Fax: (570) 882-6509
Phase III study of PSC-833 (valspodar) in combination with vincristine, doxorubicin, and dexamethasone (valspodar/VAD) versus VAD alone in patients with recurring or refractory multiple myeloma (E1A95)†
A trial of the Eastern Cooperative Oncology Group
Article first published online: 17 JAN 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 4, pages 830–838, 15 February 2006
How to Cite
Friedenberg, W. R., Rue, M., Blood, E. A., Dalton, W. S., Shustik, C., Larson, R. A., Sonneveld, P. and Greipp, P. R. (2006), Phase III study of PSC-833 (valspodar) in combination with vincristine, doxorubicin, and dexamethasone (valspodar/VAD) versus VAD alone in patients with recurring or refractory multiple myeloma (E1A95). Cancer, 106: 830–838. doi: 10.1002/cncr.21666
Presented in part at the American Society of Hematology Meeting, San Diego, CA, December 6–9,2003.
- Issue published online: 3 FEB 2006
- Article first published online: 17 JAN 2006
- Manuscript Accepted: 1 SEP 2005
- Manuscript Revised: 22 JUL 2005
- Manuscript Received: 21 APR 2005
- Eastern Cooperative Oncology Group
- Public Health Service. Grant Numbers: CA23318, CA66636, CA21115, CA73590, CA41287, CA13650
- National Cancer Institute. Grant Number: CA31946
- National Institutes of Health, and the Department of Health and Human Services
- refractory multiple myeloma;
Preliminary studies have shown valspodar (PSC-833: Novartis Pharmaceuticals, East Hanover, NJ) to be a potent inhibitor of multidrug resistance (MDR), one cause of resistance to chemotherapy. An international randomized control study (Phase III) evaluated the use of vincristine, doxorubicin, and dexamethasone (VAD) with (n = 46) and without (n = 48) valspodar in the treatment of patients with recurring or refractory multiple myeloma.
Patients with documented recurrence or refractory myeloma were stratified based on prior treatment exposure and creatinine and randomized. Because of interaction of valspodar with vincristine and doxorubicin, the doses of these drugs were reduced compared with the VAD-alone arm, and the doxorubicin was further reduced in the last 15 patients when given with valspodar based on pharmacokinetic and toxicity studies.
There were no complete or near-complete responses. There were 29% partial responses (PRs) in the VAD-alone arm and 44% with valspodar (P = 0.2). Median progression-free survival was 7 months with VAD alone and 4.9 months with valspodar (P = 0.50). Subjective response was 19% with VAD alone and 17% with valspodar (P = 1.0). Median survival with VAD alone was 18.5 months and 15.3 with the addition of valspodar (P = 0.055). Toxicity of Grade 3 or greater was higher (P < 0.0001) in the valspodar arm (89%) compared with the VAD-alone arm (58%). The reduction of doxorubicin dose reduced toxicity but not significantly (P = 0.11).
The addition of the MDR-modulating agent valspodar to VAD did not improve treatment outcome. Toxicity was increased in the valspodar-treated group compared with VAD alone. Cancer 2006. © 2006 American Cancer Society.
