Although high-dose therapy is considered the standard therapy for younger patients with multiple myeloma (MM), the advantages of performing a second transplant remain debated. The current study was conducted to evaluate the efficacy and the feasibility of a front-line double transplant program in young MM patients.
A total of 153 MM patients aged ≤65 years, the majority of whom had stage III disease (65%), were enrolled in a multicenter, nonrandomized, high-dose program including 2 transplants.
The percentage of good quality responses (complete and very good partial responses) increased from an initial 33% after induction to 91% (complete response rate of 29%) after 2 transplants. However, this increase in response did not produce an advantage in either event-free survival or overall survival, even when the analysis was performed grouping patients by response. The protocol was well tolerated and no difference in transplant-related mortality was observed between the first and second transplants. A first transplant was performed in 122 of 153 patients (80%), and 65 (42% of the enrolled patients) completed the double transplant program. Reasons for not undergoing the second autologous transplant were death (n = 2 patients), insufficient peripheral blood stem cells (n = 8 patients), severe transplant-related toxicity (n = 10 patients), allotransplants (n = 9 patients), early progression after first transplant (n = 6 patients), lost to follow-up (n = 3 patients), and patient refusal (n = 19 patients).
High-dose therapy (HDT) represents the gold standard treatment for patients with newly diagnosed multiple myeloma (MM), but to our knowledge, there is no definite agreement regarding the need for a second transplant. Two randomized trials by Attal et al.1 and Child et al.2 demonstrated a better event-free survival (EFS) and overall survival (OS) after HDT compared with conventional therapy (CC). These results have not been confirmed by other trials. Fermand et al.3 and Barlogie et al.4 observed a benefit in terms of EFS, but no difference in OS between patients treated with HDT and those treated with CC. In contrast, in the Spanish experience, in which patients who were responsive to initial chemotherapy were randomized to receive further CC or HDT, Blade et al.5 confirmed better complete response (CR) rates with HDT than with CC but found no differences in terms of EFS and OS between the 2 strategies. Given the impact of CR on outcome,6 Barlogie et al.7 explored the possibility of further increasing response rates by intensifying the treatment through the introduction of a second transplant; the EFS and OS at 12 years after this Total Therapy I strategy were found to be better than those after CC.8
Therefore, although there is a general agreement regarding the treatment of young MM patients with HDT, it remains unclear whether a second transplant adds anything to the first procedure. In the IFM94 study, Attal et al.9 found a clear advantage in terms of EFS and OS after a tandem transplant, with the best benefit from the second transplant occurring in patients with less than a good partial response after the first procedure. Conversely, preliminary results of other randomized trials did not demonstrate a clear survival benefit for the patients undergoing double transplants,10–12 raising the issues of additive costs, toxicity, and efforts to complete all the scheduled programs.13–17
The objective of the current study was to evaluate the feasibility of a double transplant program in the context of front-line high-dose chemotherapy in MM patients aged ≤65 years.
MATERIALS AND METHODS
From January 2000 to December 2004, 153 newly diagnosed MM patients aged ≤65 years in disease stages II, III, or I in progression according to the Durie and Salmon staging system and who were eligible for high-dose therapy were consecutively enrolled in a high-dose chemotherapy program including 2 tandem autotransplants.
There were 79 males and 74 females with a median age of 55 years (range, 35–65 years); 26 patients (18%) were in stage I in progression (13 patients progressed from a previous monoclonal gammopathy of undetermined significance, the remaining 13 patients had increases of M component and bone marrow plasma cell infiltration), 24 (17%) were in stage II, and 103 were in stage III disease (65%). Table 1 summarizes the characteristics of the patients at the time of enrollment.
Table 1. Main Clinical Characteristics and Prognostic Factors of 153 Patients with MM
No. of patients
MM indicates multiple myeloma; Ig, immunoglobulin.
