CSF or G-CSF; high-dose therapy: gemcitabine (1.0 g/m(2)) on day -5, BCNU (300 mg/m(2)) + gemcitabine (1.0 g/m(2)) on day -2, melphalan (140 mg/m(2)) on day -1, blood stem cell infusion on day 0; post-transplant immunotherapy (B cell NHL): rituxan (375 mg/m(2)) weekly for 4 weeks + GM-CSF (250 microg thrice weekly) (weeks 4-8); post-transplant involved-field radiotherapy (HD): 30-40 Gy to pre-transplant areas of disease (weeks 4-8); post-transplant consolidation chemotherapy (all patients): dexamethasone (40 mg daily)/cyclophosphamide (300 mg/m(2)/day)/etoposide (30 mg/m(2)/day)/cisplatin (15 mg/m(2)/day) by continuous intravenous infusion for 4 days + gemcitabine (1.0 g/m(2), day 3) (months 3 + 9) alternating with dexamethasone/paclitaxel (135 mg/m(2))/cisplatin (75 mg/m(2)) (months 6 + 12). Of the 33 patients with B cell lymphoma, 14 had primary refractory disease (42%), 12 had relapsed disease (36%) and seven had high-risk disease in first CR (21%). For the entire group, the 2-year Kaplan-Meier event-free survival (EFS) and overall survival (OS) were 30% and 35%, respectively, while six of 33 patients (18%) died before day 100 from transplant-related complications. The rituxan/GM-CSF phase was well-tolerated by the 26 patients who were treated and led to radiographic responses in seven patients; an eighth patient with a blastic variant of mantle-cell lymphoma had clearance of marrow involvement after rituxan/GM-CSF. Of the 22 patients with relapsed/refractory HD (21 patients) or high-risk T cell lymphoblastic lymphoma (one patient), the 2-year Kaplan-Meier EFS and OS were 70% and 85%, respectively, while two of 22 patients (9%) died before day 100 from transplant-related complications. Eight patients received involved field radiation and seven had radiographic responses within the treatment fields. A total of 72 courses of post-transplant consolidation chemotherapy were administered to 26 of the 55 total patients. Transient grade 3-4 myelosuppression was common and one patient died from neutropenic sepsis, but no patients required an infusion of backup stem cells. After adjustment for known prognostic factors, the EFS for the cohort of HD patients was significantly better than the EFS for an historical cohort of HD patients autografted after BEAC (BCNU/etoposide/cytarabine/cyclophosphamide) without consolidation chemotherapy (P = 0.015). In conclusion, post-transplant consolidation therapy is feasible and well-tolerated for patients autografted for aggressive NHL and HD and may be associated with improved progression-free survival particularly for patients with HD.

Results of autologous stem cell transplant in multiple myeloma patients with renal failure.

Data are presented on 81 multiple myeloma (MM) patients with renal failure (creatinine > 176.8 micromol/l) at the time of autologous stem cell transplantation (auto-SCT), including 38 patients on dialysis. The median age was 53 years (range: 29-69) and 26% had received more than 12 months of prior chemotherapy. CD34+ cells were mobilized with granulocyte colony-stimulating factor (G-CSF) alone (n = 51) or chemotherapy plus G-CSF (n = 27), yielding medians of 10 and 16 x 106 CD34+ cells/kg respectively (P = 0.003). Sixty patients (27 on dialysis) received melphalan 200 mg/m2 (MEL-200). Because of excessive toxicity, the subsequent 21 patients (11 on dialysis) received MEL 140 mg/m2 (MEL-140). Thirty-one patients (38%) completed tandem auto-SCT, including 11 on dialysis. Treatment-related mortality (TRM) was 6% and 13% after the first and second auto-SCT. Median times to absolute neutrophil count (ANC) > 0.5 x 109/l and to platelets > 50 x 109/l were 11 and 41 d respectively. Non-haematological toxicities included mucositis, pneumonitis, dysrhythmias and encephalopathy. At a median follow up of 31 months, 30 patients have died. Complete remission (CR) was achieved in 21 patients (26%) after first SCT and 31 patients (38%) after tandem SCT. Two patients discontinued dialysis after SCT. Median durations of complete remission (CR) and overall survival (OS) have not been reached; probabilities of event-free survival (EFS) and OS at 3 years were 48% and 55% respectively. Dialysis dependence and MEL dose did not affect EFS or OS. Sensitive disease prior to SCT, normal albumin level and younger age were independent prognostic factors for better OS. In conclusion, renal failure had no impact on the quality of stem cell collections and did not affect engraftment. MEL-140 had an acceptable toxicity and appeared equally effective as MEL-200. In the setting of renal failure, the role of auto-SCT early in the disease course and benefits of tandem SCT require further evaluation.