Although multiple myeloma is usually fatal, the majority of patients achieve palliation with chemotherapy.1–3 Responding patients develop resistance and require salvage treatment.4–6 A number of different mechanisms can cause resistance, including multidrug resistance (MDR) mediated by the overexpression of a p-glycoprotein (P-gp).7, 8 P-gp is a 170-kD ATP-dependent membrane transporter that acts as a drug efflux pump and prevents cytotoxicity by preventing drugs such as vincristine and doxorubicin (as well as other natural-product drugs) from having sufficient time to cause cell death.9 Previous studies in multiple myeloma have shown that the MDR-1 gene is frequently up-regulated in patients with multiple myeloma who have been previously treated with vincristine and doxorubicin causing an increased expression of P-gp 170, thus preventing cytotoxicity ex vivo and causing clinical resistance in patients with recurring multiple myeloma.10, 11, 12 Preliminary clinical studies have shown the ability of MDR modulators to reverse drug resistance and allow responses in several diseases, including multiple myeloma.13–15 Cyclosporin A has been shown to be one of the most effective ex vivo modulators of drug resistance and appears capable of reversing clinical resistance in patients with recurring or refractory multiple myeloma.15 PSC-833 (valspodar), a cyclosporin D analog, has been shown to be a much more effective inhibitor of MDR than cyclosporin A and to be less toxic.16–18 Unlike cyclosporin A, valspodar appears to be neither immunosuppressive nor nephrotoxic with the ability to achieve levels in the blood (1 μm) capable of reversing drug resistance without excessive toxicity.19–22 As a single agent, ataxia, or dizziness, is a prominent side effect requiring dose adjustments. VAD (a 4-day infusion of vincristine and doxorubicin with high-dose oral dexamethasone) has become a frequent salvage regimen for patients with multiple myeloma with a partial response of 20–30% in patients with refractory disease and 45–70% in patients with recurrence.4, 23 In a preliminary study, Dalton et al.24 found that valspodar with VAD was tolerated in a dose of 4 mg/kg by mouth four times a day for a total of 20 doses. Because valspodar inhibits endogenous P-gp expressed on hepatic parenchymal cells, Dalton et al.24 found it was necessary to reduce the usual dose of doxorubicin from 9 mg/m2/24 hours to 7 mg/m2/24 hours, and the dose of vincristine from 0.4 mg/24 hours to 0.2 mg/24 hours to allow acceptable toxicity. In a Phase 2 study, Case et. al.25 found valspodar capable of reversing clinical resistance in some patients. This Phase 3 study of VAD with or without valspodar in patients with documented recurring or refractory multiple myeloma was designed to see if there was an improvement in objective response or survival. Secondary objectives include analysis of subjective response, toxicity, progression-free survival (PFS), prognostic factors, and the correlation between treatment outcome and expression of MDR.
MATERIALS AND METHODS
The study opened for accrual on March 21, 1997, and closed on May 10, 2000. Patients must have had a confirmed diagnosis of multiple myeloma as previously described with measurable disease.23 Objective progression was required to be documented as previously described.23 Patients receiving prior vincristine were eligible, as well as those receiving prior doxorubicin if the total dose was less than 250 mg/m2 and the ejection fraction was greater than the institutional lower limit of normal. Patients were required to have an ECOG (Eastern Cooperative Oncology Group) performance status of 0–3 and to not have serious medical problems that the investigator felt would contraindicate chemotherapy. Within 2 weeks of entry to study the absolute neutrophil count needed to be at least 1000/mm3, platelets at least 50,000/mm3, creatinine less than 3 mg/dL, and bilirubin and aspartate aminotransferase (AST) less than 1.5 times the upper limit of institutional normal. Patients must have previously received adequate prior chemotherapy for multiple myeloma including at least two cycles of combination chemotherapy. Patients who received VAD must have completed their last cycle at least 3 months before study entry with no evidence of progressive disease while on treatment. Patients previously treated with autologous stem cell transplant were eligible, but not patients who had received prior allogeneic transplantation. Patients who had a preexisting peripheral neuropathy with weakness, or cerebellar disease with ataxia, were not eligible. Patients were required to give written informed consent before participating in this study and prestudy bone marrow and peripheral blood specimens were required.
Patients randomized to arm A received vincristine 0.4 mg/24 hours intravenously (i.v.) on Days 1–4 with doxorubicin 9 mg/m2/24 hours on Days 1–4, both by continuous 96-hour infusion and dexamethasone 40 mg by mouth on Days 1–4 and 15–18. Cycles were repeated every 28 days. Patients randomized to arm B received vincristine 0.2 mg/24 hours on Days 2–5, and doxorubicin 7 mg/m2 over 24 hours on Days 2–5, both by continuous 96-hour infusion, and dexamethasone 40 mg by mouth on Days 2–5 and 16–19. Cycles were repeated every 28 days. Valspodar 4 mg/kg by mouth q.i.d. was begun on Day 1 at 11 a.m. and taken through Day 6 at 6 a.m. Pharmacokinetic studies of doxorubicin and doxorubicinol levels done on 30 patients showed that the 48- and 96-hour doxorubicinol levels were significantly (P < 0.001) greater in arm B than in arm A (see Results). Subsequently, doxorubicin was reduced to 6 mg/m2/24 hours on Days 2–5 in arm B, with 15 patients receiving the reduced dose, and the other components of arm B remaining the same.
