Transplantation for Multiple Myeloma

Total Therapy I, trial intended to increase the frequency and duration of complete response, thus extending OS [9]. The concept was inspired by the St Jude’s Children’s Hospital Total Therapy (TT) programs for acute leukemia which have made unprecedented progress in curing children with both acute lymphoblastic and myeloid leukemias. It was recognized in adult acute leukemia that cures were only obtained when a CR rate of ≥40 % was accomplished. A single MEL 200 AT resulted in a CR rate of usually no more than 20 % [10, 21, 22]. Thus, using CR as a substitute for survival, the underlying hypothesis was that a more marked increase in CR rate, from less than 5 % with standard MEL-prednisone (MP) to 40 % would produce significant prolongation of EFS and OS, and perhaps attain cure [5]. These observations led to initiation of Total Therapy Trials in MM.

The schema for TT1, TT2, and TT3 trials is already published (Figs. 1 and 2). The results of TT4, TT5, and TT6 trials are still awaited. The long-term results of TT1, which enrolled 231 patients, were published and the median follow-up was an unprecedented 12 years [23]. At 10 and 15 years, respectively, 33 and 17 % of patients are alive; 15 and 7 % are event free; and 18 and 12 % of those achieving CR remain in uninterrupted remission.

The successor Phase III trial, TT2 with 668 enrollees, delivered more intense treatment by intensifying remission induction, by adding consolidation chemotherapy post-tandem AT, and by providing high-dose pulsed dexametasone during maintenance with INF-α [24]. In addition, patients were randomized to receive thalidomide from the outset until disease progression or adverse events.

The thalidomide group had significantly higher CR rates and 5-year EFS compared to the control group (62 vs. 43 %, and 56 vs. 44 %, respectively). However, the 5-year OS was approximately 65 % in both arms of the study, which could be explained by worse post-relapse survival in the thalidomide arm, suggesting that continuous exposure to thalidomide may promote drug resistance.

EFS and OS were adversely influenced by calcium (CA), elevated LDH and albumin <3.5 g/dl.

The non-thalidomide arms in TT2 and TT1 were recently compared in order to examine the potential benefit conferred by dose-intensified induction chemotherapy and post-tandem AT consolidation chemotherapy applied in TT2 without having thalidomide as a confounding variable [25]. The CR rates in both trials were similar at 41 and 43 %. However, the 5-year estimates of continuous CR (45 vs. 32 %), and 5-year EFS (43 vs. 28 %) were significantly superior in TT2, with a trend to improved OS (62 vs. 57 %). Patients who achieved CR in the first year and had tandem AT within 1 year also had superior OS.

The treatment-related mortality was similar in both trials at approximately 7 %. These data indicate that, overall, TT2 without thalidomide was superior to TT1. Since CA is an important prognostic factor, the outcome was examined in both trials in patients with good-risk (normal CA) and high-risk features (abnormal CA). TT2 especially benefited the two-thirds of good-risk patient who entered CR more frequently and had a longer duration of CR with superior EFS and OS. Thaldomide conferred benefit to patients with a standard risk GEP and abnormal metaphase cytogenetics; a surprising finding, which has not been satisfactorily explained. The TT3 trial combined bortezomib with DTPACE portion as induction and consolidation. The initial results of TT3 trial are, respectively, with an unprecedented 80 % of patients remaining in CR at 2 years with a treatment-related mortality of only 5 %.

These TT trials demonstrated that it is feasible in myeloma to achieve a long-term survival of >30 % at 10 years with tandem transplants with high-dose MEL. This also sets a new yardstick against which recently developed newer drugs should be tested.

In the IFM 94 study, French intergroup compared single transplant with MEL 140 with tandem transplants. In tandem transplant arm comprised of conditioning with MEL 140 and MEL 140/TBI with the first and second transplant, respectively [26]. A superior 7-year probability of EFS and OS in the tandem AT arm (20 vs. 10 % and 42 vs. 21 %, respectively) was observed.

