Investigational New Drugs

, Volume 30, Issue 2, pp 714–722

A phase II open-label trial of bortezomib in patients with multiple myeloma who have undergone an autologous peripheral blood stem cell transplant and failed to achieve a complete response

Authors

    • US Oncology Research, Inc.
    • Rocky Mountain Cancer Centers
    • Rocky Mountain Cancer CentersUS Oncology Research, Inc.
  • Andrew Greenspan
    • US Oncology Research, Inc.
    • Central Indiana Cancer Centers
  • John F. Schwerkoske
    • US Oncology Research, Inc.
    • Minnesota Oncology Hematology
  • Romeo A. Mandanas
    • US Oncology Research, Inc.
    • Cancer Care Associates–West
  • Joe J. Stephenson
    • US Oncology Research, Inc.
    • Cancer Centers of the Carolinas
  • George T. Kannarkat
    • US Oncology Research, Inc.
    • Fairfax Northern Virginia Hematology Oncology
  • Feng Zhan
    • US Oncology Research, Inc.
  • Kristi A. Boehm
    • US Oncology Research, Inc.
  • Lina Asmar
    • US Oncology Research, Inc.
  • Roy Beveridge
    • US Oncology Research, Inc.
    • Fairfax Northern Virginia Hematology Oncology
PHASE II STUDIES

DOI: 10.1007/s10637-010-9556-6

Cite this article as:
Rifkin, R.M., Greenspan, A., Schwerkoske, J.F. et al. Invest New Drugs (2012) 30: 714. doi:10.1007/s10637-010-9556-6

Summary

Background A majority of multiple myeloma (MM) patients fail to achieve complete response (CR) to peripheral blood stem cell transplantation (PBSCT); effective options following autologous transplantation are needed. Bortezomib (B) is active against MM. This study was conducted to determine the feasibility, safety, tolerability, and efficacy of B following high-dose melphalan therapy and PBSCT. Methods Fifty patients enrolled (48 evaluable) and 49 were treated (safety population). Treatment: 4 cycles B 1.3 mg/m2 Days 1, 4, 8, and 11/21-days; 4 additional cycles were permitted for stable or responding patients. Results Median age was 55 years (range, 38–73), 68% male, 64% ECOG PS = 0, 44% Durie-Salmon Stage IIIA prior to induction, 42% had symptomatic IgG MM; 74% had prior single transplant (26% tandem). Responses post-transplant: 70% PRs, 18% MRs. A median of 4 cycles (range, 2–8) of B were administered. Responses: CR 8%, uCR 2%, PR 23%, uPR 19%, MR 10%, and no change 35%; median time-to-treatment failure (TTF) was 6.2 months (range, 1.0–19.4). Three deaths occurred (n = 1 sepsis, n = 2 disease progression). Grade 3–4 treatment-related toxicities included: thrombocytopenia, neuropathy (14%, each); asthenia, neutropenia (10%, each); and nausea (4%). Twelve patients (24%) discontinued treatment due to toxicity and 30 patients (60%) completed the study; 20 patients started new treatment (median 5.8 months [range, 1.5–20.3]). Conclusions The study closed early due to widespread availability of B, and the lack of B-naïve patients. Bortezomib monotherapy after melphalan and autologous PBSCT was feasible, safe and well-tolerated (toxicities were manageable), but failed to produce the hypothesized response rates.

Keywords

AutologousBortezomibCommunity-basedMulticenterPeripheral blood stem cell transplantProteasome inhibition

Introduction

A significant number of multiple myeloma (MM) patients who receive high-dose melphalan therapy followed by autologous peripheral blood stem cell transplantation (PBSCT) fail to achieve a complete response (CR) [1]. Single-agent bortezomib has been shown to be an active agent in the treatment of relapsed and/or refractory MM, with reported CR (immunofixation (IF) negative) plus near CR (IF positive) rates near 10% [2]. The later publication of the APEX trial [3] also noted similar results, although presented after the current study was well under way.