Patients were continued on schedule as long as they had stable disease or response and toxicity was acceptable. Patients were evaluated after two cycles and if stable or responding, they continued on study until they achieved maximum response, or plateau phase, and then two additional cycles were given. If a patient had a cumulative dose of 650 mg/m2 of doxorubicin (total lifetime dose) and was still responding or stable, the patient was to have had the doxorubicin discontinued and to continue with the rest of the treatment. The initial cycle VAD was given at full dose regardless of initial blood counts. No dose reduction was made for renal insufficiency. Doxorubicin was reduced by 25% if radiation therapy was being administered, either concomitantly to a single bony site or within the previous 2 weeks.
Measurement of Response
Response was evaluated using the ECOG myeloma response criteria and criteria were developed for subjective response. Objective response in the serum and urine M-protein required reduction of greater or equal to 50% of pretreatment M-protein levels in both serum and urine. If only urine was used as a response criterion, the serum M-protein had to be less than 1 g/dL and the urine excretion had to be ≦ 10% of pretreatment/24 hours. Patients with objective response who had complete disappearance of an M protein and no evidence of myeloma in the bone marrow were considered to have a complete response (CR). Patients who met the complete response criteria who did not have a repeat bone marrow or who had 3–6% bone marrow plasma cells remaining or less than 3% plasma cells remaining with some sheets or clusters of malignant plasma cells were felt to have near complete response (NCR). Patients who met the criteria for objective response but who were not CR or NCR were considered to have a partial response (PR). A patient in CR, NCR, or PR was classified as being in plateau if they had an objective response and serum and urine M proteins had been stable (less than 20% variation) or had disappeared for a period of at least 4 weeks. A patient who did not meet the response criteria outlined above was felt to show no change. Patients were felt to have recurrence or progressed on treatment as previously described.23
Subjective response was based on ECOG performance status, pain status, and hemoglobin/transfusion response. Patients were given a brief pain questionnaire and an analgesic scale monthly. A response in performance status was defined as a decrease to 1 or 0. The brief pain questionnaire categorized pain as 1–4 being mild, 5–6 as moderate, and 7–10 as severe. A response in pain status was defined as an improvement by two categories in the worst pain score on the brief pain questionnaire, with no increase in narcotic use, or an improvement in the worst pain score by one category, with a decrease in narcotic consumption to occasional or none. Hemoglobin/transfusion response (in the absence of erythropoietin treatment) was considered an increase in hemoglobin of greater than 2 g/dL without transfusion or other obvious contributing factors. Patients were considered to have an excellent subjective response (SR) if they had an improvement in any two of the above three parameters. Patients who improved in either their performance status or pain status were considered to have a good SR if they did not have a hemoglobin response. Patients who improved by a decrease in transfusion requirements were considered to have had at least a subjective response. Improvement in performance status/pain status must have been maintained for at least 4 weeks to confirm an SR and the hemoglobin/transfusion benefit must have lasted at least 6 weeks.
Bone marrow aspirates were obtained within 1 month of study entry and quantitatively analyzed for multidrug resistance (MDR-1) and multidrug resistance protein (MRP-1) expression by reverse transcriptase polymerase chain reaction (RT-PCR), with histone H3 level used as an internal standard.26 Standard curves for MDR-1 and MRP-1 expression were generated by performing serial dilution experiments using RNA from sensitive and drug-resistant cell lines. Plasma cell labeling index was done as previously described. Doxorubicin and doxorubicinol levels were determined for patients in both arms during the first cycle of treatment at baseline and at 48 and 96 hours during the infusion.