In Bologna, 96 studies showed that tandem AT significantly increased the probability to achieve CR from 33 to 47 % and extended the 5-year EFS from 29 to 17 % [27]. Tandem AT prolonged EFS by approximately 1 year and was related to a higher CR rate. Twenty percent of patients who failed to achieve (n)CR post after the first transplant did so after the second transplant. Patients who were sensitive to conventional chemotherapy with VAD achieved a CR rate of 73 % with double AT, versus 52 % in the single AT group. Conversely, the CR rates in the single and tandem AT arms were similar in patients’ refractory to VAD, suggesting that resistance to conventional chemotherapy was difficult to overcome even by double AT. There was no benefit in terms of OS, which was similar after single and double AT (46 vs. 43 %). The 2-year OS from relapse in the single transplant arm was longer than in the second transplant arm (62 vs. 51 %), which can be explained by the sequential application of salvage therapies. Approximately one-third of the patients received a second, unplanned AT, obscuring the survival benefit conferred by a second planned procedure. In addition, half were treated with novel agents such as thalidomide and bortezomib, which may have the potential to reverse resistance to HDT.

The French MAG 95 used a two × two design in 230 patients younger than 56 years to study the effects of both tandem AT and purging of the autograft using CD34+ selection. Tandem AT using unselected CD34+ cells resulted in improved OS compared to single AT. CD34+ cell selection did not confer survival benefit and was associated with an increased risk of infectious complications [28].

Table 4 summarizes the studies of tandem transplants in MM patients.

We recommend that tandem transplant should not be done outside the context of clinical trials. The studies show that tandem AT is well tolerated by younger patients with good performance status. It has acceptable treatment-related morbidity and mortality. Patients who appear to benefit most from a second AT procedure are those who are not in (n)CR after the first AT. Patients with aggressive disease, e.g., abnormal CA, do not seem to do better with tandem AT.

Alkylating agents such as melphalan and the immunomodulatory lenalidomide can impair the collection of hemopoeitic stem cells (HSC). Melphalan is best completely avoided, whilst lenalidomide exposure should be limited to four cycles prior to collection. HSC are typically collected by apheresis from the peripheral blood after stimulation with granulocyte colony-stimulating factor (G-CSF). Peripheral blood progenitor cells (PBPCs) are preferable to bone marrow cells for transplantation due to quicker engraftment and a potential for less contamination of the infused cells with tumor cells. G-CSF alone and G-CSF plus cyclophosphamide are the most common regimens used for stem cell mobilization [31‐33]. The usual dose of G-CSF is 10 mcg/kg per day subcutaneous. Cyclophosphamide is usually administered at a dose of 1.5–3 g/m² intravenous for 1 or 2 days. The choice between the two methods is mainly dependent on institutional preference and experience. There are no randomized trials comparing the two approaches. Some institutions use G-CSF plus cyclophosphamide as their standard. G-CSF plus cyclophosphamide has the advantage of providing much higher PBPCs than G-CSF alone, but also carries the risk of longer time to start collection and the risk of neutropenia. Plerixafor is a chemokine receptor type 4 inhibitor (CXCR4) and it impairs the binding of hematopoietic stem cells within the bone marrow microenvironment [34]. Plerixafor is by some, primarily reserved for patients who fail stem cell collection with either G-CSF or G-CSF plus cyclophosphamide, while some use pleraxifor up-front usually together with G-CSF [35, 36]. Once the mobilization regimen is initiated, patients are monitored by peripheral blood CD34 counts. Apheresis is started when the peripheral CD34+ counts reach 20 CD34 cells/microL. Apheresis is performed with a goal of collecting between at least minimum 3 × 10⁶ CD34+ to 6 × 10⁶ CD34+ cells/kg. A minimum of 2 × 106 CD34+ cells/kg is considered essential for one transplant. We routinely collect >20 × 106 CD34+ cells/kg. Infusion of a larger number of stem cells will hasten recovery, minimize blood product use, and reduce area of CRP under the curve, the latter being a surrogate marker for infectious complication. This leaves ample cells for use as salvage transplant or for reestablishing hematopoiesis if multiple lines of salvage therapy have exhausted hematopoiesis. In general, enough PBPCs are harvested for two transplantations. PBPCs are cryopreserved in 5 % dimethylsulfoxide to be thawed at the bedside at the time of infusion [16].

The standard conditioning regimen used for AT in multiple myeloma is melphalan at a dose of 200 mg/m², with dose reductions based on age and renal function.

