Fax: (216) 444-9464
Pegylated liposomal doxorubicin, vincristine, and dexamethasone provide significant reduction in toxicity compared with doxorubicin, vincristine, and dexamethasone in patients with newly diagnosed multiple myeloma
A Phase III multicenter randomized trial
Article first published online: 10 JAN 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 4, pages 848–858, 15 February 2006
How to Cite
Rifkin, R. M., Gregory, S. A., Mohrbacher, A. and Hussein, M. A. (2006), Pegylated liposomal doxorubicin, vincristine, and dexamethasone provide significant reduction in toxicity compared with doxorubicin, vincristine, and dexamethasone in patients with newly diagnosed multiple myeloma. Cancer, 106: 848–858. doi: 10.1002/cncr.21662
- Issue published online: 3 FEB 2006
- Article first published online: 10 JAN 2006
- Manuscript Accepted: 1 SEP 2005
- Manuscript Revised: 29 JUL 2005
- Manuscript Received: 12 MAY 2005
- ALZA Corporation (Mountain View, CA)
- Tibotec Therapeutics, Division of Ortho Biotech Products, LP (Bridgewater, NJ)
- drug resistance;
- drug toxicity;
- multiple myeloma;
- pegylated liposomal doxorubicin;
- neutropenic fever
Pegylated liposomal doxorubicin has pharmacologic and safety advantages over conventional doxorubicin.
For this noninferiority trial, 192 patients with newly diagnosed, active multiple myeloma were randomized to receive either combined pegylated liposomal doxorubicin (40 mg/m2) and vincristine (1.4 mg/m2; maximum, 2.0 mg) as an intravenous infusion on Day 1 plus reduced-dose dexamethasone (40 mg) orally on Days 1–4 (DVd) (n = 97 patients) or combined vincristine (0.4 mg per day) and conventional doxorubicin (9 mg/m2 per day) as a continuous intravenous infusion on Days 1–4 plus reduced-dose dexamethasone (VAd) (n = 95 patients) for at least 4 cycles. Treatment was repeated every 4 weeks until patients either achieved maximal response, disease progression, or unacceptable toxicity or underwent transplantation. The primary endpoints were response and toxicity.
Objective response rates (DVd, 44%; VAd, 41%), progression-free survival (hazard ratio, 1.11; P = 0.69), and overall survival (hazard ratio, 0.88; P = 0.67) were similar between the treatment groups. However, DVd was associated with significantly less Grade 3/4 neutropenia or neutropenic fever (10% vs. 24%; P = 0.01), a lower incidence of sepsis, and less antibiotic use. Compared with VAd, DVd also significantly decreased the need for central venous access (P < 0.0001) and growth-factor support (P = 0.03) and resulted in less alopecia (20% vs. 44%; P < 0.001) but more hand–foot syndrome (25% vs. 1%; P < 0.001), mainly Grade 1/2.
The DVd regimen demonstrated similar efficacy with less toxicity and supportive care compared with VAd, which should improve clinical utility and optimize the opportunity for transplantation. Cancer 2006. © 2006 American Cancer Society.