Patients completed the brief pain inventory and analgesic scale for each chemotherapy cycle. The following baseline tests were obtained: complete blood count (CBC), platelet count, bone marrow aspiration/biopsy, measurement of serum M-protein in the serum and urine by electrophoresis (or nephelometry), quantitative immunoglobulins, serum soluble interleukin-6 receptor, serum beta-2 microglobulin, serum and urine immunoelectrophoresis or immunofixation, skeletal X-rays, serum calcium, phosphate, uric acid, creatinine, glucose, electrolytes, albumin, bilirubin, AST, alkaline phosphatase, and alanine aminotransferase (ALT), measurement of soft-tissue plasmacytoma, chest X-ray, multiple gated acquisition (MUGA) or echocardiogram, electrocardiogram (EKG), and pregnancy test for women of child-bearing potential. The CBC, platelet count, chemistries, and measurement of serum and/or urine M protein were repeated every 4 weeks. Repeat 24-hour urine tests were done monthly in patients with measurable disease, and every 12 weeks in those who lacked urine M-protein. X-rays were repeated after two cycles and when needed for evaluation of response. Bone marrows were done if necessary. MUGA (or echocardiogram) was to be repeated before every fourth cycle.
The primary aim of this study was to compare VAD with VAD plus valspodar for objective response (OR) and overall survival (OS). Secondary objectives were to analyze for progression-free survival (PFS), subjective response (SR), toxicity, and to correlate response with MDR status. It was estimated that 360 patients would be needed to detect a 50% difference in median survival with 80% power using a two-sided log rank test with an overall Type I error rate of 0.05. The study was closed when 100 patients had been accrued because of slow accrual. To calculate the conditional power at the current sample size, East statistical software30 was used. PFS was the time from registration until the first evidence of progression or death. Patients who had no change as their best evaluation for response were censored at the last known time they were stable. OS and PFS were estimated using the methods of Kaplan and Meier.27 The stratified log rank test28 was used to test for difference in OS and PFS between treatment arm and MDR or MRP expression. Cox proportional hazard models29 were used to determine the association between survival and other covariates.
One hundred patients were registered from 39 institutions and five cooperative groups: ECOG (64), Cancer and Leukemia Group B (CALGB) (10), National Cancer Institute of Canada (NCIC) (15), Southwestern Oncology Group (SWOG) (5), and European Organisation for Research and Treatment of Cancer (EORTC) (6). Six patients were ineligible. Of the 94 eligible patients (Table 1), 48 were randomized to VAD and 46 to valspodar plus VAD (31 patients received 7 mg/m2 of doxorubicin and 15 received 6 mg/m2 of doxorubicin). The median age of all patients was 65 years (range, 35–84) with a median time since diagnosis of 2.4 years and a median of 2.7 months from recurrence. Half of the patients had received prior radiotherapy and 54% had received vincristine, doxorubicin, or both. At study entry, 81% of the patients had a performance status of 0 or 1 and 19% had performance status of 2. There were no statistically significant differences in any of the baseline patient characteristics by treatment arm, except for the IgG level with patients receiving VAD+valspodar (P = 0.046).