The use of this dose is primarily based upon two randomized trials that have compared melphalan 200 mg/m² with a lower dose of melphalan in conditioning for HCT. The French cooperative study compared melphalan 140 mg/m² plus 8 Gy total body irradiation versus Melphalan 200 mg/m² in 282 newly diagnosed patients. Patients randomly assigned to melphalan 200 mg/m² had faster hematologic recovery, less mucositis, less transfusion requirements and, shorter hospitalizations. In this study, survival at 45 months was significantly better in patients receiving melphalan 200 mg/m² (66 versus 46 %).The Italian study comparing melphalan 200 mg/m² with melphalan 100 mg/m² as preparative regimens for HCT in 298 newly diagnosed symptomatic patients <65 years of age [37], patients who received melphalan 200 m/m² had significantly longer median progression-free survival (31.4 versus 26.2 months) and a trend towards improved projected overall survival at 5 years (62 versus 48 %). Higher percentage of gastrointestinal toxicity (11 versus 1 %), mucositis (17 versus 3 %), need for intravenous broad-spectrum antibiotics (41 versus 29 %), and need for platelet transfusions (56 versus 38 %) was noted in melphalan 200 mg/m² group. In two other randomized studies, the use of more intensive preparative regimens, such as thiotepa, busulfan, and cyclophosphamide [38] or high-dose idarubicin, cyclophosphamide, and melphalan [39] did not result in better outcomes than melphalan at a dose of 200 mg/m². Dose adjustments for melphalan for obese patients can lead to a lower dose of melphalan when calculated per kg and could potentially lead to under treatment.

(a) Patients with renal insufficiency—Randomized trials that have shown benefit with HCT compared with chemotherapy have mainly studied patients with serum creatinine <2.0 mg/dL. The data for melphalan use in patients with serum creatinine >2.0 mg/dL comes from a retrospective review of 81 patients with MM and renal failure (plasma creatinine >2 mg/dL) who underwent autologous HCT [40]. Sixty patients who received melphalan 200 mg/m², excessive toxicity was noted in patients than the subsequent 21 patients who received melphalan 140 mg/m². The patients who received melphalan 200 mg/m² had significantly higher rates of severe pulmonary toxicity (57 versus 17 %) and mucositis (93 versus 67 %). In the patients with creatinine >2.0 mg/dl, the treatment-related mortality, event-free, and overall survival were not significantly different between melphalan 200 mg/m² and melphalan 140 mg/m².

(b) Older adults—In order to reduce toxicity in patients over 65 years of age, various strategies are studied. For example, once two sequential courses of an intermediate dose of melphalan (100 mg/m²) followed by hematopoietic stem cell rescue (MEL100) has been studied as an alternative to high-dose melphalan 200 mg/m2. In a multicenter trial of 194, in the subgroup of 80 patients aged 65–70, higher rates of near complete plus partial remissions (47 versus 16 %), longer median event-free survival (28 versus 16 months), and overall survival (58 versus 37 months) were observed in patients receiving melphalan 100 mg/m². However, higher short-term toxicity (e.g., mucositis, fever, need for red cell, or platelet transfusions) was noted. The French Intergroupe Francophone du Myelome (IFM) group reported on 447 previously untreated patients with myeloma patients aged 65–75 years who received melphalan, prednisone, and thalidomide (MPT) or melphalan and prednisone alone (MP) or tandem autologous HCT using MEL100 [41]. There was no difference in median overall survival between the MP and transplant arms. This study suggests that an intermediate dose of melphalan may not be the optimal conditioning regimen. At present, we use melphalan 200 mg/m² as standard conditioning regimen for HCT myeloma patients and reduce the dose to 140 mg/m² in patients with renal impairment or those over the age of 65.