Median age, y (range)
Stage of disease
IA in progression
103 (90/13) (67%)
Monoclonal component type
Chromosome 13 deletion
Median albumin (g/dL) (range)
Median C-reactive protein (mg/dL) (range)
Median lactate dehydrogenase (IU/L) (range)
The protocol outlined in Figure 1 included: 1) 2 pulse-VAD cycles (combined intravenous bolus of vincristine [2 mg i.v.] and doxorubicin [5 mg i.v.] given on Day 1 with high-dose oral dexamethasone [40 mg] given on Days 1–4 and 14–17 of each 28-day cycle); 2) 2 DCEP cycles (combined infusion of cyclophosphamide 700 mg/m2, etoposide 100 mg/m2, and cisplatin 25 mg/m2 given on Days 1–2 associated with high-dose dexamethasone on Days 1–4 of each 28-day cycle), each followed by peripheral blood stem cell (PBSC) mobilization with granulocyte–colony-stimulating factor (5 mcg/kg) to collect at least 8 × 106 CD34+cells/kg for the 2 procedures; 3) a transplant phase consisting of 2 tandem procedures, both conditioned with melphalan at a dose of 200 mg/m2 (HD-MPH), separated by 3 to 6 months. No maintenance was scheduled after the second transplant.
Response was evaluated as follows: CR required negative immunofixation of urine and serum as well as a normal bone marrow aspirate; very good partial remission (VGPR) required at least a 90% decrease of both the monoclonal component (MC) in serum and urine and infiltration of bone marrow plasma cells (BMPC); partial remission (PR) required a minimum 25% decrease of both serum and urine MC and BMPC infiltration; stable disease indicated a reduction of MC in the serum and urine and/or BMPC infiltration of <25%; and progression was defined as the reappearance of MC or BMPC infiltration or as an MC or BMPC increase of 50% at 2 subsequent evaluations. CR and VGPR were considered together as good quality responses; stable disease and progression were considered together as limited disease control. Response rates were evaluated at the end of each phase.
Numeric variables were summarized by their median and range. Categoric variables were described by counts and relative frequencies. Any association between type of response to the first transplant and the number of transplants was tested using the chi-square test for tables. The Kaplan-Meier product-limit method was used to estimate survival curves, and the Gehan Wilcoxon test was adopted to perform comparisons between different groups of patients. For each patient, the OS was calculated as the time between the start of follow-up (either the date of diagnosis or the date of the first transplant) and the date of death or last follow-up for censored cases. EFS was calculated as the time between the start of follow-up and the date of disease progression or death, or last follow-up for censored cases.
Table 2 reports the responses after each phase demonstrating that the percentage of good quality responses (CR+VGPR) improved from the initial phase to the second transplant procedure, with a final rate of 91% (CR of 29%) achieved in patients completing the protocol as planned. At the time of the current analysis, after a median follow-up of 30 months (range, 3–78 months) the median EFS was 31 months, and the median OS had not yet been reached (Fig. 2). Patients with less than a PR after the first transplant (stable disease or progression) were found to have a worse EFS (P = .001) and OS (P = .015) compared with patients with at least a PR. Conversely, no statistically significant differences in EFS and OS were observed between patients with a good quality response and those with a PR (Fig. 3A and 3B). We compared the outcome of patients who received only 1 transplant (n = 57 patients) with that of the patients who completed the double transplant program (n = 65 patients) as originally planned. The clinical characteristics and response rates to the first transplant were comparable between the patients in the 2 groups (P = .2). The EFS (P = .549) and OS (P = .228) (Fig. 4A and 4B) were found to be similar in the 2 groups, even when the analysis was performed according to the type of response reached after the first transplant.
Table 2. Response Rate at the Completion of Each Phase of the Protocol
Type of response (%)
1st HD- MPH
VAD indicates combined intravenous bolus of vincristine and doxorubicin on Day 1 with high-dose oral dexamethasone administered on Days 1–4 and Days 14–17 of each 28-day cycle; DCEP, combined infusion of cyclophosphamide, etoposide, and cisplatin on Days 1–2 associated with high-dose dexamethasone on Days 1–4 of each 28-day cycle; HD-MPH, melphalan given at a dose of 200 mg/m2.
Good quality response
Very good partial remission
Partial disease control
Limited disease control
Feasibility of the Program
Figure 1 shows the patient-flow throughout the protocol and summarizes the reasons for withdrawal at each step. Among the 153 patients enrolled into this study, 31 patients (20%) dropped out before the first transplant. Three patients dropped out after pulse-VAD (1 because of progression and 2 were lost to follow-up) and 28 patients dropped out after DCEP for the following reasons: progression (n = 11 patients), severe toxicity (n = 1 patient), insufficient PBSC mobilization (n = 7 patients), allogeneic transplant (n = 4 patients), withdrawal of consent (n = 2 patients), and missing follow-up (n = 3 patients).