An estimated 15,270 patients will be newly diagnosed with multiple myeloma in 2004, and an estimated 11,070 patients will die of their disease.1 For many years, the mainstay of treatment for multiple myeloma was the combination of melphalan plus prednisone, which produces response rates of 50–60% as first-line therapy but may preclude subsequent stem cell transplantation because of progenitor cell toxicity, stressing the importance of stem cell protection during induction regimens.2, 3 Subsequently, it was demonstrated that the combination of vincristine and doxorubicin (both administered as a continuous, 96-hour infusion through a central venous line) plus intermittent high-dose dexamethasone (the VAD regimen) improved response rates and the rapidity of responses in patients with newly diagnosed and refractory multiple myeloma.3–5 Response rates of 55–84% have been reported in newly diagnosed patients, with median remission durations of approximately 18 months.3, 5–7 Consequently, VAD became the first-line regimen in the 1980s, ultimately followed by intensive preparative regimens supported by autologous bone marrow or stem cell support. However, despite its demonstrated efficacy, the VAD regimen has many disadvantages that have reduced its popularity in recent years. The requirement for a continuous 96-hour infusion during each cycle necessitates placement of a central venous line, thus increasing the risk of infection and making it difficult to administer in the outpatient setting. Moreover, conventional VAD is associated with neutropenia, alopecia, doxorubicin-related cardiotoxicity, and steroid toxicity.5, 6, 8, 9 Because of these toxicities and the advanced age of many patients with multiple myeloma, treatment with VAD is not optimal in most patients.10 Thus, other regimens have been studied within and outside the confines of clinical trials. Dexamethasone alone or thalidomide plus dexamethasone (thal/dex) increasingly are being used because of their relative ease of administration. However, to date, there are no substantive data showing that these regimens are either superior or inferior to the use of VAD in the pretransplantation or nontransplantation setting. Moreover, there are scant data regarding the durability of response (progression-free survival [PFS] or overall survival [OS]) to dexamethasone alone or thal/dex, although some data suggest that the durability of response is inferior to that achieved with chemotherapy-based regimens.11
Given the toxicity concerns and difficulty of administering VAD, modifications of the regimen have been explored, including replacement of conventional doxorubicin with pegylated liposomal doxorubicin (Doxil®; Tibotec Therapeutics, Division of Ortho Biotech Products, LP, Bridgewater, NJ; also known as CAELYX® outside the United States, marketed by Schering-Plough Corporation, Kenilworth, NJ). Pegylated liposomal doxorubicin has several pharmacologic and safety advantages over conventional doxorubicin, including an extended circulation time (t½ = 55 hours), which allows 1-hour infusion every month instead of a 96-hour infusion with conventional doxorubicin; increased extravasation through abnormal angiogenic vessels, thereby providing increased tumor exposure; and a significantly lower risk of cardiotoxicity, myelosuppression, and alopecia.9, 12–15
Several groups have investigated the efficacy and safety of pegylated liposomal doxorubicin, vincristine, and reduced-dose dexamethasone (DVd) in multiple myeloma and have demonstrated that this regimen is tolerated well and produces response rates similar to those produced by the conventional VAD regimen. In a Phase II trial of DVd in 33 patients with newly diagnosed multiple myeloma, a response rate of 88% (12% complete responses; 76% partial responses) was demonstrated.16 The durability of response also was noteworthy, with a median time to disease progression of 23 months and a 3-year survival rate of 67%. Longer term follow-up from this study now estimates the median OS to be approximately 5 years (unpublished results). On the basis of these promising results, the current randomized, multicenter trial was designed to compare the efficacy and safety of DVd directly with those of VAd in patients with newly diagnosed multiple myeloma. Although there are doubts regarding the antimyeloma activity of vincristine in the VAD and DVd regimens, we included vincristine in the regimen, because it has been used traditionally in different myeloma regimens based on the work by Jackson et al.17 Moreover, we conducted the current study to compare conventional doxorubicin with pegylated liposomal doxorubicin, not to assess the role of vincristine. Thus, the objective of this study was to assess whether the use of pegylated liposomal doxorubicin could provide efficacy similar to that of conventional doxorubicin while reducing the toxicities commonly seen with conventional doxorubicin-based therapy for patients with newly diagnosed multiple myeloma.
MATERIALS AND METHODS
Patients age 18 years or older with previously untreated Durie–Salmon18 ≥ Stage I multiple myeloma were eligible. Eligible patients had to have symptomatic disease that required immediate treatment, measurable disease based on quantifiable serum and/or urine M-components, a Karnofsky performance status ≥ 60%, adequate bone marrow function (absolute neutrophil count ≥ 1000 cells/mm3, platelet count ≥ 75,000/mm3), adequate liver function (total bilirubin ≤ 2.0 mg/dL, aspartate aminotransferase ≤ 2 × the upper limit of normal), adequate renal function (serum creatinine ≤ 2.5 mg/dL), and a life expectancy of at least 3 months. Patients were ineligible if they had had any other malignancy within the past 5 years, with the exception of basal cell or squamous cell carcinoma of the skin or in situ carcinoma of the cervix, a history of cardiac disease (New York Heart Association Class II or greater with congestive heart failure), myocardial infarction within the past 6 months, unstable angina, uncontrolled hypertension or cardiac arrhythmias, uncontrolled diabetes mellitus, uncontrolled systemic infection, nonsecretory myeloma, monoclonal gammopathies of undetermined significance, or smoldering myeloma. Prior chemotherapy for multiple myeloma, radiotherapy to an area > 33% of the skeleton, local radiotherapy within 1 week of study treatment, any investigational supportive-care agent within 30 days of study treatment, or single-agent corticosteroids to treat multiple myeloma were not allowed. All patients provided informed written consent, and the study was approved by all local institutional review boards.