|Patient characteristics||VAD||VAD+ valspodar|
|Age, median (range)||61 (35–84)||67 (46–83)|
|Response to prior treatment|
|Recurring on or off therapy||40||83||42||91|
|Refractory to all treatment||8||17||4||9|
|Median yrs since diagnosis (range)||2.7 (0.4–7.6)||2.1 (0.4–12.3)|
|Median mos since recurrence (range)||2.7 (0–59.5)||3.0 (0–74.7)|
|Time since BMT in yrs, median (range)||2.2 (0.5–5.5)||2.2 (0.5–5.5)|
Analysis of Survival and Response
The median survival in the VAD arm was 18.5 months (95% confidence interval [CI], 15.7–24.4) and in the VAD plus valspodar arm median survival was 15.3 months (95% CI, 7.0–20.3), which was of marginal statistical significance (Fig. 1), P = 0.055. The median followup time for survivors was 31.1 months (range, 1.8–53.5). The current observed hazard ratio of VAD plus valspodar arm (B) over VAD arm (A) was 1.48 (95% CI, 0.95, 2.3). This interval does not contain the target hazard ratio for an improved survival on the VAD plus valspodar arm (B) (0.667). Therefore, it is unlikely that VAD plus valspodar yields a survival benefit. In addition, the conditional power, the probability that we would reject the null hypothesis of equal hazard rates on the two arms given the results at the current analysis, assuming a true hazard ratio of VAD plus valspodar arm (B) over VAD arm (A) of 0.667, is 0.041. This indicates that given the current results, if we were to have continued the study to full information, the probability of rejecting the null hypothesis in favor of VAD plus valspodar is very small even if VAD plus valspodar is truly better than VAD.
Of the 94 eligible patients, there were 10 inevaluable for response, 2 in the VAD arm and 8 in the VAD plus valspodar arm (Table 2). No patient achieved a complete or near complete response. In the VAD arm (A), 14 patients (29%) achieved a partial response, 25 (52%) did not change, and 3 (6%) progressed. In the VAD plus valspodar arm (B), 20 patients (44%) achieved a partial response, 14 (30%) did not change, and 3 (6%) progressed. Comparing treatment arm A with B, there was no statistically significant difference (29% vs. 44%, respectively), P = 0.2 (Fisher exact test).
|Overall objective response||Treatment|
|n = 48||n = 46|
|Near complete response||0||(0)||0||(0)|
Analysis of Progression-Free Survival and Subjective Response
Of the 89 eligible patients with objective response data, 44 received VAD and 45 VAD plus valspodar (Fig. 2). The median PFS in the VAD arm (Fig. 2) was 7 months (95% CI, 3.8–9.2) and 4.9 months (95% CI, 2.5–8.2) in the VAD plus valspodar arm (P = 0.5).
Nine (19%) patients in arm A (VAD) and 8 (17%) in arm B (VAD plus valspodar) (Table 3) experienced either excellent or good subjective response (P = 1.0).
|Best overall subjective response||Treatment|
|n = 48||n = 46||N = 94|
|Excellent subjective response||1||(2)||3||(6)||4||(4)|
|Good subjective response||8||(17)||5||(11)||13||(14)|
Ninety-six patients had data to analyze for Grade 3 or higher toxicities regardless of their eligibility status (Table 4). Hematological toxicities were excluded in the calculation of the worst degree because of the difficulty in distinguishing between treatment-related nadirs and those due to the disease. There were more patients (89%) in arm B (VAD plus valspodar) than arm A (58%) that experienced Grade 3 or higher toxicities (P = <0.0001). More patients in arm B had Grade 3 or higher toxicity for leukopenia (74% vs. 47%), granulocytopenia (67% vs. 45%), thrombocytopenia (45% vs. 11%), infection (28% vs. 11%), liver dysfunction (26% vs. 4%), and ileus (12% vs. 0%), all of which were statistically significant (P < 0.05). One patient treated with VAD (multiple toxicities) and six patients treated with VAD plus valspodar had lethal toxicities (multiple toxicities in two and infection in three and one cerebral hemorrhage, P = 0.11). Severe, persistent ataxia did not occur in either arm.
|n = 47||n = 49|
|Grade (%)||Grade (%)|
|Worse degree (nonhematologic)||43||13||2||63||14||12|
Analysis of Multidrug Resistance Data
MDR-1 was assessed in 30 patients and MRP-1 in 24 patients. There were no statistically significant differences between arms in the MDR-1 or MRP-1 data, and no significant relation to objective response, subjective response, PFS, and OS. The small number of patients precluded a multivariate analysis of the association of MDR with treatment outcome, so only univariate analysis was done.
