Maintenance therapy in multiple myeloma

Abstract

Despite recent advances, multiple myeloma remains an incurable disease. Induction therapy followed by autologous transplantation has become the standard of care. The idea of maintenance therapy in multiple myeloma is not new. Starting with chemotherapy in 1975, to interferon in 1998, to novel agents recently, a multitude of agents have been explored in patients with multiple myeloma. In spite of the novel agents, multiple myeloma continues to be an incurable disease with the progression-free survival after autologous transplant rarely exceeding 3 years. The goal of using maintenance therapy has been to improve the outcomes following autologous transplantation by increasing the progression-free survival, deepening remissions and perhaps increasing overall survival. It has been shown that patients with a stringent complete response (CR) have a better outcome [Kapoor et al. 2013]. It is becoming increasingly common to check minimal residual disease (MRD) as a means of assessing depth of response. It has also been shown that patients with no MRD have not only a better progression-free survival but also a better overall survival compared with patients who are MRD positive. This makes it even more important to find agents for maintenance therapy, which can further deepen and maintain responses. Here, we present a comprehensive review of the agents studied as maintenance for multiple myeloma and their efficacy, both in terms of overall survival, progression-free survival and toxicity.

Introduction

Despite recent advances, multiple myeloma remains an incurable disease. Induction therapy, followed by autologous transplantation has become the standard of care. The idea of maintenance therapy in multiple myeloma is not new. Starting with chemotherapy in 1975 [Alexanian et al. 1975], to interferon in 1998, to novel agents recently, a multitude of agents have been explored in patients with multiple myeloma. In spite of the novel agents, multiple myeloma continues to be an incurable disease with the progression-free survival (PFS) after autologous transplant rarely exceeding 3 years. The goal of using maintenance therapy has been to improve the outcomes following autologous transplantation by increasing the PFS, deepening remissions and perhaps increasing overall survival (OS). It has been shown that patients with a stringent CR have a better outcome [Kapoor et al. 2013]. It is becoming increasingly common to check MRD as a means of assessing depth of response. It has also been shown that patients with no MRD have not only a better PFS but also a better OS compared with patients who are MRD positive [Paiva et al. 2008; Rawstron et al. 2013; Martinez-Lopez et al. 2014]. This makes it even more important to find agents for maintenance therapy, which can further deepen and maintain responses. Here, we present a comprehensive review of the agents studied as maintenance for multiple myeloma and their efficacy, both in terms of OS, PFS and toxicity. All the agents studied and used in practice currently or in past are listed in Table 1 and the agents currently in trial are listed in Table 2.

Table 1.

Agents studied in maintenance therapy of multiple myeloma.