Autologous hematopoietic cell transplantation (HCT) has been performed in both the inpatient and outpatient settings. Approximately 24 h after completion of the preparative chemotherapy, peripheral blood progenitor cells (PBPCs) are reinfused. A period of pancytopenia follows. Red blood cell and platelet transfusions are administered as necessary while hematopoietic colony-stimulating factors (i.e., G-CSF) are used to speed neutrophil engraftment. Neutrophil engraftment usually occurs by day 10–11 and platelet counts are expected to recover to greater than 20,000 by day 16 [42]. Red blood cell transfusion requirements during autologous HCT are usually minimal. HCT without transfusion support, although not ideal has recently been reported in a series of 50 Jehova’s witnesses with acceptable toxicity [43]. At our institution, approximately 80–90 % of patients undergo autologous transplantation entirely as an outpatient, with daily monitoring until full engraftment has occurred [44]. Patients who undergo HCT are at risk for bacterial, viral, and fungal infections, the time course of which varies in the post-transplant period, according to the degree of immune deficiency and cytopenia induced by the transplantation procedure. Approximately 40 % of patients with multiple myeloma undergoing autologous HCT will experience febrile neutropenia [45]. As a result, prophylactic therapies to prevent infection including antiviral and antifungal drugs are recommended during the period of increased risk. In addition, all markers of potential infection must be investigated thoroughly.

The absence of a plateau on survival curves post HDT and AT with ongoing relapses justifies the exploration of maintenance strategies with the prolongation of response duration as the goal. In addition, AT may result in a very low tumor burden, which may be susceptible to long-term suppression by drugs or immunomodulatory agents. A number of agents have been or are being explored for maintenance, including interferon alpha (IFN-α), corticosteroids, biphosphonates, thalidomide, lenalidomide, bortezomib, and combinations of these drugs. Several studies show that consolidation and maintenance post AT can further reduce tumor burden and improve outcome.

IFN was one of the first drugs to be studied as maintenance post AT. It yielded a small benefit in a meta-analysis from the European Group for Blood and Marrow Transplantation (EBMT) [46]. The study by Australian myeloma group (with a median follow-up 5.4 years), thalidomide administered in combination with prednisolone for 1 year was found to significantly prolong PFS and OS compared with prednisolone alone: 5-year PFS was 27 % for Thal/Pred versus 15 % for Pred, P = 0.005; 5-year OS was 66 versus 47 %, respectively, P = 0.007) [47]. Minimum thalidomide exposure required was at least 8 months to gain a PFS and OS advantage and there was no difference between the two arms regarding overall response rate to salvage therapy or post relapse. This suggested that acquired resistance in this study was not an important issue for thalidomide-treated patients.

In the meta-analysis of five trials involving patients, Morgan et al. found a significant late OS benefit for thalidomide (P < 0.001, 7-year difference hazard ratio (HR) = 12.3; 95 % confidence interval, 5.5–19.0). Similarly, Lenalidomide has been investigated in the post-transplant setting in three large studies. In the study conducted by the IFM group, patients were randomized to lenalidomide maintenance until progression or no maintenance following a single or tandem ASCT step and two cycles of lenalidomide consolidation in both arms [48]. The lenalidomide was stopped at a median of 2 years (range 1–3 years) due to concerns regarding second primary malignancies. With a median follow-up of 67 months from randomization, the PFS for patients who had received lenalidomide maintenance was significantly longer than for those who had not received any maintenance therapy (46 versus 24 months, P < 0.001) [49]. Although OS did not achieve statistical significance in two arms (82 versus 81 months, respectively, P = 0.8), the second PFS and survival after the first progression were shorter in the lenalidomide maintenance arm. In addition, the cumulative incidence of second primary malignancy was significantly higher with lenalidomide. When examining cytogenetic risk, progression was superior for the lenalidomide arm for patients with or without 13q deletion and without t(4;14) or 17p deletion, but did not reach significance for patients with either t(4;14) or 17p deletion. The CALGB conducted a large placebo-controlled randomized study of lenalidomide maintenance following ASCT [50]. In this study, lenalidomide was administered until disease progression. 86 of 128 eligible patients received lenalidomide maintenance. At a median follow-up of 34 months, 37 % who received lenalidomide maintenance and 58 % who received placebo had disease progression or had died. The median time to progression was 46 months in the lenalidomide group and 27 months in the placebo group (P < 0.001). The median OS in both arms had not been reached, with 85 % of the lenalidomide-arm patients and 77 % of the placebo-arm patients who were alive at the time of the analysis having died (P < 0.03). The rate of second primary malignancy appeared to plateau at 2 years post SCT. This study was updated in 2013. At median follow-up of 48 months, the intent-to-treat analysis demonstrated that the OS was 80 % for the lenalidomide arm and 70 % for the placebo group (P = 0.008) with a continued PFS advantage in the lenalidomide arm. In the study by GIMEMA group, at a median follow-up of 51.2 months, the median PFS was superior with lenalidomide maintenance (PFS: 41.9 months for lenalidomide versus 21.6 months for placebo, P < 0.0001), but 3-year OS was not significantly prolonged (3-year OS: 88.0 % versus 79.2 %, respectively, P = 0.14).