Study Design and Treatment
Patients were assigned randomly (1:1) to receive either DVd or VAd every 4 weeks either until they achieved maximal response, disease progression, or stable plateau disease or until unacceptable toxicity occurred. Patients who were eligible for hematopoietic stem cell transplantation and who demonstrated a response to either DVd or VAd could proceed to high-dose chemotherapy and transplantation after they completed at least four cycles of study drug (at least four cycles are considered standard before transplantation). The DVd regimen consisted of pegylated liposomal doxorubicin 40 mg/m2 intravenously (IV) over 1 hour plus vincristine 1.4 mg/m2 to a maximum of 2.0 mg IV over 5 minutes on Day 1 and dexamethasone 40 mg per day orally on Days 1–4 of each 28-day cycle. The VAd regimen consisted of vincristine 0.4 mg per day and conventional doxorubicin 9 mg/m2 per day IV over 96 hours and dexamethasone 40 mg per day orally on Days 1–4 of each 28-day cycle.
The primary endpoints were objective response rate and prospectively defined comparisons of toxicities and hospitalizations because of treatment-related adverse events (defined as clinical benefit). Objective response (defined as complete remission, remission, or partial remission) was assessed according to the modified Southwest Oncology Group (SWOG) criteria. The toxicities studied (i.e., clinical benefit) were defined prospectively based on the incidence of Grade 3 or 4 neutropenia or neutropenic fever, documented sepsis, antibiotic treatment, and the incidence and number of days of hospitalization for adverse events. Secondary endpoints included the time to disease progression, OS, safety, and tolerability. Adverse events were graded according to National Cancer Institute Common Toxicity Criteria (version 2). Additional analyses were performed to compare treatment groups with respect to the number of cycles administered in the hospital, the number of days required for drug administration, the number of cycles administered through a central venous line, and the requirement for growth-factor support.
Baseline assessment included quantification of serum and/or urine M-components by electrophoresis and immunofixation, unilateral bone marrow biopsy or aspirate, skeletal survey, multigated angiogram or echocardiogram to evaluate left ventricular ejection fraction (LVEF), electrocardiogram, and complete serum chemistry and urinalysis. After each cycle of therapy, quantification of serum and/or urine M-components by electrophoresis and immunofixation was performed to assess response, and radiologic assessment was performed if it was indicated clinically. At the end of treatment, LVEF was reevaluated with the same test that was used at baseline. Bone marrow biopsy or aspirate was obtained after every two to four cycles. Adverse events were monitored at each visit.
Modified SWOG criteria were used to assess objective status according to measurable and quantifiable protein levels. Accordingly, acceptable protein criteria are quantitative immunoglobulin and/or urine M-component (i.e., Bence Jones protein); if both are present, then the quantitative immunoglobulin is followed for response determination. On the basis of these criteria, a complete remission was defined by the absence of bone marrow or blood findings of multiple myeloma, including the disappearance of all evidence of serum and urine M-components on electrophoresis and by immunofixation studies. Remission (i.e., major remission) was defined as a 75–99% reduction in quantitative immunoglobulins and, if present, a 90–99% reduction in urine M-components or urine M-components < 0.2 g per day. Partial remission was defined as a 50–74% reduction in quantitative immunoglobulin and (if present) a 50–89% reduction in urine M-components. Stable disease was defined as a reduction < 50% in quantitative immunoglobulins or, if the patient had light-chain disease only, then a reduction < 50% in urine M-component. Progressive disease was defined as an increase > 100% from the lowest level of protein production, an increase > 25% above the baseline protein level after a confirmed remission, the reappearance of any myeloma peaks that had disappeared with treatment, or an increase in the number of lytic bone lesions. At the study end, an independent medical reviewer who was blinded to study treatment assessed objective response. If, at the time of this review, neither quantifiable immunoglobulin nor urine M-component data were available, then patients were considered not assessable for response. All patients who were candidates for bone marrow or stem cell transplantation were followed for approximately 1 month posttransplantation to determine engraftment success. All patients who did not proceed to transplantation were followed for 6 months after they discontinued the study treatment. In April 2004, a data analysis was performed on all patients to update disease progression and survival status.