Class of agent	Type of agent	Comments	Induction	Consolidation	Efficacy	Reference or ClinicalTrials.gov identifier
Cytokines	Interferon	Tested in phase III trial	Vincristine, adriamycin	Auto transplant with bone marrow stem cells	Increase in PFS; no change in OS	Alexanian et al. [1975]
	Interferon	Tested in phase III trial (84 patients)	Methylprednisolone (VAMP) or cylophosphamide + VAMP	Melphalan conditioning	No change in PFS (p = 0.09) and OS (p = 0.109)	Cunningham et al. [1998]
Glucocorticoids	Prednisone (prednisone 10 mg versus 50 mg)	Retrospective data, off label (250 patients)	Vincristine, doxorubicin, and dexamethasone with prednisone (VAD-P) or VAD-P plus quinine (VAD-P/Q)	No consolidation	Increase in PFS (14 versus 5 months; p = 0.003); no change in OS (37 versus 26 months; p = 0.05)	Berenson et al. [2002]
	Prednisone + interferon	Randomized trial (233 patients)	VAD versus VAD plus verapamil or quinine	No consolidation	Increase in PFS median, (19 versus 9 months for IFN; p = 0.008); no change in OS (57 months for IFN/P versus 46 months for IFN; p = 0.36)	Salmon et al. [1998]
	Dexamethasone	Randomized trial (307 patients)	Melphalan prednisone versus melphalan dexamethasone	No consolidation	Increase in PFS (2.76 years versus 1.97 years p = 0.000); no change in OS (3.86 years versus 3.65 years p = 0.74).	Shustik et al. [2004]
IMiDs	Thalidomide	Randomized phase III clinical trial (597 patients)	VAD combination chemotherapy of cyclophosphamide, etoposide, vinctistine, adriamycin and dexamethasone	Double ASCT with melphalan	Increase in PFS (p = 0.003) and OS (p = 0.04)	Attal et al. [2006]
		Randomized phase III clinical trial (668 patients)	VAD versus thalidomide, AD	Double ASCT with melphalan, followed by combination chemotherapy of cyclophosphamide, etoposide, vincristine, adriamycin and dexamethasone	Increase in PFS (56% versus 44%, p = 0.01); no difference in OS	Barlogie et al. [2006]
		Randomized phase III clinical trial (536 patients)	CTD, CVAD, MP	ASCT with melphalan	Increase in PFS (34 months versus 25 months p < 0.001); no difference in OS (60 months versus 73 months p = 0.77)	Lokhorst et al. [2010]
		Meta-analysis		No consolidation	Increase in PFS (23 versus 15 months; p < 0.001); no difference in OS (p = 0.40)	Morgan et al. [2012]
IMiDs	Lenalidomide	Randomized phase III clinical trial (614 patients)	VAD or bortezomib, dexamethasone	ASCT with melphalan	Increase in PFS (41 months versus 23 months p < 0.001); no difference in OS (80% in lenalidomide group versus 84% in the placebo p = 0.29)	Attal et al. [2012]
		Randomized phase III (460 patients)	Bortezomib, lenalidomide or thalidomide-based regimens in various combinations	ASCT with melphalan	Increase in PFS (39 months versus 21 months, p < 0.001); OS not reached	McCarthy et al. [2012]
Proteasome inhibitors	Bortezomib	Randomized phase III clinical trial (386 patients)	VBMCP/VBAD/B versus TD versus VTD	ASCT with melphalan	PFS longer with thalidomide/bortezomib compared with thalidomide alone and with alfa2-IFN (78% versus 63% versus 49%, respectively, at 2 years, p = 0.01); OS was not significantly different in the individual arms	Rosiñol et al. [2012]
		Randomized phase III clinical trial (627 patients)	VAD versus PAD	ASCT with melphalan	Increase in PFS (13–30 months; HR, 0.45; 95% CI, 0.26–0.78; p = 0.004) and OS (21–54 months; HR, 0.33; 95% CI, 0.16–0.65; p < 0.001), especially in high-risk patients	Sonneveld et al. [2012]

ASCT, autologous stem cell transplant; CTD, cyclophosphamide, thalidomide, and dexamethasone; MP, melphelan prednisone; VBMCP, vincristine, carmustine, melphalan, cyclophosphamide, prednisone; VBAD/B, vincristine, BCNU, doxorubicin, dexamethasone/bortezomib; VTD, bortezomib, thalidomide and dexamethasone; TD, thalidomide and dexamethasone; CI, confidence interval; PFS, progression-free survival; PAD, bortezomib, doxorubicin and dexamethasone; HR, hazard ratio; OS, overall survival; IFN, interferon; IMiDs, immunomodulating drugs.

Table 2.

Newer agents.

Class of agent	Type of agent	Comments	Induction	Consolidation	Efficacy	Reference or ClinicalTrials.gov identifier
Proteasome inhibitors	Ixazomib	Phase II study	Ixazomib, lenalidomide and dexamethasone	No consolidation	Deepening responses	Kumar et al. [2014]
		Phase III study	Proteasome inhibitor with or without immunomodulating drug (IMiD)-based regimens	ASCT with melphalan	Ongoing	NCT02181413
	Carfilzomib	Phase II study	CarCyDex induction	ASCT with melphalan versus four cycles of CarCyDex	Ongoing	NCT02315716
IMiIDs	Pomalidomide	Phase II study	Four cycles of clarithromycin, pomalidomide and dexamethasone (ClaPD)	ASCT with melphalan versus ClaPD for nine cycles	Ongoing	NCT01745588

ASCT, autologous stem cell transplant; CarCyDex, carfilzomib, cyclophosphamide and dexamethasone.

