Introduction

Autologous stem-cell transplantation (ASCT) is considered as the standard of care for transplant-eligible patients with newly diagnosed multiple myeloma (NDMM) [1]. In the past years, the efficacy of induction therapy prior ASCT has been strongly improved, and triplet combination containing both proteasome inhibitor and immunomodulatory agent is now considered as the standard of care [2]. The triplet combination bortezomib, thalidomide, and dexamethasone (VTD) demonstrated strong efficacy and is approved for the treatment of transplant-eligible NDMM patients [3, 4]. The induction regimen bortezomib, lenalidomide, and dexamethasone (VRD) resulted in high response rate with favorable safety profile and is also considered as a standard of care prior ASCT for de novo MM patients [1, 5, 6]. In addition to improve response rate, induction therapy should preserve an adequate stem-cell collection in order to ensure the feasibility of ASCT. In the present study, we retrospectively analyzed stem-cell collection performed in our institution for newly diagnosed myeloma patients who received either VTD or VRD induction therapy between 2009 and 2019.

Material and methods

Patients and induction therapy

We conducted a retrospective institutional review board-approved analysis of all consecutive patients who fulfilled the following criteria: (i) diagnosis of symptomatic multiple myeloma according to international criteria [7], (ii) eligible for high-dose melphalan followed by ASCT, (iii) treated with at least three cycles of VTD or VRD induction, and (iv) received stem-cell mobilization. Patients treated in the context of clinical trials were excluded from the study. VTD induction consisted of bortezomib 1.3 mg/m2 SC (day 1, 4, 8,and 11), thalidomide 100 mg/day, and dexamethasone 20 mg (day 1–2, 8–9, J15–16, and 22–23) (28-day cycle). VRD induction consisted of bortezomib 1.3 mg/m2 SC (day 1, 4, 8, and 11), lenalidomide 25 mg (day 1–14), and dexamethasone 20 mg (day 1–2, 4–5, 8–9, and 11–12) (21-day cycle).

Stem-cell mobilization and harvesting

Stem-cell mobilization consisted of intravenous cyclophosphamide (3 g/m2) followed by G-CSF administered at a dose of 10 μg/kg/day subcutaneously starting 24 h after cyclophosphamide administration until peripheral blood stem-cell collection (planned at day 10 post cyclophosphamide). Apheresis was performed at least 3 weeks after the last dose of lenalidomide. Apheresis for peripheral blood stem-cell harvesting was performed until an optimal harvest of ≥5.106 CD34/kg [8]. Plerixafor (0.24 mg/kg) was administered preemptively to rescue mobilization in patients with blood CD34-positive cells <15/mm3 at time of apheresis. Collection failure was defined by stem-cell harvest <2.106 CD34/kg. Stem-cell harvesting was performed at Etablissement Français du Sang, Nantes, France.

Statistical analysis

The Fisher test was used to compare parameters between VTD and VRD groups. All p values were two-tailed. The significance level was set at 5% and statistical analyses were carried out using R Software version 3.0 R Development Core Team (2015). R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3-900051-07-0, http://www.R-project.org/.

Results

Patient characteristics

From January 2009 to December 2019, 325 patients with previously untreated symptomatic MM who received at least three cycles of VTD or VRD before stem-cell mobilization were identified. The characteristics of patients are summarized in Table 1. Two hundred and forty-two (74.5%) patients received VTD induction and 83 (25.5%) patients received VRD induction. The median age (61, range 25–71) was similar in both groups. Reflecting the change of practice over the past years, VTD was the predominant induction regimen before 2018 and VRD induction was used instead of VTD. Patients received a median number of four cycles before stem-cell collection.

Table 1 Characteristics of patients.
Full size table

Plerixafor use and apheresis

All patients received stem-cell mobilization with cyclophosphamide (3 g/m2) plus G-CSF. Apheresis was planned 10 days after cyclophosphamide. Plerixafor was added to rescue the mobilization failure in 29 (8.9%) patients (Table 1). The plerixafor use was more frequent in patients receiving VRD induction (n = 16/83, 19.3%) in comparison with patients receiving VTD (n = 13/242, 5.4%), p = 0.0004) (Fig. 1). Moreover, 33/83 (42.3%) patients needed at least two apheresis to reach adequate stem-cell collection after VRD induction, versus 73/242 (30.2%) patients after VTD induction (p = 0.05).

Fig. 1: Impact of VTD or VRD induction on stem-cell collection.
figure 1

a Impact of VTD or VRD induction on Plerixafor use and number of apheresis. b Impact of VTD or VRD induction on stem-cell yield. VTD bortezomib, thalidomide, dexamethasone, VRD bortezomib, lenalidomide, dexamethasone.