The present study integrated maintenance therapy with bortezomib into the treatment plan for myeloma patients who have received high dose melphalan therapy followed by PBSCT and failed to achieve a CR. Based on Richardson’s work [2], it was hoped that bortezomib would convert 10% of this patient population with minimal residual disease (MRD) [defined according to the criteria established by Blade et al. [4] as patients with a partial response (PR), minimal response (MR), or no change (NC)] to complete responders (CR IF-) or to near complete responders (CR IF+) as described by Richardson [2]. Prolonging the progression-free survival (PFS) was also a goal of the study. Due to the need to determine the role of bortezomib therapy following stem cell transplantation in MM [5], the feasibility of the administration single-agent bortezomib after transplantation was explored.

Patients and methods

Study design

This was a Phase II open label study in patients with multiple myeloma. The primary objective of the study was to determine if bortezomib treatment starting 60–120 days after high dose melphalan therapy followed by an autologous PBSCT would convert patients with minimal residual disease (ie, less than a complete response [PR, MR, NC]) to complete responders (CR IF-) according to the criteria published by Blade et al. [4] or near complete responders (CR IF+) as described by Richardson [2]. The secondary objective of this study was to evaluate the safety and tolerability of the bortezomib regimen in this patient population and setting.

Patients received 1.3 mg/m2 of bortezomib (VELCADE, Millennium Pharmaceuticals, Inc., Cambridge, MA) via IV push over 3–5 s on Days 1, 4, 8, and 11, every 21-days for 4 cycles. Dexamethasone administration with bortezomib was not permitted. Patients could receive up to 4 additional cycles if a response was noted after the completion of Cycle 4. Antiviral prophylaxis utilizing acyclovir, 400 mg orally twice daily during the study, starting on Study Day 1, was required to prevent herpes zoster reactivation.

The protocol was approved by a Central Institutional Review Board with jurisdiction over specific sites that registered patients on study. All patients were required to sign an informed consent form prior to any study-related testing being performed.

Patients

Eligible patients had to have all of the following:(1) a confirmed diagnosis of symptomatic MM and (2) undergone high-dose chemotherapy (melphalan 200 mg/m2/transplant) followed by 1 or 2 autologous PBSCTs. Patients may have received any number of prior therapies prior to SCT provided they had received an autologous PBSCT and had PR, MR, or NC ≥60 to ≤120 days post single or double PBSCT for MM without progression of disease during the study window. Additionally, successful completion of a standard mobilization/apheresis regimen with the collection of a minimum of 2.5 × 106 CD34+ cells/kg recipient weight was required. ECOG PS of 0 or 1; absolute neutrophil count (ANC) values, on 3 consecutive days, ≥500/mm3 to ensure engraftment; and, normal renal and hepatic function with platelets ≥50,000 cells/μL without a platelet transfusion for last 7 days, ANC ≥1,500 cells/μL, and hemoglobin ≥8.0 g/dL were all required.

Patients were excluded if they had any of the following: non-secretory multiple myeloma, prior allogeneic transplant; history of hypersensitivity to components of bortezomib (boron or mannitol), receiving concurrent immunomodulatory therapy (ie, lenalidomide or thalidomide) or any treatment for myeloma post-autografting. Any evidence of central nervous system (CNS) involvement; any uncontrolled intercurrent medical or psychiatric illness, or serious infection; known HIV positive; history and/or symptoms related to a cardiovascular accident within the prior months; a myocardial infarction within 6 months prior to enrollment or had New York Heart Association (NYHA [6]) Class III or IV heart failure, uncontrolled angina, severe uncontrolled ventricular arrhythmias, or electrocardiographic evidence of acute ischemia or active conduction system abnormalities at the time of enrollment; or had a history of other malignancy within the last 5 years.

Assessments

A comprehensive re-evaluation of the patient’s disease response was conducted at study entry utilizing the EBMT response criteria. The following were done at study entry (following PBSCT): complete medical history; brief physical examination; blood work including CBC, CMP, LDH, phosphorous, calculation of ClCr; assessment of serum β2-microglobulin; serum electrophoresis (EP) (24-h urine was done if M-protein had been detectable in urine previously); serum and urine immunofixation electrophoresis (IFE), for patients with no detectable M-spike in serum or in urine EP; serum immunoglobulins (IgA, IgM, IgG); bone marrow examination (biopsy and aspirate) for histopathology and cytogenetics; EKG if clinically indicated; lytic bone skeletal survey; and quality of life assessment via FACT & GOG-NTx [7]. The majority of assessments were repeated prior to each cycle; skeletal surveys were repeated every 4 cycles to accommodate the study design and correspond to serial determinations of response. Bone marrow biopsies were done only to confirm a CR. Assessments at the end of treatment and in the follow-up phase (1-year following last dose) were similar to those described above.