Interferon

In 1990, Mandelli and colleagues first published the use of interferon after induction chemotherapy, prolonging PFS [Mandelli et al. 1990]. The strongest benefit was seen in patients who have the best response to chemotherapy. However, it was in 1998 when a randomized control trial was undertaken where patients were randomized to interferon maintenance postautologous transplant versus no maintenance [Cunningham et al. 1998]. The duration of response was higher in the interferon arm. Even though the PFS and OS were both higher at 52 months, the PFS and OS at 77 months, though increased, were not significant. But it was a large randomized controlled study that showed no difference in PFS or OS [Barlogie et al. 2006a]. Given the toxicity of interferon and lack of survival data, it never became standard of care.

Glucocorticoids

Steroids alone have demonstrated efficacy in multiple myeloma for a long time [McIntyre et al. 1985; Alexanian et al. 1986]. Steroids work by suppressing cytokine production important in multiple myeloma. Thus, a few studies were also performed to evaluate the efficacy of steroids during maintenance [Berenson et al. 2002]. Steroids were initially studied in combination with interferon. Salmon and colleagues randomized patients to receive interferon with prednisone as maintenance versus interferon alone [Salmon et al. 1998]. Even though there was an increase in PFS in the prednisone group, there was no change in OS. Later, dexamethasone 40 mg for 4 days every 28 days after induction was compared with placebo after induction therapy. It showed an increase in PFS (2.7 years versus 1.9 years) but no change in OS [Shustik et al. 2004]. Given the lack of benefit on OS accompanied by side effects from steroids, it did not become routinely used as maintenance therapy.

Immunomodulatory agents

Thalidomide

Following the efficacy of thalidomide in patients with refractory multiple myeloma, thalidomide was studied as maintenance treatment following autologous transplantation with or without other agents. It was one of the first agents that showed an improvement; not only in PFS, but also OS. Attal and colleagues initially found a higher survival rate amongst patients randomized to the thalidomide arm [Attal et al. 2006]. Thalidomide maintenance was instituted after induction with the chemotherapy regimen VAD (vincristine, doxorubicin, and dexamethasone) followed by tandem autologous stem cell transplant (ASCT). But due to side effects, the drug was discontinued in 39% of the patients; the main one being peripheral neuropathy. Barlogie and colleagues also studied thalidomide in the maintenance setting, but in their study, thalidomide was used during induction, consolidation and maintenance [Barlogie et al. 2006b]. Even though this study showed better CR rates, better event-free survival in the thalidomide maintenance arm, there was no OS observed, likely due to the dose of thalidomide and using it during all the phases. In 2009, HOVON conducted a phase III randomized trial in which patients were randomized to receive induction as well as maintenance with thalidomide versus no thalidomide, again, showing an increase in PFS but no increase in OS [Lokhorst et al. 2010]. A meta-analysis conducted by Morgan et al. confirmed these findings that showed an increase in PFS in the thalidomide group, but with no increase in OS, and worse side effects [Morgan et al. 2012]. With no consistent benefit in OS and a significantly worse adverse-effect profile, thalidomide maintenance never became standard of care.

Lenalidomide

Given the efficacy of thalidomide, it was only natural to try lenalidomide, a derivate of thalidomide, a more effective and less toxic agent in the maintenance setting as well. It has been shown to be effective in both frontline and relapsed refractory settings alone and as a part of various different regimens [Rajkumar et al. 2005; Dimopoulos et al. 2007]. Two main studies were conducted by IFM 05-02 and CALGB 100104, respectively, in regards to maintenance lenalidomide postautologous transplant. While they both showed an increase in PFS, only CALGB 100104 showed an increase in overall survival [Attal et al. 2012; McCarthy et al. 2012]. The IFM 2005-02, a phase III study, assessed 614 patients and randomized them to receive lenalidomide maintenance post autologous transplantation at the dose of 10 mg/day for 3 months which could be increased to 15 mg/day if tolerated versus placebo. It was continued until disease progression.

