Autologous stem cell transplantation in multiple myeloma is not dead but alive and well

Abstract

In multiple myeloma (MM), the use of high-dose chemotherapy with autologous stem cell transplantation (ASCT) led to incremental advances in patient management in the 1990s. The clinical results for patients dramatically improved further in the 2000s with the introduction of immunomodulatory drugs and proteasome inhibitors. In the ‘modern’ era for MM treatment, transplant trials strongly support the use of upfront ASCT in the context of novel agents, and until proven otherwise, the old ASCT remains the standard of care for eligible patients. Nevertheless, some issues remain unresolved and this editorial aims to highlight the concerns to be addressed in the future.

Keywords:

• autologous stem cell transplantation
• high-dose melphalan
• immunomodulatory drugs
• multiple myeloma
• new drugs
• proteasome inhibitors

1. Introduction

From the history of multiple myeloma (MM), we have learned that among the multiple combinations of chemotherapy available until 1996, only melphalan and prednisone could improve the overall survival (OS) of patients. In that year, a French group [1] identified ‘something’ that could make a difference in the treatment and survival and that ‘something’ was high-dose chemotherapy (HDC) and autologous stem cell transplantation (ASCT). In the 2000s, a dramatic increase in improvement rates was observed with the advent of immunomodulatory drugs (IMiDs) and proteasome inhibitors, and if this trend continues, the 5-year survival for a patient diagnosed in 2014 should be ∼ 66% [2].The 2010s are characterized by studying the optimal combination, sequence and duration of therapies; however, the specific role and timing for ASCT has not been established with modern induction regimens. In the scientific community, there has been much debate on whether to withdraw the procedure, but we believe that ASCT in MM is not dead but alive and well.

2. Evidence in favor of ASCT

Before the use of the new drugs in induction therapy, the complete remission (CR) rate did not exceed 10%. In this scenario, the rationale for proceeding to HDC with ASCT was to increase the depth of response. Over the past decade, this approach has been considered the standard of care for eligible patients with newly diagnosed MM, based on increased rate of CR, prolonged disease-free (DFS) 1, 3, 4, 5, 6, 7 and OS 1, 4,compared with conventional chemotherapy (CC).

The major limitation to these studies is that all randomized trials of ASCT after induction therapy were designed and implemented before the availability of novel agents, but we believe that the role of transplantation is equally as strong in the ‘modern era’ (Table 1).

Table 1. Multiple myeloma: the role of autologous stem cell transplantation in the era of immunomodulatory drugs and proteasome inhibitors.

Author (year)	Study design	Induction therapy	HDC/CC	Results
Harousseau et al. (2008) [11]	Phase III, 482 patients	Vel/Dex VAD	MEL	Vel/Dex CR + nCR/≥ VGPR vs VAD post-induction (21.3%/46.7% vs 8.3%/18.6%, p = 0.0023/< 0.0001); CR + nCR/≥VGPR Vel ≥ Dex advantage was maintained post-ASCT, both in patients who received ASCT (40.8%/71.8% vs 28.8%/51%, p = 0.0089/< 0.0001)
Cavo et al. (2009) [12]	Phase III	VTD (n = 236); TD (n=238)	Double MEL	CR (≥VGPR) rate with VTD induction therapy was 19% (62%) vs 5% (31%) with TD (p < 0.001 for both CR and ≥ VGPR comparisons); progression through the subsequent treatment phases was associated with an increase in the frequency of CR and ≥ VGPR up to a final value of 44 and 80%, respectively, in the VTD arm; the corresponding rates in the TD arm were 32% (p = 0.02) and 65% (p=0.001)
Harousseau et al. (2010) [10]	Phase III, 482 patients	VAD (n = 121), VAD plus DCEP (n = 121), Vel/Dex (n = 121), Vel/Dex plus DCEP (n = 119)	MEL	Post-induction: CR/nCR (14.8 vs 6.4%), at least VGPR (37.7 vs 15.1%), and overall response (78.5 vs 62.8%) were higher with bortezomib plus dexamethasone versus VAD; response rates were similar in patients who did and did not receive DCEP Post-first transplantation: CR/nCR (35.0 vs 18.4%) and at least VGPR (54.3 vs 37.2%) rates remained significantly higher with bortezomib plus dexamethasone; PFS was 36.0 vs 29.7 months (p = 0.064) with bortezomib plus dexamethasone versus VAD; respective 3-year survival rates were 81.4 and 77.4% (median follow-up, 32.2 months)
Rosinol et al. (2012) [8]	Phase III, 386 patients	VTD (n = 130); TD (n = 127); VBMCP/VBAD/B (n = 129)	MEL	CR rates in VTD arm: 35% pre-transplant; 57% post-transplant; CR rates in TD arm: 14% pre-transplant; 40% post-transplant
Palumbo et al. (2014) [9]	Phase III, 273 patients	RD	MEL (n = 141) vs MPR (n = 132)	PFS: MEL 43.0 vs MPR 22.4 months; HR for progression or death = 0.44; 95% CI: 0.32 – 0.61; p < 0.001; 4-year OS MEL 81.6% vs MPR 65.3%; HR for death = 0.55; 95% CI: 0.32 – 0.93; p = 0.02)