Analysis of Prognostic Factors
Sixty-nine patients had complete data for a Cox regression model for survival using treatment arm, age, plasma cell labeling index (PCLI), percentage of bone marrow plasma cells, beta-2 microglobulin (B-2M), C-reactive protein (CRP), soluble interleukin-6 receptor (SIL-6R), and interleukin 6 (IL-6) as explanatory variables. Only the treatment arm was statistically significant (P = 0.037), indicating a survival benefit in the VAD arm with a hazard ratio = 1.90 (95% CI, 1.03–3.47). When the stepwise variable selection method was used (Table 5), treatment arm, PCLI, and IL-6 were significant or marginally significant (P = 0.006, 0.064, 0.003, respectively). Similarly, a Cox regression model for PFS using the same covariates in 67 patients found the PCLI to be statistically significant (P = 0.03). When the stepwise variable selection method was used (Table 6), PCLI and CRP were marginally significant.
|Variable||P value||Hazard ratio||95% CI|
|Arm (VAD+ valspodar vs. VAD)||0.006||2.16||1.24–3.75|
|PCLI (≥ 0.60 vs. <0.60)||0.064||1.66||0.97–2.85|
|IL-6 (≥ 10.85 vs. <10.85)||0.003||2.39||1.34–4.24|
|Variable||P value||Hazard ratio||95% CI|
|PCLI (> 0.60 vs. < 0.60)||0.07||1.64||0.95–2.84|
|CRP (> 0.47 vs. < 0.47)||0.09||1.60||0.92–2.76|
Pharmacokinetic Studies of Doxorubicin
Pharmacokinetics study of doxorubicin infusion in 30 of the initial patients put on study revealed a statistically significant increase in doxorubicinol levels at both 48 and 96 hours in the patients who also received valspodar (P < 0.05). In addition, a preliminary look at the toxicity revealed a trend towards increased toxicity in arm B. The dose of doxorubicin was subsequently decreased to 6 mg/m2/24 hours in 15 of the 46 patients. Grade 3 or higher toxicity on the valspodar arm before doxorubicin reduction was 93% (n = 31) compared with 67% (n = 15) after reduction (P = 0.11).
The hypothesis of this study was that inhibition of the drug efflux pump would improve survival in patients with recurring or refractory myeloma. Valspodar had been shown to be effective ex vivo in reversing drug resistance16, 17 and in preliminary clinical studies to be both efficacious and minimally toxic.18–22, 24, 25 The study was closed early due to slow accrual. Although not statistically significant, there was an increase in partial response with the addition of valspodar (44% vs. 29%), but this was more than counteracted by an increase in toxicity, which led to a marginally inferior survival with the addition of valspodar (P = 0.055). Both arms had similar subjective response. Although there was a suggestion from ex vivo data that valspodar might delay expression of MDR,31 a higher objective response did not result in an improvement in PFS.
A Phase I study with VAD and valspodar determined the maximum tolerated doses of vincristine and doxorubicin with valspodar,24 but the subsequent pharmacokinetic studies and toxicity analysis revealed that the initial dose of doxorubicin was excessive. Although this was not statistically significant, the numbers for comparison were small. It appears likely that the excessive toxicity in arm B was due to the increased toxicity of doxorubicin and vincristine because valspodar inhibits the hepatic excretion of these drugs. The fact that six patients on arm B died of toxicity (before dose adjustment) compared with one patient in arm A contributed to the lack of benefit of valspodar.
In a previous Phase II study, valspodar did show improvement in some patients with myeloma who received the agent.25 Phase III studies in acute myelogenous leukemia (AML)32–34 did not show a statistically significant benefit with quinine or valspodar, although a benefit was shown with cyclosporin A modulation in a study of poor-risk AML.35 In this SWOG study, the rate of CR was not improved, but recurrence-free survival and OS were significantly increased. The dose of daunorubicin was the same in both arms, and steady-state concentrations of both daunorubicin and daunorubicinol were significantly higher in the cyclosporin A-treated arm. This suggests that the benefit was due to higher area under the curve of both daunorubicin and its metabolite, and therefore a higher effective dose of daunorubicin being administered to the cyclosporin A arm because of the reduced excretion of the chemotherapeutic agent. In our study, the dose of doxorubicin was decreased initially, based on Phase I data, yet the dose had to be further decreased based on blood levels obtained on the first 30 patients, and toxicity analysis suggesting higher levels correlated with higher toxicity. Even with this dose adjustment, toxicity was higher in the valspodar (B) arm. The poor results with PSC-833 might have been due to the combined effect of lower dose and higher toxicity, although the pharmacokinetics suggest the effective dose of adriamycin was excessive, not deficient.