There were few important differences between the two trials. The PFS was significantly higher in the lenalidomide group at 41 months versus 23 months. There was no difference in overall survival at 4 years or 5 years [Lauwers-Cances et al. 2013]. Both groups received consolidation with lenalidomide. But in the lenalidomide group, there was an increased incidence of secondary malignancies at 2.3 versus 1.3 per 100 patient years. Most patients in this study received induction with Velcade, dexamethasone (VAD) or VD.

CALGB 100104, also a phase III study, assessed 460 patients under the age of 70 who had stable disease or some response at 100 days postautologous transplantation, and randomized them to lenalidomide maintenance versus placebo until disease progression. Unlike IFM 2005-02, most patients received an induction regimen with an immunomodulating drug (IMiD). This trial again showed an increase in PFS at 46 versus 27 months. Unlike the IFM 2005-02, the percentage of patients alive at 34 months was 85% in the lenalidomide group versus 77% in the placebo group. This survival advantage persisted at 4 years at 80% versus 70% [McCarthy et al. 2013]. The incidence of second malignancies was again higher in the lenalidomide group at 4.3% versus 2.2%.

Apart from the induction regimens, other important distinctions between the two trials were

1. the presence of lenalidomide consolidation in the IFM 2005-02 trial versus CALGB 100104;

2. 21% patients had two autologous transplants in the IFM 2005-02 versus 100% had just one autologous transplant in CALGB 100104;

3. there was an increased incidence of acute myeloid leukemia (AML)/ myelodysplastic syndrome (MDS) seen in patients in the CALG 100104 trial that was not seen in the IFM 2005-02 trial;

4. maintenance was stopped at a median of 32 months in the IFM 2005-02 trial and it was continued in the CALGB 100104.

The results of the meta-analysis presented at American Society of Clinical Oncology (ASCO) of these three trials did show that the OS in the lenalidomide group had not been reached versus 86 months with no maintenance. The survival was statistically significant at 5, 6 and 7 years [Michel et al. 2016].

Pomalidomide

Pomalidomide was found to be one of the most potent IMiDs in preclinical studies [Koh et al. 2005]. A phase II study initially showed the efficacy of pomalidomide with dexamethasone in patients with relapsed refractory disease [Lacy et al. 2009]. This led to a phase III study which randomized 302 patients to receive pomalidomide with low dose dexamethasone or high dose dexamethasone. It was found that the median PFS was 4 months versus 1.9 months (p < 0.0001). This led to the approval of pomalidomide in the relapsed refractory setting [San Miguel et al. 2013]. There is currently a trial ongoing with the use of pomalidomide in the maintenance setting. In this phase II study all patients start by receiving four cycles of clarithromycin, pomalidomide and dexamethasone (ClaPD). After four cycles, half of the patients will undergo an ASCT followed by pomalidomide (group 1). The other half of the patients will continue to receive ClaPD for nine cycles to be followed by pomalidomide maintenance (group 2). The study is currently not recruiting and the results are awaited [ClinicalTrials.gov identifier: NCT01745588].

Proteasome inhibitors

Bortezomib

Given the efficacy of bortezomib in patients with high-risk disease, several trials were conducted to study the efficacy of bortezomib post-transplantation. It has also been shown that patients with high free-light chains at relapse have a very poor prognosis, possibly also making them high risk [Chilkulwar et al. 2016]. The Spanish Myeloma Group conducted a phase III trial where patients were randomized to three different induction regimens: VTD (bortezomib, thalidomide and dexamethasone), TD (thalidomide and dexamethasone) and vincristine, BCNU, melphalan, cyclophosphamide, prednisone/vincristine, BCNU, doxorubicin, dexamethasone/bortezomib [Rosiñol et al. 2012]. All the groups underwent autologous transplantation with melphalan conditioning. After 3 months, patients were then randomized to receive maintenance with interferon alpha or thalidomide, or thalidomide plus bortezomib. The PFS at 2 years was higher in the group that received maintenance with bortezomib plus thalidomide versus thalidomide versus interferon alpha (78% versus 63% versus 49%, respectively).