ASCT: Autologous stem cell transplantation; CC: Conventional chemotherapy; CR: Complete remission; DCEP: Cyclophosphamide, etoposide, and cisplatin; HDC: High-dose chemotherapy; HR: Hazard ratio; MEL: Melphalan 200 mg/mq; MPR: Melphalan–prednisone–lenalidomide; nCR: Near complete remission; OS: Overall survival; PFS: Progression-free survival; RD: Lenalidomide plus dexamethasone; TD: Thalidomide/dexamethasone; VAD: Vincristine/doxorubicin/dexamethasone; VBMCP/VBAD/B: Vincristine, BCNU, melphalan, cyclophosphamide, prednisone/vincristine, BCNU, doxorubicin, dexamethasone/bortezomib; Vel/Dex: Bortezomib/dexamethasone; VGPR: Very good partial remission; VTD: Bortezomib/thalidomide/dexamethasone.

The results of the Phase III PETHEMA trial [8] support the use of upfront ASCT in the context of novel drugs. In this study, the response rates were evaluated after induction therapy and after ASCT. The CR rates increased from 35% pre-transplant to 57% post-transplant, in the group treated with bortezomib/thalidomide/dexamethasone (VTD) as induction therapy and from 14 to 40% in the group treated with thalidomide/dexamethasone (TD).

Recently, Palumbo et al. [9]. published results of a randomized, Phase III study comparing melphalan 200 mg/mq (HDM) plus ASCT with melphalan–prednisone–lenalidomide (MPR). Both progression-free survival (PFS) and OS were significantly longer with HDM plus ASCT than with MPR (median PFS: 43.0 vs 22.4 months; hazard ratio [HR] for progression or death = 0.44; 95% CI: 0.32 – 0.61; p <0.001; and 4-year OS: 81.6 vs 65.3%; HR for death = 0.55; 95% CI: 0.32 – 0.93; p = 0.02).

Results from a French trial [10], with patients receiving primary induction therapy with bortezomib/dexamethasone (Vel/Dex) versus vincristine/doxorubicin/dexamethasone (VAD) showed a marked improvement in overall responses with Vel/Dex over VAD. Responses were evaluated after induction treatment and post-ASCT. After the first ASCT, CR/near CR was 35% in the Vel/Dex arm, compared with 18.4% in the VAD arm. The very good partial remission (VGPR) was 54.3 versus 37.2%. PFS was 36 versus 29.7 months (p = 0.064) with Vel/Dex vs VAD. The PFS was longer in patients achieving greater than or equal VGPR after ASCT [11].

In another study, patients were randomized to induction therapy with VTD or TD before double ASCT. The three-drug regimen yielded high response rates, with a CR of 19 versus 5% and VGPR of 62 versus 31%. After ASCT, improved incremental responses were still seen with VTD compared to TD [12].

Taken together, these studies suggest that improved responses with the novel induction therapies result in improved outcomes after ASCT, and currently, the most common indication for ASCT in Europe remains MM [13].

