Intensive Treatment for Multiple Myeloma: Where Do We Stand?

It is now more than 10 years since high-dose therapy for multiple myeloma (MM) was introduced by 60). The Royal Marsden group was the first to establish a dose–response effect for alkylating agents in MM. Using high-dose melphalan (HDM), 140 mg/m2, it was possible to overcome resistance to lower doses of melphalan and to induce complete remissions in approximately 35% of patients (66). The encouraging results of HDM prompted additional studies which concentrated on high-dose chemo/radiotherapy followed by (autologous) stem cell transplantation. Initial studies were performed in refractory patients who were treated with 100–140 mg/m2 melphalan supported by reinfusion of autologous bone marrow (10, 12). In subsequent studies, myeloablative regimens followed by transplantation with autologous bone marrow or peripheral blood stem cells (PBSC) were explored (46; 7). The use of mobilized PBSC, usually collected after chemotherapy together with haemopoietic growth factors, has improved the feasibility of autologous transplantation (39; 24; 31). So far, thousands of myeloma patients have been autotransplanted worldwide. The majority of published studies concerns phase I and phase II trials and include a heterogenous patient population. In these studies a variety of strategies have been used such as repeated dose-intensification, regimens combined with double transplants, purging of autografts, different sources of stem cells and alternative myeloablative regimens (1; 5; 13; 17; 35; 42; 53; 58). It is difficult to draw definite conclusions from these studies because of the different approaches and the possibility that a selection bias of patients may have determined the outcome. Although it has been shown that intensive therapy improves the overall and complete response rates, a survival benefit has not been shown. There has been only one randomized study to demonstrate that autologous bone marrow transplantation (ABMT) improves the outcome of newly diagnosed MM patients when compared with conventional chemotherapy (6). Therefore it seems premature to conclude that intensive therapy is the standard approach for the younger MM patient. Apart from summarizing what has been achieved so far with high-dose therapy in MM, the aim of this review is to discuss the rationale and status of ongoing phase II and III studies from which a more precisely defined role of autologous and allogeneic stem cell transplantation in myeloma may be expected. New promising modalities for treatment such as post transplant immunotherapy and non-myeloablative allogeneic stem cell transplantation will also be discussed. In patients with chemosensitive disease who were transplanted after a short period of induction therapy, remarkably high complete remission (CR) rates and prolonged event-free survival (EFS) and overall survival (OS) rates have been reported (Table I) (29; 17; 5; 42; 45; 52). Most impressive results were achieved by the Arkansas group with their ‘total therapy program’ which consisted of non-cross-resistant induction regimes followed by a double autotransplantation procedure (15). In addition, interferon (IFN) maintenance was given until relapse. The updated results of 231 newly diagnosed patients using an intention-to-treat analysis showed that the CR/PR rate was 26%/75% after the transplantation and 38%/81% after the second transplantation, respectively (11). The median event-free survival and overall survival were 42 and 65 months, respectively, with a median time to progression of 54 months. Favourable prognostic factors were the absence of ‘unfavourable’ karyotypes (any translocation or abnormalities of chromosomes 11q and 13), low levels of β2 microglobulin (B2M), less than 1 year of prior therapy, chemosensitive disease and the achievement of CR within 6 months after induction (45; 69). In most other published phase II studies (Table I) only patients who had responded to previous conventional chemotherapy were allowed to proceed to high-dose treatment with autologous stem cell transplantation. The Royal Marsden study with melphalan 200 mg/m2 and autologous bone marrow transplantation (ABMT) included a subset of 53 responsive patients out of a group of 105 patients (28); 5) gave melphalan 140 mg/m2 plus total body irradiation (TBI) and monoclonal-antibody purged bone marrow in 52 responsive patients. In the Dutch Hovon Study intensification was given to 50 patients responding to induction therapy with vincristine, adriamycin, dexamethasone (VAD) and intermediate-dose melphalan (70 mg/m2 IDM), originally given to 76 patients (52). 42) summarized the results of 133 patients from 18 French centres and reported a median overall survival of 46 months. Notably the overall survival was 54 months in 103 patients responding to primary treatment as compared to 30 months in 30 non-responders. It has become clear that favourable prognostic factors for conventional treatment, e.g. a low B2M and responsive disease, are also significant for the outcome of autotransplantation. Thus, although the response and survival in these phase II studies (see Table I) appears to be better than historical control groups of patients who were treated conventionally, one should still be careful to draw definite conclusions from the so far published non-randomized trials. This is illustrated by historical comparisons performed by 23) and 3), who showed that patients who did not receive high-dose chemotherapy but were eligible (younger age, chemosensitive disease), had the same prolonged survival (>5 years) as patients who received an autologous transplant. So far the results of only one prospective trial, the IMF (Intergroup Myeloma Française), has been published which shows a significant improvement by