In the HOVON-65 trial/GMMG-HD4, a randomized phase III trial was undertaken [Sonneveld et al. 2012]. In the trial, 827 patients were randomized to different induction regimens, VAD: vincristine, doxorubicin and dexamethasone, or PAD: bortezomib, doxorubicin and dexamethasone. The two groups underwent autologous transplantation with melphalan conditioning. The patients who received induction with VAD received maintenance with thalidomide, while patients in the PAD group received maintenance with bortezomib. The maintenance was continued for 2 years. The PFS was significantly improved in patients receiving bortezomib maintenance versus thalidomide maintenance (28 versus 35 months). In high-risk patients, such as patients with 17p deletion, both PFS (12 versus 22 months) and OS (24 versus 54 months) was significantly better in the bortezomib arm. Even patients with high creatinine (>2) had a significantly better PFS (13 versus 30 months) and OS (21 versus 54 months). In this trial, it is hard to discern the benefit of maintenance, as the patients in both arms received different induction regimens. It could be that the benefit seen in the bortezomib arm is also due to the induction regimen, which contains bortezomib, as the CR rates seen in patients who received PAD were much higher than those seen in patients who received VAD.

Ixazomib

Ixazomib is an oral proteasome inhibitor (PI) recently approved in combination with lenalidomide and dexamethasone for induction in patients with relapsed refractory multiple myeloma [Moreau et al. 2015]. In a phase II study, 50 patients were enrolled to receive induction with ixazomib, lenalidomide and dexamethasone for six cycles. Patients who were transplant eligible then moved off the study for autologous transplantation. Of these, 21 patients moved on to receive maintenance with ixazomib at 3.6 mg weekly. It was well tolerated for a median of 1.5 years and 33% of the patients also achieved deeper responses. The median duration of response was 26.5 months [Kumar et al. 2014]. A phase III trial is currently underway to evaluate the efficacy of ixazomib as maintenance postautologous transplantation. The study requires standard-of-care induction therapy (induction therapy must include PI with or without IMiD-based regimens as primary therapy for multiple myeloma), followed by a single ASCT with a high-dose melphalan (200 mg/m²) conditioning regimen, within 12 months of diagnosis. Maintenance post-transplant is with ixazomib citrate 3 mg on days 1, 8 and 15 of each 28-day cycle, for up to 26 cycles (approximately 24 months) [ClinicalTrials.gov identifier: NCT02181413].

Carfilzomib

Carfilzomib is a more potent PI and was found to be an irreversible inhibitor of the ubiquitin−proteasome pathway [Jakubowiak et al. 2012]. Carfilzomib has shown efficacy in patients with relapsed refractory myeloma [Siegel et al. 2012]. In the trial by Siegel and colleagues, 792 patients were randomized to receive either lenalidomide with dexamethasone in combination with carfilzomib, or alone. The median PFS was 26.3 months in the carfilzomib group versus 17.6 months with lenalidomide−dexamethasone alone (p = 0.0001). It is currently approved for the treatment for relapsed refractory patients. At the Mayo Clinic, a risk-stratification model termed Mayo Stratification of Myeloma and Risk-Adapted Therapy has been developed to guide treatment for patients based on their fluorescent in situ hybridization (FISH) results. In high-risk patients, t(14;16), t(14;20), 17p deletion, the recommendation is to use Carfilzomib, Revlimid, dexamethasone (KRd) for four cycles as induction, followed by maintenance treatment with carfilzomib or bortezomib for at least 1 year. Trials are currently ongoing to evaluate the efficacy of carfilzomib as a maintenance agent. The Cardamon trial is a phase II trial using carfilzomib/cyclophosphamide/dexamethasone with maintenance carfilzomib in untreated transplant-eligible patients with symptomatic multiple myeloma. Patients are randomized after induction to receive either four further cycles of carfilzomib, cyclophosphamide and dexamethasone (CarCyDex) treatment, followed by 18 months of maintenance carfilzomib or melphalan-conditioned ASCT, followed by 18 months of maintenance carfilzomib [ClinicalTrials.gov identifier: NCT02315716].