2.1 Toxicity, morbidity and quality of life

HDM has become substantially less toxic over time and it should be considered a safe procedure. The Center for the International Blood and Marrow Transplant Registry showed that the risk of death after ASCT in the 2000s has decreased and published data suggest that day 100 mortality of upfront ASCT is 1% [14]. The introduction of myeloid growth factors after transplantation, the use of peripheral blood stem cells (SCs) in place of bone marrow SCs, modern, better-tolerated induction regimens and other improvements in supportive care have reduced the transplant-related mortality rate, and many transplant teams routinely perform ASCT in the outpatient setting [15]. Also, in contrast to popular perception, mortality, morbidity and societal cost from ongoing non-transplant therapy are not negligible.

3. Questions for the future

In the ‘modern’ era for MM treatment, transplant trials strongly support the use of upfront ASCT in the context of novel agents, and until proven otherwise, ASCT remains the standard of care for eligible patients.

However, there are some highly debated issues that the scientific community needs to resolve. First, the type and duration of induction therapy has not been well defined and there is little consensus on the optimal non-melphalan-containing induction regimens. Although randomized trials have been conducted comparing older regimens with newer regimens, containing IMiDs or proteasome inhibitors, there are few randomized trials that have compared these modern combinations [16]. Thus, the best regimen can only be extrapolated indirectly from Phase III trials showing improved responses before transplant 8, 9, 10, 17. However, in the context of evidence-based medicine, bortezomib-based induction seems to be most appropriate for patients who are on their way to an ASCT [18].

The second issue is to better clarify the definition of CR. The International Myeloma Working Group response criteria were developed from the European Society for Blood and Marrow Transplantation, International Bone Marrow Transplant Registry and Autologous Blood & Marrow Transplant registry response criteria, with revisions and improvements to help uniform reporting 19, 20. However, we believe that the current electrophoresis-based response definitions are inadequate because improvements in sensibility of molecular and flow-cytometry techniques have clearly demonstrated the presence of significant residual MM disease using the current CR definition [21]. In this context, one needs to reconsider the criteria for defining CR (i.e., whether based on flow-cytometry data, immunohistochemistry, positron emission tomography, free light chain levels, etc.).

The third open question is to understand the best SC mobilization procedure to be administered. Mobilization with G-CSF alone is well tolerated but its use can be limited by suboptimal SC yields [22]. Adding chemotherapeutic agents to G-CSF may increase SC yields. The most commonly used chemotherapy-based mobilization in MM includes high-dose cyclophosphamide [23]. There is indication that chemo-mobilization reduces the tumor load in vivo and tumor cell contamination in the apheresis product. The benefit of higher cell yields (compared with G-CSF alone) may be offset by less predictability of timing and an increased risk for the patient [24]. In the era of novel agents, chemotherapy mobilization is clearly not superior to mobilization by growth factors alone. The impressive results achieved by incorporating newer novel agents into frontline therapy are observed not only in response rates but are also correlated with the potential for obtaining minimal residual disease-negative, tumor-free hematopoietic progenitor cell grafts [25]. In this way, it becomes unnecessary to expose the patient to the risks of chemotherapy, the effectiveness of which on the disease is questioned [26].

The fourth issue to be clarified is the number of transplants and the timing of transplantation. ASCT may be single or tandem (a planned second course of HDM within 6 months of the first). Although the studies were performed in the old era, the tandem ASCT approach achieved improvement in OS [27], even though a survival benefit was mainly seen in those patients who failed to achieve at least a VGPR 28, 29. We underline that patients who achieve a CR and stop treatment is a clinical error, because what we may define as an immunofixation-negative CR may still have tumor cells present. The improved depth of response with newer induction strategies and the efficacy of continuous treatment with IMiDs and proteasome inhibitors is a driving force behind the shift in the timing of ASCT. We believe that clinicians should deliver the best treatment to all patients, particularly when faced with a patient with a potentially good prognosis. From this stand point, it makes much more sense to intensify treatment in patients with a sensitive disease, instead of intensifying treatment in patients with aggressive disease.

Recent results with the new induction regimens suggest that there is a role for tandem ASCT in MM in the presence of adverse cytogenetic abnormalities. In the Italian trial [19] investigating VTD versus TD as induction and consolidation, PFS, DFS or OS at 3 years was comparable for patients with or without the t(4;14) translocation, indicating that the double procedure is efficacious in the presence of this high-risk marker. In another trial comparing bortezomib/doxorubicin/dexamethasone with VAD as induction therapy, the impact of del(17p13) on PFS and OS can be reduced by a novel agent induction regimen followed by a double ASCT [30].