The sample size determination was based on two 1-sided tests with α = 0.025 for therapeutic comparability between treatment groups based on the objective response rate. On the basis of the noninferiority design of this trial, the treatments were considered therapeutically equivalent if the corresponding 2-sided 95% confidence interval for the difference in objective response rates was within ± 20%. A sample size of 200 patients (100 patients per treatment group) provided 80% power to demonstrate therapeutic equivalence, assuming that the common objective response rate would be approximately 50%. This sample size provided > 80% power to detect a 20% difference in clinical benefit. All patients who received at least a partial dose of study drug were assessable for safety (modified intent-to-treat analysis), and all patients who completed at least one cycle of study drug were assessable for response. Statistical comparisons between treatment groups for response rate and clinical benefit were performed using the two-sided Fisher exact test; statistical comparisons of drug administration parameters were performed using the Wilcoxon two-sample test. PFS and OS were estimated by using the Kaplan–Meier method.19 In the analysis of PFS, patients were censored at the date of bone marrow or stem cell transplantation.
In total, 192 patients were randomized to receive either DVd (n = 97 patients) or VAd (n = 95 patients) and were included in all analyses (Fig. 1). Patients were enrolled at 23 academic and community centers between January 2001 and August 2003, and the last patient completed treatment in November 2003. Baseline patient demographics and disease characteristics were balanced well across treatment groups (Table 1). The mean patient age was 60 years in both treatment groups. Nearly 10% of patients in both treatment groups had a Karnofsky performance status ≤ 60, nearly 50% of patients had > 3 osteolytic lesions, and the mean percentage of plasma cells in the bone marrow was approximately 40% in both groups.
|Characteristic||DVd (n = 97)||VAd (n = 95)|
|Gender: No. of patients (%)|
|Male||57 (58.7)||58 (61.0)|
|Female||40 (41.2)||37 (38.9)|
|Mean age in yrs (range)||60 (37–84)||60 (44–81)|
|KPS: No. of patients (%)|
|≤ 60||10 (10.3)||8 (8.4)|
|70–80||34 (35.1)||36 (37.9)|
|90–100||53 (54.6)||51 (53.7)|
|Mean β-2 microglobulin, mg/L||3.96||4.99|
|Mean serum albumin, g/dL||3.51||3.29|
|Mean LDH, U/L||257||235|
|No. of lytic lesions: No. of patients (%)a|
|0–3||54 (55.7)||47 (49.5)|
|> 3||41 (42.3)||47 (49.5)|
|Percentage of plasma cells in bone marrow, mean ± SD||40.0 ± 26.0||41.7 ± 24.5|
|Percent LVEF, mean ± SD||62.1 ± 7.5||63.2 ± 7.6|
All patients received at least one cycle of treatment, and the median number of cycles received was three in both treatment groups (Table 2). In the DVd group, 4% of cycles were administered in the hospital compared with 32% of the VAd cycles (P < 0.001). The DVd group required significantly fewer days for drug administration (1.3 days vs. 5.2 days per cycle; P < 0.001). In addition, 45% of DVd cycles were administered through a central venous line compared with 96% of VAd cycles (P < 0.0001). Although 4% of DVd cycles were administered in the hospital, the majority of these were the first cycle; ≥ 98% of subsequent cycles were administered in the outpatient clinic. In contrast, approximately 30% of all VAd cycles were administered in the hospital. Dose modifications were required in 16.9% of DVd cycles and in 11.4% of VAd cycles (Table 2). The most common reason for dose modification in both treatment groups was cycle delay. The most common reasons for treatment discontinuation were similar between treatment groups (Table 2) and included proceeding to transplantation, response to treatment, and disease progression.