Discussion

Currently, we have data for use of thalidomide, lenalidomide and bortezomib as maintenance agents. Given the toxicity of thalidomide, lenalidomide had largely replaced its use. It has been proven multiple times that maintenance therapy improves PFS. The data on prolonging OS is not consistent. However, the recent meta-analysis presented at ASCO did show a benefit towards OS. Patients who achieved ⩽PR post-ASCT benefited from lenalidomide (HR = 0.86; 95% CI, 0.65–1.15) as well as patients with CR/very good partial response (VGPR) (HR = 0.70; 95% CI, 0.54–0.90). OS benefit was generally consistent across subgroups. Heterogeneity test showed significant difference across trials (p = 0.047). We should be selective in choosing the patients in which we should use maintenance therapy. It should definitely be used in patients with a higher chance of relapse postautologous transplantation. Patients who haven’t achieved complete remission going into transplantation and patients with high-risk disease are the most likely to benefit from maintenance, as we know that patients with MRD negativity at day 100 post-transplantation have a better outcome than patients who don’t. The influence of cytogentics should also be considered when starting patients on lenalidomide maintenance. In the IFM study of maintenance lenalidomide versus no maintenance, an increase in PFS was observed with lenalidomide maintenance in the overall study population (median 42 versus 24 months, p < 0.0001) and in patients with del 17p (median 29 versus 14 months, p < 0.02), but not in patients with t(4;14) (median 28 versus 24 months, p < 0.04; based on analysis presented at the annual meeting of the American Society of Hematology in 2010). In this setting, the prolongation of PFS with lenalidomide maintenance appears less than other subgroups [Attal et al. 2016].

Given the efficacy of bortezomib in patients with high-risk disease, PIs should be used as maintenance in high-risk disease for at least a year after consolidation, with the data we have currently. With the advent of newer agents such as ixazomib and carfilzomib, they will likely become the standard of care for these patients in the future once we have enough trial data to support their use in this setting.

In other settings where maintenance is required such as patients not in CR but without high-risk disease by FISH, one could use lenalidomide maintenance after taking into account the long-term adverse of these agents such as second malignancies and balancing the risk−benefit ratio.

Questions still to be answered are: what the impact of maintenance on OS in different population subsets may be, how and when MRD negativity should be incorporated into practice and the exact role of the novel agents in the maintenance setting.

Relapse after autologous stem cell transplant occurs over the course of time in almost all patients with multiple myeloma. Whether relapse patterns are changing in the era of lenalidomide or bortezomib maintenance has yet to be established. Moreover, when patients have asymptomatic relapse/progression on maintenance therapy, it is unclear whether the patient should be continued on their current maintenance regimen, whether the dose should be escalated, whether a different regimen should be used, or even whether the patient should be taken off maintenance and simply observed until the time of clinical progression. With the approval of several novel agents in the past few years there are several options available for treatment of relapsed disease. The three main approaches for management of symptomatic relapse after initial ASCT are reinduction followed by salvage ASCT, reinduction followed by allogeneic SCT, or most commonly, reinduction with continuation of conventional-dose chemotherapy using rational combinations of novel therapies for relapsed/refractory disease. There is no clear-cut benefit of one salvage regimen over the other. In the current era, the choice of chemotherapy needs to be individualized based on prior therapies’ side-effect profile and patients’ comorbidities. Even with the approval of numerous drugs for relapsed multiple myeloma, reinduction followed by salvage ASCT might still have a role to play in the relapsed setting, especially in the setting of prolonged remission from the first transplant. There are also several trials being conducted using immunotherapy and chimeric-antigen-receptor-T-cell-therapy (CAR-T) cells in relapsed myeloma. Where these newer modalities of treatment fit in the current paradigm of relapsed myeloma needs to be further investigated.