Criteria for assessing response and toxicity

This study utilized the European Group for Blood and Marrow Transplant (EBMT), International Bone Marrow Transplant Registry (IBMTR) and Autologous Blood and Marrow Transplant Registry (ABMTR) criteria adapted from Blade [4] to define response, relapse and disease progression. Best responses were determined using the sequence of disease status with corresponding best response.

Toxicities were graded according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events (NCI CTCAE) version 3.0 [8].

Statistical analysis

The primary objective of this Phase II, open-label study of bortezomib was to determine the conversion rate to complete response (CR) rate in patients with multiple myeloma who had undergone high-dose melphalan followed by autologous PBSCT and failed to achieve a CR. Using chi-square (χ2) test, with a predetermined significance level α = 0.05 and a desired power of 80% (ie, β = 0.20), a sample size of 68 was needed to reject the null hypothesis H0: P = 10% in favor of the alternative hypothesis HA: P = 20%, where “P” stands for response rate.

Based on the EBMT criteria, a patient’s best overall response was determined using 7 categories including CR, unconfirmed CR (uCR), partial response (PR), unconfirmed PR (uPR), minimal response (MR), no change (NC), and progressive disease (PD). For each of the 7 categories, a rate was calculated with a 95% CI using exact binomial method. This analysis was done using the evaluable population (all assessable, eligible patients who received at least 1 dose of study treatment).

Progression-free survival (PFS) was defined from the treatment start date (date of first dose) to either the date the patient was first recorded as having disease progression (even if the patient went off treatment because of toxicity), or the date of death if the patient died due to any causes before progression. Survival was measured from the date from start of treatment with B to the date of death or the last date the patient was known to be alive and overall survival (OS) was also used to evaluate the efficacy of B. Kaplan-Meier method [9] was used for both PFS and OS. PFS, median PFS, and PFS rates were calculated at 3-month timepoints with 95% CI provided. Similar methods were applied for analysis of OS. As this was a single-arm study, only descriptive statistics are presented; analysis of key variables was limited to what is presented here.

In addition, “quality-of-life” (QOL), was also analyzed using the ITT population. QOL data were collected from registration to the end of treatment using FACT&GOG-Ntx (version 4) [7] questionnaires; 7 subscales of QOL scores were calculated at baseline, each treatment cycle, and end of study. For each treatment cycle (and end of treatment), all subscale scores were compared with the patient’s baseline score(s) using Wilcoxon signed-rank test.

Dosage, dose intensity, and dose changes made were summarized using descriptive statistics (mean, median, and range) and frequency tables. Adverse events, coded in COSTART terms and graded by NCI CTCAE (version 3) [8], were recorded throughout the whole trial, and those adverse events considered drug-related were tabulated by COSTART term and highest grade. The safety population, defined as all registered patients who received at least 1 dose of study treatment, was used for these analyses.

Results

Patient characteristics

Between March 2004 and November 2007, 50 patients (intent-to-treat population) enrolled at sites throughout the US Oncology Transplant Network; 49 patients were treated (safety population). The study was closed earlier than planned, before the required 68 patients could be enrolled because of accrual difficulties (difficulty identifying B-naïve patients). Patient demographics at baseline, including prior therapies are summarized in Table 1.
Table 1

Patient characteristics at baseline

Number of Subjects Enrolled

50

Age (Years)

 Median

55.9

 Range

38.8–73.5

 

Number and Percentage (%) of Subjects

Sex

 Male

34 (68)

 Female

16 (32)

Race

 White

41 (82)

 Black

8 (16)

 Hispanic

1 (2)

ECOG Performance Statusa

 0

32 (64)

 1

18 (36)

Prior Therapyb

 Any

50 (100)

 Chemotherapy

48 (96)