intensive therapy. In this study conventional therapy (BVAP/VMCP) was compared to ABMT which was performed after four to six cycles of BVAP/VMCP. α-Interferon maintenance was administered in both arms until relapse. Points of criticism of this trial are the relatively small number of patients and the poor response rate of the conventional arm which compared unfavourably with other trials of conventional chemotherapy (41). Does more tumour reduction translate into a longer event-free and overall survival? Several studies including the large analysis performed by the European Group for Blood and Marrow Transplantation (EBMT) on prognostic factors have shown that the achievement of CR post transplantation had a significant positive effect on event-free survival and overall survival (21; 42; 6). This indicates that the ultimate aim of intensive treatment should be to attain more complete remissions at the molecular level. Such a goal may be achieved by further dose escalation by using double transplants. 44) and 20) performed, for the first time, double transplants in myeloma. However, because of excessive toxicity, only 36% of 97 patients in the French study proceeded to the second transplant. The CR rate after the second transplant was 34%. 20) observed a CR rate of 40% after the second transplant in a small group (n = 13) of patients. The largest series of double transplants has been performed by the group led by Barlogie. In previously untreated patients the CR rate increased from 26% after the first transplant to 38% after the second. In addition to cytogenetics (absence of abnormalities of chromosomes 11q and/or any translocation), low C-reactive protein (CRP), high haemoglobin, and CR within 6 months after induction were favourable prognostic features for event-free survival and overall survival (45). This study showed that more intensive treatment results in a better tumour reduction and a higher CR rate. It will be interesting to measure longitudinally the tumour load of minimal residual disease more accurately by quantifying the malignant cells of patients with a very good partial remission or complete remission. This can be done by the recently described quantitative PCR methods that are based on amplification of the unique heavy chain sequence of the malignant clone using allele-specific oligonucleotide (IgH ASO-PCR) (37). The sensitivity of this method is 10−4 cells. Using this technique the relative contribution of each treatment phase to tumour reduction can be precisely assessed and whether molecular remissions are achieved after double transplants. It has been shown that molecular CRs are frequently achieved after allogeneic BMT but rarely after single autologous transplants (26; 55). Definite evidence, however, that double transplants are superior to other treatment modalities including single transplants, can come only from randomized studies. Such a study was initiated by the IMF in 1994 and included 400 newly diagnosed patients <60 years. Patients were randomized to receive a single autologous transplant with HDM 140 mg/m2 and total body irradiation (TBI) or a double autologous transplant: the first cycle consisted of HDM 140 mg/m2 and the second contained HDM 140 mg/m2 and total body irradiation (TBI). An interim analysis of 200 patients showed that double transplants did not significantly improve the response rate, the event-free survival or overall survival (8). However, a longer follow-up and analysis of all 400 patients will be necessary before it can be definitely concluded whether ‘more is better’. Peripheral blood stem cells are now mostly used for autologous stem cell support. Important advantages of PBSCT over ABMT are the more rapid haematological recovery and a lower probability of tumour cell contamination of the graft (73). A prospective randomized trial of the IMF confirmed that the duration of aplasia is shorter and that there is a reduction of the transfusion requirements after PBSC. However, the relapse rates after PBSCT and ABMT were not different (43). Numerous studies have shown that peripheral blood stem cells are contaminated with myeloma cells and, theoretically, these tumour cells may contribute to the occurrence of relapses after autologous stem cell transplantation (38; 64). CD34 selection of PBSC significantly reduces the number of contaminating MM cells, but sensitive methods such as PCR analysis with patient-specific primers for the CDR III region still show persistence of minimal residual disease in the grafts, even if positive CD34 selection is combined with B-cell depletion. 70) combined centrifugal elutriation and chemical lysis to deplete PBSC collections of monocytes, granulocytes, platelets and erythrocytes with high-speed cell sorting of CD34+ Thy+ Lin− progenitor cells. The grafts from three patients which were tested with CDR III specific primers contained no detectable myeloma cells. However, patients who were transplanted with these highly purified progenitor cells showed substantially delayed engraftment post-transplantation. In addition to the delay of neutrophil and platelet recovery following CD34 selected autologous stem cell transplantation the immunorecovery may also be affected, especially of the CD4 cells, and this has been associated with an increase of opportunistic infections. As the procedure for CD34+ selection is expensive and time-consuming, the potential benefits should be weighed against the disadvantages of delayed haemopoietic and immunological reconstruction. Evidence that cells from the graft contribute to relapse will require large randomized clinical trials with a long follow-up. Such studies are being performed in France and in the U.S.A. So far, the interim analysis of the French IMF study does not show any benefit from CD34 selection