Bortezomib maintenance therapy has been investigated in two phase 3 studies, the HOVON and GMMG studies. A landmark analysis from the start of maintenance showed that OS was superior for patients on the bortezomib arm, while PFS was comparable. The limitation of these studies was that, while the results of these trials confirmed the important role of bortezomib-containing regimen in the treatment of newly diagnosed MM, the design of the trials made it difficult to interpret the role of bortezomib maintenance difficult as it was not possible to delineate the individual contributions of induction versus the maintenance treatment as part of this design. Furthermore, the OS benefit was seen only for patients with del17p cytogenetic abnormalities and those who presented in renal failure. In a large randomized phase 3 study conducted by the Spanish myeloma group, the combination of bortezomib and thalidomide was compared to thalidomide and to Interferon as maintenance therapy and a significant benefit in PFS was found for the VT combination with a median follow-up of 34.9 months [15].

In the context of AT trials for myeloma, disappearance of the M-protein is generally thought to play a pivotal role in predicting superior survival [9, 14, 18, 47, 49, 51]. However, several observations have challenged the notion that a CR is an absolute requirement for prolongation of survival in myeloma. The Mayo Clinic has reported that patients who underwent AT have similar PFS and OS irrespective of whether they achieved CR [52]. Patients with CA in TT2 had similar CR rates of 40 % compared to those without CA, yet their median survival was significantly shorter post high-dose MEL therapy. A similar observation applied to the thalidomide arm in TT2 where a higher CR rate and EFS did not translate into superior survival. The median time to CR in TT2 was 12 months despite the application of an intensive therapeutic program [24] In addition, the disappearance of MRI-defined focal lesions (FL), which are potential sites for surviving resistant myeloma cells, lags 2 years behind the disappearance of M-protein from blood and urine [53]. Paradoxically more aggressive disease features such as CA, elevated LDH, and IgA isotype are predictors of CR [25]. We have recently reported that high serum free light chain levels at diagnosis and a rapid reduction after induction therapy are predictive for achieving CR [54]. These parameters reflect more proliferative myeloma which, on the one hand, is inherently more sensitive to combination chemotherapy, and on the other is linked to shorter EFS and OS due to rapid myeloma regrowth when tumor reduction is insufficient.

Conversely, patients with myeloma evolving from monoclonal gammopathy of unknown significance (MGUS) or with smoldering myeloma have significantly lower CR rates, yet equal EFS and OS compared to patients with de novo myeloma. The lower CR rate in these patients is likely to be due to a lower proliferative rate, reflecting more indolent disease, which is less susceptible to eradication by both standard- and high-dose chemotherapy. Interestingly, these patients also have fewer MRI-defined FL. Patients with MGUS-like myeloma have favorable clinical characteristics, with lower CR rates yet superior survival compared to patients with non-MGUS-like myeloma. Attainment of CR in myeloma is often a gradual and cumulative process. It appears, therefore, that achieving a CR is only critical for prolonging survival in a truly high-risk group of myeloma that thus far can only be captured by gene expression profiling [55]. These data also suggest that the use of CR as a surrogate marker for OS and EFS in clinical trials of novel agents cannot be applied without characterizing myeloma at the molecular level and should be used with extreme caution.

Results of multicolor flow cytometry and deep-sequencing studies suggest that among patients achieving a complete response, MRD-negative status is associated with significant improvements in PFS and OS. The UK Medical Research Council IX trial assessed the importance of achieving MRD. There was a significant improvement in OS for each log depletion in MRD level. The median OS was 1 year for ≥10 %, 4 years for 1 to <10 %, 5.9 years for 0.1 to <1 %, 6.8 years for 0.01 to <0.1 %, and more than 7.5 years for <0.01 % MRD level. The detection of minimal residual disease MRD in myeloma using a 0.01 % threshold (10(−4)) after treatment is an independent predictor of PFS [56].