 Radiotherapy

13 (26)

 Surgery

10 (20)

 Single transplant

37 (74)

 Tandem transplant

13 (26)

Durie-Salmon Stage at Baseline

 IA

7 (14)

 IIA

13 (26)

 IIB

1 (2)

 IIIA

22 (44)

 IIIB

7 (14)

Type of Myeloma

 IgG/K

21 (42)

 IgG/L

9 (18)

 IgA/K

9 (18)

 IgA/L

1 (2)

 Kappa

3 (6)

 Lambda

5 (10)

 unknown

2 (4)

a Assessed prior to first treatment with study drug

b Subjects may have had more than 1 prior therapy.

Treatment outcomes

Overall, 10% of patients had complete responses (8% confirmed, 2% unconfirmed) as their best response; 23% had PRs and another 19% had uPRs. Patients missing assessments used to confirm response were labeled as “unconfirmed” for whatever best response was observed. Overall, the response rate was 31% confirmed and 21% unconfirmed (total 52%). Disease stabilization was seen in 35% of patients, who had no change noted in disease status from baseline to end of treatment. Median time to response was 2.6 months (range, 0.7–4.3); the median duration of response has not yet been reached as <50% of patients have progressed to date. Response durations have ranged from 9.3 months to 17.7 months; at 12-months 100% of responses were durable. Best response, reasons for treatment discontinuation, and patient status are summarized in Table 2.
Table 2

Best response after treatment and reasons for treatment discontinuation

Total Number of Patients

50

 

Number of Assessable Patients

48a

 
 

Number and Percentage (%) of Subjects

Best Response

N (%)

95% CI

 Complete response (CR)

4 (8.3)

2.3, 20.0

 Unconfirmed complete response (uCR)b

1 (2.1)

0.1, 11.1

 Partial response (PR)

11 (22.9)

12.0, 37.3

 Unconfirmed partial response (uPR)b

9 (18.8)

8.9, 32.6

 Minimum response (MR)

5 (10.4)

3.5, 22.7

 No change (NC)

17 (35.4)

22.2, 50.5

 Progressive disease (PD)

0

 Best response not evaluable (NE)

1 (2.1)

−.1, 11.1

Time to Response (months)

 Median

2.6

 

 Range

0.7–4.3

 

Duration of Response (months)

 Median

Not yet reached

 

 Range

9.3–17.7

 

Status

 Alive

47(94.0)

 

 Dead

3(6.0)

 

Cause of Death

 Progressive disease

2

 

 Sepsis

1

 

Reason for Discontinuation

 Normal study completion

30 (60.0)

 

 Adverse eventc

12 (24.0)

 

 Patient request

3 (6.0)

 

 Progressive disease

2 (4.0)

 

 Deemed ineligible after registration

2 (4.0)

 

 Investigator decision

1 (2.0)

 

Total cycles received

Patients treated

49

 

 Median

4.0

 

 Range

2–8

 

a2 patients were not included in the evaluable population; both were ineligible (one received no treatment and the other was determined to be ineligible after receiving 4 cycles).

bUnconfirmed responses were required when confirmatory assessments could not/were not performed in a timely manner

cAEs that were responsible for discontinuation of treatment included: gait abnormal (n = 1), asthenia with neuropathy (n = 1), asthenia with constipation (n = 1), neuropathy (n = 4), nausea with diarrhea (n = 1), neuritis peripheral with diarrhea (n = 1), neuritis peripheral with neuropathy (n = 1), unrelated pneumonia (n = 1), and unrelated carpal tunnel syndrome (n = 1).

A separate analysis of response based on stratification by transplant status (single [n = 35] vs. tandem [n = 13]) was done; however, no statistical analyses could be done as the study was not powered for this subanalysis. There were some differences between these subgroups, with regard to response, based on transplant status. Namely, more than twice the proportion of patients with tandem transplants had PR compared to single transplant patients (38.5% vs 17.1%); 1 tandem patient had a uCR while no single transplant patient did, and no tandem patient had MR while 14% of single transplant patients did. Complete response results for single/tandem patients were: CR (8.6%/7.7%), uCR (0%/7.7%), PR (17.1%/38.5%), uPR (20.0%/15.4%), MR (14.3%/0%), NC (37.1%/30.8%), no PD, and not evaluable (2.9%/0%).