|Parameter||DVd (n = 97)||VAd (n = 95)|
|Median no. of cycles (range)|
|All patients||3 (1–15)||3 (1–12)|
|Patients proceeding to transplant||4 (1–7)||4 (1–8)|
|Total no. of cycles administered||474||420|
|No. of cycles that required dose modification (%)|
|Yes||80 (16.9)||48 (11.4)|
|No||394 (83.1)||372 (88.6)|
|Type of dose modification: No. of cycles with dose modification (%)|
|Cycle delay||54 (67.5)||29 (60.4)|
|Infusion interruption||13 (16.3)||14 (29.2)|
|Infusion dose level reduction||25 (31.3)||6 (12.5)|
|Modified dexamethasone regimen||1 (1.0)||6 (12.5)|
|Reasons for discontinuation: No. of patients (%)|
|Transplantation||30 (31.3)||34 (35.8)|
|Response to treatment||30 (31.3)||28 (29.5)|
|Progression of multiple myeloma||13 (13.5)||11 (11.6)|
|Decline in LVEF||1 (1.0)||4 (4.2)|
|Adverse events||10 (10.4)||8 (8.4)|
|Death||3 (3.1)||1 (1.1)|
The objective response rate was similar in both treatment groups (Table 3). In the DVd group, 43 patients (44.3%) had an objective response, including 3 complete remissions, 15 remissions, and 25 partial remissions. In the VAd group, 39 patients (41.0%) had an objective response, including 15 remissions and 24 partial remissions. No patient who received VAd had a complete remission. The difference in response rate was 3.3% (95% confidence interval, − 17.3 to 10.7%; P = 0.66), thus satisfying the protocol-specified criteria for equivalence. Fourteen patients in the DVd group and 10 patients in the VAd group were not assessable for response, because they had neither quantifiable serum immunoglobulin nor urine M-component data available and, thus, were considered nonresponders.
|Response||No. of patients (%)||P|
|DVd n = 97||VAd n = 95|
|Objective response||43 (44.3)||39 (41.0)||0.66|
|Complete remission||3 (3.1)||0 (0.0)|
|Remission||15 (15.5)||15 (15.8)|
|Partial remission||25 (25.8)||24 (25.3)|
|Stable disease||38 (39.2)||46 (48.4)|
|Progressive disease||2 (2.1)||0 (0.0)|
|Not assessablea||14 (14.4)||10 (10.5)|
Thirty-four patients (35%) who received DVd and 35 patients (37%) who received VAd proceeded to autologous stem cell transplantation. All but two patients in each group underwent a successful graft. Patients who proceeded to transplantation received a median of four cycles of DVd or VAd before transplantation (Table 2). According to the protocol, 30 of 34 patients (88%) in the DVd group and 32 of 35 patients (91%) in the VAd group received ≥ 4 cycles of treatment before they underwent transplantation.
The prospectively defined comparison of toxicities (i.e., clinical benefit) of DVd compared with VAd is shown in Table 4. Patients who received DVd had a significantly lower incidence of Grade 3/4 neutropenia or neutropenic fever (P = 0.02); they also had a lower incidence of documented sepsis and antibiotic use. Consistent with the lower incidence of severe myelosuppression, significantly fewer patients in the DVd group required growth factor support (46% vs. 61%; P = 0.03). Growth factors for neutropenia were administered to 10% of patients in the DVd group compared with 21% of patients in the VAd group (P = 0.04), and growth factors for anemia were administered to 44% of patients in the DVd group compared with 60% of patients in the VAd group (P = 0.03).
|Incidence type||No. of patients (%)||Pa|
|DVd n = 97||VAd n = 95|
|Neutropenia (Grade 3 or 4) or neutropenic fever||11 (10.3)||23 (24.2)||0.02|
|Documented sepsis,||3 (3.1)||8 (8.4)||0.13|
|Antibiotic treatment||61 (62.9)||65 (68.4)||0.45|
|Hospitalizations for adverse events||36 (37.1)||34 (35.8)||0.88|
PFS and OS
At a median follow-up of 21 months in the DVd group and 20 months in the VAd group, no significant differences were seen in either PFS or OS between treatment groups (Figs. 2, 3). The hazard ratio for PFS was 1.11 (P = 0.69), and the hazard ratio for OS was 0.88 (P = 0.67). The 1-year PFS rates were 78% for the DVd group compared with 76% for the VAd group, and the 2-year PFS rates were 53% and 56%, respectively. The 1-year OS rates were 88% for the DVd group and 85% for the VAd group, and the 2-year OS rates were 79% and 72%, respectively.