Allogeneic hematopoietic cell transplantation (HCT) is potentially the curative treatment for multiple myeloma. However its use is limited because of high rate of treatment-related mortality and its efficacy compared with autologous HCT is not fully established. With the advent of new proteasome inhibitors and IMiDs, role of allogeneic transplant needs to be scrutinized. In subsequent sections, we will review the role of allogeneic transplant in MM.

Myeloablative transplants require that patients receive high-dose chemotherapy with or without total body radiation, followed by donor stem cell infusion. Myeloablative transplants are not the preferred modality and are very infrequently performed. Hence this is not discussed further in this chapter.

Syngeneic allogeneic transplants are from the identical twin donor and only a limited number of syngeneic transplants have been performed in multiple myeloma. The EBMT analyzed 25 syngeneic transplants and compared the results with 250 patients who underwent either autologous or allogeneic HCT (n = 125, each) HCT [57]. The TRM was substantially lower with only two patients dying due to transplant-related toxicity as compared to myeloablative allogeneic transplant. Overall survival for the patients undergoing syngenic transplants was 73 months, significantly better than that of the case-matched autologous transplants (44 months); both groups outperformed the allogeneic transplants. The Seattle Marrow Transplant Team performed syngeneic transplants on five patients with multiple myeloma [58]. Four patients responded to therapy, while one patient died one month after transplant from cytomegalovirus-associated interstitial pneumonitis. Response durations for these four patients were 6, 17, 18, and more than 72 months. These reports are useful in exploring the various treatment options but do not support the use of syngeneic HCT in place of an autologous HCT if a twin donor is available.

The results of non-myeloablative T-cell depleted transplants have been summarized in Table 5. Longer-term follow-up of these studies is awaited. It is unclear what impact non-myeloablative HCT will have given the improved outcomes with autologous HCT after the introduction of novel agents utilized in induction, consolidation, and maintenance, in the era of novel agents. In current scenario, non-myeloablative-allogeneic transplantation remains investigational and its role needs to be validated in the era of novel agents.

The role of auto–allo versus tandem autologous transplant is debatable. A meta-analysis of 1822 patients with previously untreated myeloma comparing double autologous HCT versus a single autologous HCT followed by non-myeloablative allogeneic HCT [62], allogeneic HCT was associated with greater treatment-related mortality (relative risk [RR] 3.3; 95 % CI 2.2-4.8 (RR 1.4; 95 % CI 1.1–1.8), but similar overall survival at and beyond 36 months.

The following is a brief summary of the largest trials conducted in this setting

BMT CTN Trial [63]:

Hovon 50 [64]:

French IFM [65]:

Spanish PETHEMA [15]:

However, the Italian study of 162 patients showed better event-free survival in patients who underwent sib matched allogeneic transplant [66].

The authors of this chapter noted that hardly any genetic data was included in these trials. Hence in the era of genomically driven personalized medicine, the role of auto/allo remains investigational and the data needs to be carefully analyzed.

It is important to recognize that MM is complex disease with significant clonal heterogeneity. Transplantation helps in early reduction of clonal diversity and increases the likelihood of cure. Long-term follow-up studies show that AT can achieve profound cytoreduction and likely cure a portion of patients even before the introduction of novel drugs. Total Therapy 1 (TT1), the first tandem AT trial for myeloma, enrolled 231 patients; with a median follow-up of 17 years, 23 remain alive and progression-free with a plateau on the overall survival (OS) curve appearing around 14 years [23]. Martinez-Lopez et al. reported on 344 patients who received AT between 1989 and 1998 who had a median follow-up of 153 months. A plateau in OS appeared after 11 years and, with a further follow-up of 5 years, no relapses were observed [67]. Targeted therapy has already demonstrated some efficacy, but may merely select for resistant subclones, unless a given mutation drives the disease as recently has been described for KRAS/NRAS activating mutations. High-risk myeloma even with modern genetic analysis will likely remain a challenging disease to treat in years to come. AT, tandem autotransplants, autotransplant followed by reduced-intensity allogeneic transplant and allogeneic transplant will remain an important option in the therapeutic armamentarium of myeloma. Novel immune therapeutic agent anti-CD38 antibodies, i.e., daratumomab, checkpoint blockade inhibitors, cellular therapies, and vaccines will likely become available very soon and this can enhance the anti-myeloma response without transplant-associated risks. It is conceivable that such strategies will be complimentary and not replacement for transplant.