Median duration of follow-up for all patients was 15.9 months (range, 3.2–23.1 months). Median OS and PFS have not been reached; however, OS ranged from 4.5 to 23.0 months and PFS ranged from 1.3 to 20.3 months. At 12-months, OS was 95.8% (95% CI 84.4, 98.9) and PFS was 91.4% (95% CI 78.6, 96.7). Details of OS and PFS at 3-month intervals are provided in Fig. 1a and b; time to treatment failure is summarized in Fig. 2.
https://static-content.springer.com/image/art%3A10.1007%2Fs10637-010-9556-6/MediaObjects/10637_2010_9556_Fig1_HTML.gif
Fig. 1

Estimated Overall Survival and Progression-free Survival (per protocol population)

https://static-content.springer.com/image/art%3A10.1007%2Fs10637-010-9556-6/MediaObjects/10637_2010_9556_Fig2_HTML.gif
Fig. 2

Estimated Time to Treatment Failure (per protocol population) a progression, death, off treatment other than normal completion, start new treatment

Drug delivery

The median number of cycles completed was 4 (range, 2–8). The median relative dose-intensity of bortezomib was 85% (range, 45–111). Doses were delayed or reduced in 43% of patients. These delays or reductions were largely attributed to toxicity, specifically neuropathy, thrombocytopenia, neutropenia, and asthenia.

Toxicity

Toxicity in this study was infrequent most events were mild or moderate (Grade 1–2). The most frequent AEs were hematologic or neurologic in nature, except for nausea. Grade 3–4 treatment-related adverse events are summarized in Table 3. Treatment discontinuation due to AEs occurred in 24% of ITT patients, including 2 patients (n = 1 each, pneumonia and carpal tunnel syndrome that were unrelated to study treatment). Other reasons for discontinuation are detailed in Table 3.
Table 3

Treatment-related Toxicities Reported in ≥2 patients, (n = 49)

Adverse Event

Grade 3

Grade 4

Total (%)

Gr 1–4%

Hematologic

 Neutropenia

5

0

5 (10.2)

14.3

 Thrombocytopenia

4

3

7 (14.3)

20.4

Nonhematologic

 Asthenia

5

0

5 (10.2)

16.3

 Nausea

2

0

2 (4.1)

12.2

 Neuropathy

6

1

7 (14.3)

26.5

Quality of life

Of the various subscales that comprise the FACT&GOG-Ntx QOL assessment tool, Social Well-being reached significant declines (P = 0.05) at Cycles 3 and 8. Physical Well-being was significantly decreased at Cycle 4 (P < 0.01), and Emotional Well-being was significantly increased in Cycles 1–4 compared to baseline (P = 0.02–0.04). Total Outcome Index reflected a significant (P = 0.01) decrease at Cycle 4 and the overall assessment of QOL reflected a trend towards decrease (with the exception of Emotional Well-being) throughout the course of the study, albeit, generally not significant. It should be noted that 21 patients received no more than 4 cycles of study treatment; thus, the number of QOL questionnaires completed in Cycle 5 were drastically reduced (from 41 in Cycle 4 to 21 in Cycle 5). Changes in QOL are reflected graphically in Fig. 3.
https://static-content.springer.com/image/art%3A10.1007%2Fs10637-010-9556-6/MediaObjects/10637_2010_9556_Fig3_HTML.gif
Fig. 3

Quality of Life (intent-to-treat population) FACT&GOG-NTX Functional Assessment of Cancer Therapy (specific for patients with neurotoxicity) SWB Social Well-Being FACT-G Functional Assessment of Cancer Therapy (General Version) EWB Emotional Well-Being TOI Trial Outcome Index (PWB+FWB+additional concerns) PWB Physical Well-Being FWB Functional Well-Being

Discussion

With the introduction of novel agents such as thalidomide, lenalidomide, liposomal doxorubicin, and bortezomib, many options are now approved for the treatment of multiple myeloma; however, in 2003 there was a paucity of data for the treatment of myeloma patients in the post-transplant setting who have failed to achieve a complete response. Contemporary combination chemotherapies employing novel agents now have overall response rates in excess of 50%, and are being studied extensively in the post-transplant setting. Multiple studies have now examined the role of improving response rates with stem cell transplantation, and have shown that converting patients to complete responders may improve overall survival.