Both regimens were tolerated well acutely, and the majority of adverse events were mild to moderate in severity (Table 5). The incidence of nausea, asthenia, constipation, anemia, fever, and stomatitis was similar between treatment groups. Alopecia, injection-site reactions, and Grade 3 or 4 neutropenia were significantly less common in the DVd group, whereas hand–foot syndrome was significantly more common in the DVd group. The majority of hand–foot syndrome was Grade 1 or 2: Only 4 patients developed Grade 3 symptoms, and none of those patients discontinued treatment because of hand–foot syndrome. Grade 2 alopecia (i.e., pronounced or complete hair loss) also was significantly more common in the VAd group (20% vs. 4%; P < 0.0001). Treatment with VAd was associated with Grade 3 or 4 congestive heart failure in 2 patients, and 1 of those patients also had a cardiomyopathy. In contrast, no patient in the DVd group had Grade 3 or 4 cardiac adverse events. Moreover, DVd had less effect on cardiac function compared with VAd (Fig. 4). The mean absolute decrease from baseline LVEF was significantly less in the DVd group (P < 0.01). Nine deaths occurred on study in the DVd group compared with 14 deaths in the VAd group, primarily from disease progression (n = 9 patients). No deaths were attributed to thromboembolic events.
|Toxicity||Percentage of patients|
|All grades||Grade 3 or 4|
|DVd n = 97||VAd n = 95||DVd n = 97||VAd n = 95|
The results of this trial confirm that DVd and VAD have similar efficacy for the treatment of newly diagnosed multiple myeloma and are consistent with the recent report by Dimopoulos et al.20 This study highlights the disadvantages and toxicities of VAd that have led to its decreased utility. Patients who received DVd had a significantly lower incidence of severe neutropenia or neutropenic fever, significantly less need for growth factor support, less need for antibiotics, and overall fewer days for drug administration compared with patients who received VAd. Treatment with DVd was associated with significantly higher incidence of hand–foot syndrome, but it was mostly (84%) Grade 1 and 2 and was not dose limiting. Patients who received DVd had significantly less alopecia and decrease in cardiac function. In addition, the majority of DVd cycles were administered in the outpatient setting without the need for a central venous line. Accordingly, the similar efficacy of DVd with less bone marrow suppression and Grade 3/4 neutropenia or neutropenic fever compared with VAd offers a clear strategy when selecting an anthracycline-based regimen for initial treatment of multiple myeloma. The DVd regimen also theoretically permits continued therapy (i.e., greater number of cycles) because of fewer serious complications in responsive patients, thereby minimizing tumor burden at the time of transplantation. Although rapid response and relief of symptoms for patients with multiple myeloma are desirable, the quality of response to nonmyeloablative doses of chemotherapy does not appear to correlate with improvements in survival for patients with multiple myeloma, as opposed to transplantation approaches.7, 21, 22 In contrast, the importance of the time to disease progression as an indicator of survival and, thus, as an indicator of prognosis in patients with newly diagnosed multiple myeloma was demonstrated recently in a landmark analysis of SWOG chemotherapy trials.7 That analysis showed that patients with either an objective response or stable disease (and, thus, delayed disease progression) had a similar survival prognosis, which was significantly better compared with the prognosis for patients who progressed. Therefore, PFS is an important efficacy endpoint for predicting survival. In the current trial, the median PFS was equivalent to that reported previously for DVd16, 20 and higher than that reported for VAD.3, 5–7 Nearly all patients had either an objective response or stable disease (84% in the DVd group, 89% in the VAd group); only 2 assessable patients in the DVd group had disease progression on study. It is noteworthy that the 2-year survival rate of 79% in the DVd group (based on data from 90% of patients) compares favorably with that reported with other regimens for patients with multiple myeloma.2, 4, 5 Therefore, on the basis of mature PFS and OS data, DVd appears to be an effective regimen for the treatment of patients with newly diagnosed multiple myeloma, and it has an improved safety profile compared with VAd.
Compared with previously reported studies of pegylated liposomal doxorubicin and vincristine with either full-dose or reduced-dose dexamethasone, the response rate observed in this trial (based on strict, modified SWOG criteria) was lower.16, 20, 23 The lower response rate observed in this study may have been influenced by several factors, such as patients' baseline disease characteristics, the decision to proceed to transplantation before patients achieved maximal response, or the use of a reduced-dose dexamethasone regimen. In the Phase II trial reported by Hussein et al.,16 which demonstrated a response rate of 88%, patients received a mean of 9 cycles of DVd and achieved maximal response after 5 months. In comparison, patients in the current trial received a mean of 3.4 cycles of DVd. In addition, patients in the Phase II trial were younger and had better performance status than patients in the current trial. These factors may have contributed to the higher response rate observed in the earlier Phase II trial.