Full size image

Stem-cell collection failure and stem-cell yield

Six (1.8%) patients experienced stem-cell collection failure, 5/83 patients (6%) in the VRD group, and 1/242 patient (0.4%) in the VTD group (p = 0.004). The median number of CD34-positive cells was slightly higher in the VTD group: 9.3 versus 8.5 × 106 in the VRD group (p = 0.05). Among patients with autologous stem-cell harvest >2.106 CD34/kg (n = 319), stem-cell harvest was insufficient for tandem transplantation (>4 × 106 CD34/kg) in two patients (2.6%) in the VRD arm versus six patients (2.5%). Ninety-eight percent and ninety-three percent of patients actually underwent ASCT in the VTD and VRD group, respectively.

Discussion

To the best of our knowledge, the present real-life single-center retrospective study is the first to compare autologous stem-cell collection following either VTD or VRD induction therapy in patients with de novo MM. Despite the biases inherent to the retrospective design of the study, all patients received homogenous mobilization using cyclophosphamide (3 g/m2) plus G-CSF, which is considered as a standard of care [9, 10]. Plerixafor has been shown to improve stem-cell mobilization in patients with multiple myeloma [11]. Stem-cell mobilization with cyclophosphamide (1.5 g/m2) plus G-CSF versus plerixafor plus G-CSF in the era of novel agents has been recently compared [12]. In this study from Chaudhary and colleagues, mobilization with plerixafor plus G-CSF was found to be more effective in terms of CD34-positive cells collection. In our institution, a preemptive use of plerixafor in addition with Cyclophosphamide + G-CSF was performed. The efficacy of such an approach has been reported [13]. Our study confirmed that stem-cell collection following both VTD and VRD induction allowed the completion of ASCT in the vast majority of patients (98% and 93%, respectively). The feasibility of stem-cell collection after VTD or VRD induction has been confirmed in the context of phase-3 studies [1, 4]. However, our study revealed that VRD induction was associated with a higher plerixafor use with a 4:1 ratio in comparison with VTD, which imply a cost increase for the procedure, as recently published [14, 15]. Moreover, stem-cell collection after VRD induction required a higher number of apheresis to reach adequate stem-cell collection. The median number of CD34-positive cells obtained in the VRD group was also found to be inferior in comparison with the VTD group. Importantly, stem-cell mobilization failure was more frequent in patients who received VRD induction. The triplet bortezomib, cyclophosphamide and dexamethasone (VCD) is considered as a standard of care prior ASCT [6]. Recently, Skerget and colleagues demonstrated similar stem-cell collection following VCD or VTD induction, in terms of CD34-positive cell yields and rate of mobilization failure [16]. Stem-cell collection impairment following lenalidomide-based induction has been previously reported [17,18,19]. In addition with its myelotoxicity, lenalidomide exposure has been associated with suppression of myeloid progenitors in collected graft [20]. The triplet combination carfilzomib–lenalidomide–dexamethasone (KRD) demonstrated strong efficacy as induction therapy prior ASCT [21]. Interestingly, Bal and colleagues recently showed that KRD induction resulted in a higher rate of collection failure and lower rate of stem-cell yields in comparison with VRD [22]. Recently, the anti-CD38 monoclonal antibody daratumumab has been approved in combination with VTD in newly diagnosed myeloma patients who are eligible for ASCT, based on the results of the phase-3 trial Cassiopeia [23]. In a phase-2 randomized study, the addition of daratumumab to VRD induction has been shown to significantly improve response rate [24]. However, the addition of daratumumab to VRD also resulted in lower rate of CD34-positive cells collection [24]. The phase-3 randomized trial EMN-02 recently demonstrated a progression-free and overall survival benefit favoring tandem ASCT versus single ASCT, especially for patients with high-risk cytogenetic [25]. This survival advantage following a tandem-transplant approach for selected MM patients reinforces the need of an optimal stem-cell collection. In the future context of lenalidomide and/or anti-CD38-based induction, attention should be paid to the optimal timing of stem-cell collection. Indeed, stem-cell harvesting should ideally be planned after 3– 4 cycles of induction, as suggested by the recent data issued from the German CONCEPT trial (NCT03104842) [26]. Moreover, mobilization with cyclophosphamide + G-CSF should be systematically considered [27].

To conclude, this real-life study confirms the possibility to obtain adequate autologous stem-cell collection for the vast majority of MM patients after VRD induction, even if more patients experienced collection failure in comparison with VTD. Moreover, VRD induction required a more frequent use of plerixafor and a higher number of apheresis, implying time, and cost-consuming consequences. The use of tandem transplantation for selected patients and the increasing addition of anti-CD38 monoclonal antibody during induction highlight the need of an optimal stem-cell collection strategy.