In 2012, the Cochrane review compared tandem ASCT with single ASCT as first-line treatment in patients with symptomatic MM with respect to OS, event-free survival, quality of life and treatment- or transplantation-related mortality [31]. The authors did not consider any study to be sufficiently informative for contemporary treatment decisions in terms of single versus tandem ASCT in view of inherent biases in these studies. In addition, none of the trials integrated the so-called novel agents, which are now considered standard treatment for MM.

A second transplant need not always be performed in tandem. The second transplant can be delayed until progression and becomes a useful salvage strategy. The second ASCT in relapsed MM can be associated with superior OS and PFS compared with CC if the response duration from the first transplant is at least 12 – 24 months [32]. A single trial conducted prior to the availability of novel agents provides the only Phase III data comparing ‘early’ upfront ASCT versus ‘late’ ASCT at disease progression/relapse, with comparable OS noted in both groups [33]. Analyses of the randomized trial, conducted by Rajkumar et al., which compared lenalidomide in combination with high-dose or low-dose dexamethasone as initial therapy, showed that patients choosing ASCT after 4 cycles of therapy had a better 3-year OS than patients choosing to continue initial therapy [34]. In the recently published randomized trial by Palumbo et al. [9], the results suggest that upfront ASCT improves both PFS and OS compared to the arm without HDM.

Further ongoing randomized trials comparing upfront ASCT with new drugs without high-dose therapy will provide a definitive answer to the question if, with modern induction regimens, ASCT can add benefit. It is possible that some subgroups of patients may need early frontline ASCT, whereas in some other patients the procedure may be postponed. It is also true that an appropriate follow up is needed before any conclusion.

The fifth point is the definition of ‘transplantation eligible’. Patients treated are often older than the ones included in the published trials, and based on the increasing number of patients aged over 65 years who undergo ASCT, the performance status and comorbidities should be considered as factors defining transplantation eligibility rather than age 14, 35.

The sixth issue of debate is the role of consolidation and maintenance in the treatment of MM. Consolidation/maintenance therapy with IMiDs and proteasome inhibitors is a reasonable option for long-term disease control and improvement in OS, although no agent has been approved by the European Medicines Agency for this use. Regardless of the agent that we choose, there must be a risk–benefit discussion. The risk of side effects must be considered with the use of lenalidomide, and an increased risk of second primary malignancies has been reported with long-term exposure, which appears associated with its use in conjunction with melphalan [36]. Toxicities must be considered with long-term use of bortezomib, but the risk of peripheral neuropathy may be lessened with the use of subcutaneous administration. Peripheral neuropathy, thromboembolic and hematologic side effects must be considered with the use of thalidomide, and, in patients with selective adverse molecular cytogenetic lesions determined by interphase fluorescence in situ hybridization (FISH), thalidomide should not be used as maintenance therapy [37].

The last issue to consider is that not all MM are the same. We know that we can risk-stratify the malignancy based on very basic demographics and variables such as the free light chain assay, cytogenetic or FISH markers that can help us identify differences in the structural and functional manifestations of the disease and those indices translate into differences in clinical outcomes. Induction and HDC are good approaches for achieving CR, and in high-risk patients we need to consider maintaining these CRs. In standard-risk patients, maintenance therapy may also improve not only the duration of a response but also OS.

In conclusion, new agents and new drug combinations with ASCT have improved our ability to obtain rapid control of a patient’s disease. The critical issues associated with ASCT in MM are summarized in Table 2. Using our best strategies upfront may convert MM into a much more indolent and chronic disease.

Table 2. High-dose chemotherapy with autologous stem cell transplantation strategy in multiple myeloma: open questions.

Type and duration of novel induction therapies
Real definition of complete remission
Definition of the best stem cell mobilization procedure
Single or tandem autologous stem cell transplantation
Early upfront versus late autologous stem cell transplantation
Definition of ‘transplantation eligible age’
Role of consolidation and maintenance
Risk stratification therapy

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents, received or pending, or royalties.