Because PFS appears to be a primary predictor of survival, it is important to note that the median PFS of > 24 months observed in the current trial is the same as that reported previously in the Phase II trial.16 Moreover, PFS with DVd appears to be similar to that reported in the randomized, multicenter study conducted by the Greek Myeloma Study Group that compared DVD with VAD directly as first-line therapy. Full-dose dexamethasone was used in that trial, and VAD was administered by bolus rather than continuous infusion. The objective response rate was 61%, and the median time to disease progression was 24 months for patients who received either regimen.20 Therefore, it appears that regimens using pegylated liposomal doxorubicin (with either full-dose or reduced-dose dexamethasone) are highly active, yielding a median time to progression of approximately 24 months, and provide similar antimyeloma activity compared with conventional anthracycline-based regimens in newly diagnosed patients. On the basis of an indirect comparison of the current trial and the Greek Myeloma Study Group, it also appears that reducing the dose of dexamethasone in the DVd regimen may have improved tolerability without affecting PFS or OS.
These data suggest that DVd and VAd have similar efficacy and that DVd has significant advantages in terms of convenience of dosing and safety. Most important, DVd is associated with less myelosuppression and, thus, fewer Grade 3/4 neutropenic events or fever, results in less cardiotoxicity, and does not require a central line or continuous infusion. However, the use of an anthracycline-based regimen, such as VAD, for initial treatment of multiple myeloma has declined, and thal/dex is used increasingly, primarily because of the convenience of its oral formulation and use as an outpatient therapy. Single-center, Phase II studies reportedly have produced response rates for thal/dex between 64% and 72%.24, 25 More recently, Rajkumar et al.26 presented the final results from a multicenter, randomized study comparing dexamethasone alone with thal/dex. Response rates in that study were 42% for dexamethasone alone and 58% for thal/dex. However, because of significant toxicities seen in the thal/dex arm (e.g., 18% Grade 3/4 deep vein thrombosis for thal/dex vs. 3% for dexamethasone alone; 34% any Grade 4 adverse event for thal/dex vs. 17% for dexamethasone alone), those authors noted that the use of this combination regimen should be weighed carefully.26, 27 Longer follow-up of large multicenter studies will be necessary to define the activity of thalidomide in front-line regimens. Although several new compounds have been introduced into the armamentarium for multiple myeloma therapy, chemotherapy continues to have a major role in the management of the disease. Recently, however, Cavo et al.28 challenged this paradigm when they reported results from a retrospective, matched case–control trial suggesting that thal/dex is superior to VAD. Certainly, this is consistent with the observed change in clinical practice from an anthracycline-based regimen to thal/dex for first-line management of symptomatic myeloma. However, those authors indicated that these data should be interpreted with caution, because the study, albeit well controlled, was not randomized. Moreover, the recent landmark analysis of the SWOG trials suggests that the time to disease progression, and not the quality of response (complete response, partial response, stable disease), is predictive of survival.7 Therefore, because Cavo et al. did not report on the time to progression, interpretation of their data are further confounded. Thus, the optimal management of patients with newly diagnosed myeloma requires further study; and, currently, the evidence does not point toward one particular treatment as the standard of care. There is also a critical need for long-term follow-up data from studies examining new treatment options in the first-line setting. To date, for example, there are no substantive reports on long-term survival in newly diagnosed patients who are treated with thal/dex. On the basis of the SWOG analysis, this is critical. Moreover, the effect of the early use of immune modulators, such as thalidomide, in newly diagnosed patients is not clear and needs to be established. This is highlighted by data from the University of Arkansas showing a possible negative impact of thalidomide when it is used in newly diagnosed patients with regard to their response to salvage therapy.29 Longer follow-up and larger studies are needed to establish the role of these new compounds and regimens.29
Finally, the relative role of chemotherapy also is under question. The data presented here suggest that the chemotherapy component of DVd or VAd (pegylated liposomal doxorubicin and/or vincristine) was responsible for the prolonged and durable PFS and OS, because there appeared to be no differences in the PFS and OS rates reported in our study and those reported by Dimopolous et al.,20 despite different dosages of dexamethasone (reduced-dose vs. full-dose dexamethasone). This raises the question whether vincristine and/or the anthracycline component of the regimen were responsible for the durability of response. This question may have been answered by a study recently reported by Offidani et al.,30 who investigated the use of thal/dex plus pegylated liposomal doxorubicin in 59 patients with newly diagnosed or recurrent myeloma. The overall response rate in their study was 79% (30% complete responses, 10% very good partial responses, 39% partial responses); and at a median follow-up of 12 months, the projected 2-year OS and PFS rates were 73% and 50%, respectively. This is comparable to the outcome achieved with DVd-T, the same regimen described above but with the addition of vincristine.31 To examine the role of chemotherapy further and to build on the study we report here, we are now comparing pegylated liposomal doxorubicin plus thal/dex versus thal/dex alone in a randomized study in newly diagnosed patients. This will allow us to assess the role of pegylated liposomal doxorubicin in the treatment of patients with multiple myeloma and also to assess more globally the role of chemotherapy as an anchor component in treatment regimens for patients with myeloma.