In all, 122 patients (80% of the enrolled patients) entered the transplant phase and of these, only 65 (42% of the enrolled patients) completed the double transplant program. Reasons for not undergoing the second autologous transplant were death during the first procedure (n = 2 patients), insufficient PBSC for a second transplant (n = 8 patients), severe transplant-related toxicity (n = 10 patients), allogeneic transplant from a well-matched human leukocyte antigen (HLA) donor (n = 9 patients), early progression after high-dose melphalan (n = 6 patients), patient lost to follow-up (n = 3 patients), and patient refusal of the second transplant (n = 19 patients).
No therapy-related mortality was observed during the first 2 phases of the program (2 pulse-VAD cycles and 2 DCEP cycles). The most common adverse events noted during pulse-VAD were infections (8%), constipation (4%), and peripheral neuropathy (2%). A comparable limited toxicity was observed with the DCEP chemotherapy, with infections occurring in 10% of the patients and nausea and constipation in <1%.
The transplant-related mortality was limited, with no difference noted between the 2 transplants; 2 patients died during the first procedure and 1 during the second transplant. Toxicities were the same during the first and the second transplant procedures: oral mucositis (14%), nausea and emesis (5%), diarrhea (9%), and infections (7%).
The scenario of therapy in MM patients in the last 15 years has been largely influenced by the introduction of autologous transplantation. Some studies demonstrated a superiority of HDT compared with CC both in terms of EFS and OS1, 2; other experiences3, 4 confirmed a better EFS but found no difference in OS for patients randomly assigned to HDT; and finally, Blade et al.,5 despite the improved response rate achieved with HDT, found no difference in outcome when considering both EFS and OS in terms of transplantation versus CC. Moreover, to our knowledge, no clear advantage has been shown in performing 2 tandem transplants compared with only 1.7–12 Given the complexity of an HDT schedule including a double transplant, 1 important emerging issue is to define patients who could benefit from 2 procedures.
We performed this study on a cohort of 153 MM patients who were younger than 65 years and were enrolled in a high-dose program including 2 transplants with the aim of evaluating not only the efficacy of this therapeutic strategy but also its real feasibility.
In keeping with data in the literature,7–12 the response rate improved throughout the protocol. In particular, the percentage of patients who achieved a CR+VGPR after 1 transplant was 68% (CR rate of 19%) and this figure rose to 91% (CR rate of 29%) after the second procedure. Nevertheless, these results did not lead to longer EFS and OS in patients who received 2 transplants compared with those who underwent only 1 transplant (Fig. 4), even when the analysis was performed on the basis of the type of response after first transplant. These findings are in contrast with those of other studies reporting an advantage from performing a second transplant in those patients not achieving a CR after the first procedure.9, 10
Considering the feasibility of this double transplant program, 31 patients (20%) did not reach the first transplant, mainly because of disease progression or failed collection of PBSC (Fig. 1). This is consistent with previous studies that reported a comparable withdrawal rate of 20% to 25% due to disease progression occurring before transplantation.7–11 The patient flow increased after the first transplant; of the 122 patients who underwent the first transplant (80% of those enrolled), only 65 patients (42% of the enrolled and 53% of those who underwent 1 transplant) underwent the second procedure, revealing the limiting complexity of the entire program, together with no clear advantage in terms of outcome. This is in accordance with data previously reported by Rosinol et al.12 In the PETHEMA experience, only 45% of the patients referred to a tandem transplant program completed the assigned therapy. Both in our study and in the Spanish report, the readiness of patients to interrupt treatment was the most frequent reason for not continuing the program after the first transplant (Fig. 1). This most likely is due to the good performance status and the good quality of life patients usually achieve after the first procedure, with a consequent refusal to receive further treatment.
In conclusion, although the indication for HDT in younger patients appears clear, the role of a second transplant remains undefined. The results of the current study did not demonstrate an advantage from undergoing a second procedure, independently of the quality of response obtained with the first transplant. Therefore, the lack of a clear advantage in terms of efficacy, the high percentage of dropouts observed throughout the program, and the additive toxicity and costs of the second transplant appear to favor the strategy of a single transplant as the initial management program in younger MM patients.