as compared to unpurged PBSCT (M. Attal, personal communication). Retroviral gene marking of autografts might also help to identify the source of relapse. Low transduction efficiency of myeloma cells with the current gene marking protocols may, however, hamper the interpretation of such studies. Should newly diagnosed myeloma patients receive high-dose therapy as part of their initial treatment or only after resistance to, or relapse following, conventional therapy? As mentioned above, two historical comparisons (3; 23) showed that patients with chemosensitive disease may not benefit from myeloablative therapy administered early in the course of the disease, in contrast to patients with secondary refractory disease. In agreement with these findings is the final analysis of the French multicentre study comparing ‘upfront high-dose chemotherapy with conventional chemotherapy followed by high-dose therapy in case of primary refractory disease or relapse after response (36). In both arms the median overall survival was exactly the same and exceeded 5 years. 36) concluded that the only reason for the administration of high-dose therapy early during the disease was based on the shorter period of chemotherapy, which may have a clinical benefit. The current Dutch HOVON multicentre trial has a comparable design and includes a quality-of-life evaluation. In this protocol, patients receive a maximal ‘conventional’ therapy consisting of VAD three times, followed by cyclophamide and PBSC collection, and two courses of intermediate-dose melphalan (IDM), 70 mg/m2, followed by HDT with PBSCT immediately or after relapse. Quality-of-life and costs aspects, measured in both arms, may further help to determine the role and optimal timing of HDT. Various conditioning regimens have been used in heterogenous groups of myeloma patients (2; 34; 56). It is difficult to conclude which scheme is optimal in myeloma. In a recent multivariate analysis of the EBMT registry on prognostic factors a non-TBI preparative regimen was independently associated with a better survival (21). In a historical comparison of patients receiving double transplants it was demonstrated that melphalan 200 mg/m2 was superior to lower doses of melphalan combined with TBI (32). Another potential advantage of melphalan 200 mg/m2 is that it induces less immunosuppression than TBI-including regimens. A more definitive evaluation of toxicity and efficacy of melphalan alone versus a TBI regimen comes from a randomized IMF study in which patients between 60 and 65 years receive either melphalan 200 mg/m2 or melphalan 140 mg/m2 plus TBI. The number of treatment-related deaths appears higher in patients who received melphalan plus TBI (M. Attal, personal communication). This is an important issue, as all patients eventually relapse after a previous autologous transplant. The first indication that patients who relapse after intensive treatment may respond again to (high-dose) therapy came from studies by the Royal Marsden group. Twenty-nine patients who had relapsed after single HDM 140 mg/m2 were treated with reinduction therapy followed by melphalan 200 mg/m2 plus ABMT. The response rate was 93% and response duration was 17 months (57). the Arkansas group evaluated the outcome of 196 patients with a relapse after a single or a double transplant. Patients received standard-dose therapy or a further transplant. 10% of patients died within 2 months, 8% achieved a CR, and the post salvage durations of event-free survival/overall survival were 8 and 14 months, respectively. Patients with a low B2M responded best with an ultimate CR rate of 32%. Multivariate analysis showed that this category of patients had a significant better outcome after a repeat transplant in terms of prolonged event-free survival/overall survival than after standard-dose therapy (64+/64+v 9/11 months respectively). In patients who relapsed after a double transplantation, the time to relapse (>1 year) was the predominant prognostic factor rather than further intensification. From these data it can be concluded that patients with a relapse after prior transplantation may benefit from continuous treatment (14; 71). Patients with a low B2M, one prior transplant and late relapse are probably the best candidates for a second transplantation. The implication of this is that sufficient PBSCs should be collected before the first transplant to perform a second (salvage) transplant procedure. It cannot be excluded, however, that selection factors leading to treatment decisions explain the favourable outcome after repeat transplant when compared to standard-dose therapy. Randomized trials in which different treatment options are tested (e.g. standard dose versus intensive treatment) in patients with a relapse after high-dose therapy would be more decisive, but will probably be difficult to organize. Patients with resistant disease or with a very early relapse might be candidates for novel strategies such as non-myeloablative chemotherapy followed by HLA-identical donor cell transplantation to induce a graft-versus-myeloma effect. The role of interferon maintenance after standard-dose chemotherapy for MM is the subject of a continuous debate, because the results of randomized trials are conflicting. Several groups found a favourable effect of interferon with respect to survival and/or response duration, whereas others found no beneficial effect of IFN (54; 47). There are few published trials concerning the efficacy of IFN maintenance treatment after intensive therapy. In a French registry study on autologous stem cell transplantation (42) no significant impact on event-free survival and overall survival was observed, whereas in the registry study of the EBMT, IFN maintenance seems to improve the overall survival (21). So far, only one randomized trial with IFN maintenance after autologous stem cell transplantation has been published (30). Evaluation at a median follow-up of 52 months after HDM showed a significantly better outcome for the IFN group: the median event-free survival in the 42 patients randomized to IFN was 46 months versus 27 months in the control group (P < 0.0025). The overall survival was significantly better in the IFN group. At a median time of 77 months, however, the survival advantage had disappeared. For those patients who achieved a CR, a trend towards a better event-free survival was noted, whereas overall survival was not significantly different. It may be concluded that, to conventional treatment, there is no benefit for IFN maintenance after HDT. randomized studies including groups of patients studies will be to whether IFN a in the of minimal residual disease of multiple myeloma patients. The of an allogeneic bone marrow transplantation in multiple myeloma is The median survival after allogeneic BMT in different from 18 to months, because the is significantly higher than with autologous stem cell transplantation (Table A survival advantage for patients receiving an allogeneic transplant compared to patients with the same receiving an autologous transplant has not been shown. In a analysis of patients reported to the European Marrow Transplantation (EBMT) the overall survival of patients receiving an autologous transplant was significantly better than that of allogeneic BMT patients 18 This survival however, was only in but not in This analysis did not identify a of including patients with chemosensitive disease, who might benefit from allogeneic for patients at 1 year was a trend found for prolonged overall survival and a significantly prolonged event-free survival for allogeneic The reason for the outcome in allogeneic BMT is probably to a higher treatment-related which is not for by the higher CR rate and the lower relapse rate. of allogeneic BMT reported by the EBMT, by the by the and the Arkansas groups from 35% to (Table An important factor for this is the of a high of patients and refractory patients. results from ongoing studies that lower was observed if allogeneic BMT was performed during the initial of disease or after From these results it is that the of allogeneic BMT in MM should on early transplantation and on conditioning regimens that are less It will be however, to determine the role of allogeneic BMT in MM by a randomized trial with other intensive Such a comparison however, been in the current Dutch HOVON In this phase III patients are randomized between interferon maintenance therapy and PBSCT after induction therapy with VAD and Patients years and an HLA-identical are to allogeneic BMT after VAD and and compared with the patients in the other two The occurrence of molecular remissions after allogeneic BMT and the lower relapse rate from CR in patients after allogeneic BMT as compared to patients after PBSCT the of a beneficial graft-versus-myeloma effect was recently by remissions after donor in patients with a relapse after allogeneic BMT In a recent of 17 patients treated in the the response rate was including CR patients responded after dose escalation of donor cells. At months the overall survival was factors for a positive response were a dose of and the of and/or on these findings a clinical was using a dose of 1 which may be increased to 5 if there is no response and no and may be in of no response after the since these may the as illustrated by the follow-up of a patient with minimal residual disease who achieved a molecular remission after The of is the of The use of donor which are with the gene or the of purified allogeneic cells may be to control As is probably by cells against on myeloma cells, of cells might the as administered defined number of cells after of to patients with relapsed haematological after allogeneic in this study was low and were achieved in of six patients with relapsed MM. of responding patients clinical that other than can be in the as patients with relapsed multiple myeloma after allogeneic bone marrow transplantation. follow-up of disease in a patient with multiple myeloma as measured by and bone marrow cell remissions were after allogeneic BMT and after the first donor in with disease was observed after the second series of the graft-versus-myeloma effect was doses are in number of body of tumour load by a patient-specific quantitative showed disease in the bone marrow during CR and during relapse. remission was achieved after the of three The quantitative PCR was based on and allele-specific primers to the region of the heavy chain gene as previously described It is that further improvement in the outcome of myeloma will be achieved by treatment either by more intensive strategies or by the of the increase in overall and complete response after autologous stem cell transplantation has not the ultimate outcome of MM, although the achievement of CR in a small category of patients that survival can be on the current phase III studies, patient may be who will benefit from different treatment of patients should be done by with prognostic such as B2M with the recently important abnormalities and of response and relapse should be done with the of molecular to minimal residual disease. such as maintenance therapy with e.g. and recently shown to be after allogeneic and with tumour or may help to relapse. The of a graft-versus-myeloma effect and the recent of less regimens for allogeneic stem cell the possibility to novel strategies use of the graft-versus-myeloma effect after such transplants may be achieved by repeated until maximal response has this approach is on a is because these transplants may be in patients to 65 years and may be from donor