In the current study, we demonstrated that, by replacing conventional doxorubicin with pegylated liposomal doxorubicin (changing from VAd to DVd), we were able to overcome many of the disadvantages of VAd while providing an effective and convenient outpatient treatment option for patients with newly diagnosed, symptomatic multiple myeloma. In addition, by reducing the toxicities associated with conventional doxorubicin, the DVd regimen optimizes the opportunity for patients to proceed in a timely manner to single or tandem transplantation. Moreover, this regimen is a base for further development of new regimens for the management of multiple myeloma. For example, the combination of an immune modulator, such as thalidomide or lenalidomide, with DVd may have synergistic antiangiogenic and antitumor effects. It has been shown that the DVd regimen reduces microvessel density in myeloma patients,32 and thalidomide may prevent new blood vessels from forming and sensitize myeloma cells to cytotoxic agents, thereby improving the quality of response.33
The authors gratefully acknowledge the patients and families who made this study possible and the participating investigators, including A. Briggs, H. Burris, C. DeCastro, M. Gautier, J. Gurtler, F. Ravandi, Y-H. Chen, L. Heffner, J. Wall, K. Stewart, J. Ganey, J. Smith, D. Vafai, J. Hajdenberg, B. Mason, T. Pluard, R. Smith, D. Gravenor, J. Gandhi, J. Kirshner, and F. Yunus. The authors also thank the statisticians at ALZA Corporation—Ramanan Gopalan and Chinglin Lai—and Pam Jacobs, Colin Lowery, Steven Fruchtman, and Mark Wildgust in Clinical Affairs at Tibotec Therapeutics.
- 2Plasma cell neoplasms. In: DeVitaVTJr., HellmanS, RosenbergSA, editors. Cancer: principles and practice of oncology, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2001: 2465–2499., , .
- 11Melphalan-prednisone versus dexamethasone-based regimens for newly diagnosed myeloma patients aged 65–75 years. Safety analysis of the IFM 95 trial on 457 patients [abstract 4609]. Blood. 1999;94: 309b., , , et al.
- 26Thalidomide plus dexamethasone versus dexamethasone alone in newly diagnosed multiple myeloma (E1A00): results of a Phase III trial coordinated by the Eastern Cooperative Oncology Group [abstract 205]. Blood. 2004; 104: 63a., , , , .
- 27National Comprehensive Cancer Network. Clinical practice guidelines in oncology: multiple myeloma. Version 1. Available at URL: http://www.nccn.org/professionals/physician_gls/PDF/myeloma.pdf (accessed February 9, 2005).
- 29Management of patients with multiple myeloma (MM) failing total therapy 2 (TT 2) according to thalidomide (THAL) randomization [abstract 1483]. Blood. 2004; 104: 414a., , , .
- 30Thalidomide, dexamethasone and pegylated liposomal doxorubicin (ThaDD) in elderly/relapsed multiple myeloma (MM) patient [abstract]. Haematologica. 2005; 90(Suppl 2): 9., , , et al.
- 31Pegylated doxorubicin(D), vincristine(V), reduced frequency dexamethasone(D) and thalidomide(T) (DVd-T) in newly diagnosed (Nmm) and relapsed/refractory (Rmm) multiple myeloma patients [abstract]. Blood. 2003;102: 237a., , , , , .
- 32The effects of different chemotherapeautic regimens on bone marrow microvessel density in patients with multiple myeloma [abstract 2424]. Proc Am Soc Clin Oncol. 2003;22: 603., , , , , .
- 33Doxil (D), vincristine (V), decadron (d) and thalidomide (T) (DVd-T) for relapsed/refractory multiple myeloma (RMM) [abstract 1566]. Blood. 2002;100:403a., , , , .