Although limited by the number of patients (30) who had MDR assessed, a statistically significant improvement in response was not achieved in those patients who expressed MDR and received valspodar. Of the 30 patients who had MDR assessed, 17 had previously received neither vincristine nor doxorubicin, 11 had received vincristine alone, and no patient had received doxorubicin alone, with two receiving both. There were no differences in MDR levels according to the previous treatment and no differences in OR, SR, or PFS according to the previous treatment received. Because increased expression of P-gp is related to prior exposure to both vincristine and doxorubicin, it is likely that more patients would have expressed higher levels of P-gp, as evaluated by RT-PCR, if they had been previously exposed to these drugs, and a relation to OR, SR, of PFS may have been achieved. The RT-PCR technique used to evaluate for MDR was flawed26 because plasma cells were diluted with other bone marrow constituents, making quantification problematic. Alternative strategies, such as immunohistochemical techniques11 or functional assays36, 37 that identify the target cells, might have shown a better relation to treatment with valspodar but conclusions would still have been limited by small numbers.
Analysis of prognostic factors confirm previous data for untreated patients that PCLI, IL-6, and CRP are related to OS or PFS. The combination of valspodar and VAD was worse than VAD alone in the population of patients (69) that had data available for prognostic factor evaluation.
In conclusion, the multidrug modulating agent valspodar did not improve survival and increased toxicity in patients with recurring or refractory multiple myeloma despite the statistically insignificant improvement in partial responses. Survival was better with VAD than with valspodar/VAD due to the altered pharmacokinetics of doxorubicin and vincristine. Because this study was limited by accrual, definitive conclusions regarding the concept of reversing drug resistance, either through the mechanism of MDR or alternative mechanisms, cannot be definitively proven or disproven. Newer agents that reverse MDR but do not have a significant effect on the pharmacodynamics of the chemotherapy38, 39 may be able to improve survival in diseases that frequently express MDR, such as elderly patients with AML, or patients with recurring multiple myeloma.
- 14R-verapamil reverses drug resistance in some relapsed lymphoma patients. Anticancer Drugs. 1994; 5(Suppl 1): 72., , , et al.
- 18Clinical modulation of multidrug resistance in VAD-refractory multiple myeloma: studies with cyclosporin A and SDZ valspodar. Anticancer Drugs. 1994; 5(Suppl 1): 72., , , et al.
- 19A phase I trial of doxorubicin (DOX) and valspodar, a modulator of multidrug resistance (MDR). Anticancer Drugs. 1994; 5(Suppl 1): 42., , , et al.
- 20SDZ valspodar in combination with doxorubicin: a phase I and pharmacologic study in solid tumors. Anticancer Drugs. 1994; 5(Suppl 1): 42., , , et al.
- 22The pharmacokinetics and bioavailability of a new chemosensitizer, SDZ valspodar, in patients with advanced cancer. Anticancer Drugs. 1994; 5(Suppl 1): 44., , , et al.
- 24Phase I study of the chemosensitizer SDZ valspodar in combination with VAD chemotherapy to reverse multidrug resistance [abstract]. Blood. 1996; 88(Suppl 1): 662a., , , et al.
- 25Phase II study of PSC 833 (PSC) and VAD chemotherapy in patients with VAD-refractory multiple myeloma (MM). Blood. 1998; 92: 427a., , , et al.
- 29Regression models and life tables (with discussion). J R Stat Soc Ser B. 1972; 34: 187–220..
- 30East, v. 3.1. Cytel Software, Cambridge, MA, 2003.