Patients in the current study had been previously treated with high-dose melphalan followed by either single or tandem transplant and had failed to achieve a CR following that procedure. Such patients may be expected to fare poorly when compared to those that obtained a complete response with transplant. Since all patients had been previously treated with high-dose melphalan followed by transplant, the investigators were challenged in treating these transplant-refractory patients given their multiple prior regimens of and intensity of chemotherapy; it was not possible to separate whether any observed disease stabilization was due to PBSCT or B. At the inception of the study, bortezomib as post-transplant therapy had not been previously investigated and Durie-Salmon staging was the standard (ISS had not yet become the norm). The safety, efficacy, and toxicity of bortezomib had been explored as monotherapy or in combination with dexamethasone prior to giving post-transplant combination therapy [10]. Additionally, Uy and colleagues [11] had included bortezomib as part of the pre-transplant regimen (prior to melphalan) and again post-transplant weekly 5 of 6 weeks for 6 cycles. Uy’s study produced ORR of 83% but cannot be compared to the current study as our patient population consists of patients who did not respond completely to PBSCT.

During conduct of this study bortezomib was approved for the treatment of newly diagnosed and previously treated multiple myeloma, thus only 50 of a planned 68 patients were enrolled. This is 18 fewer patients than what was required to power the study. Given the fact that bortezomib failed to induce a CR in 10% of patients as planned (CR rate was 8.3%); the ORR in the current study, including unconfirmed responses was 52.1%, minimal responses were recorded in 10.4% of patients and no change (disease stabilization) occurred in 35.4%. The study was not powered at 80%, but rather only had 71% power.

The standard dosing schema with bortezomib was used in the current study and was based upon previously published work [12]. Although the toxicities were generally mild to moderate, the patients assessed their quality of life as remaining generally stable with slight, although not significant, decreases. The significant increase in emotional well-being during the first 4 cycles of treatment suggests that patients remained hopeful during the course of their treatment.

Currently, there are multiple treatments available for previously-treated MM including pegylated liposomal doxorubicin plus bortezomib and immunomodulatory agents plus bortezomib with or without dexamethasone.Intravenous bisphosphonates are approved for newly diagnosed (untreated) MM. Many agents are currently being developed and investigated for use in MM [13, 14].

With the advent of novel agents and the introduction of maintenance therapy, the treatment of multiple myeloma has become increasing complex and in a constant state of evolution. In the future, autologous stem cell transplantation should be viewed as a “modality of therapy” and the use of it will be integrated within the clinical context of the natural history of the disease. Most likely, it will be employed after achieving at least a very good PR following an induction therapy, and patients will then likely be “maintained” with one or more simple combination therapies post-remission. The results provided in this study will assist in the design of post-transplant therapy and will form the foundation for the development of more effective post-transplant combination therapies which are now being shown to deepen responses further.

Although this study was terminated with fewer than planned patients, the data can provide a foundation for developing more active combination therapies in the post-transplant setting in an attempt to improve conversions of suboptimal responders to complete responders, thereby improving overall and progression-free survival. In addition, this would translate to a longer time of therapy, and prolonged time to next therapy. It is postulated that the low response rates observed in this study are in part due to selecting patients with an overall poor prognosis and no prior therapy known to overcome unfavorable cytogenetics given their less than optimal responses to transplantation. These prognostic factor analyses are becoming the standard of care and will allow us to select more active combination therapies, which will be personalized for the treatment of MM.

Acknowledgments

We thank the patients who shared their experiences with US Oncology physicians (see Appendix), the site coordinators in the field (especially Kim K. Seeley, BSN, OCN and Nichole M. Stephens, BSN), and program manager Julie Boston, RN; project managers Sharon Ambro, RN and Mary Ann Rauch, BS; and, clinical data specialist Tracy Locke, BA, RHIA, who all ensured the accuracy of the data.

Conflict of Interest Disclosures

The following disclosures apply: Dr. Rifkin is on the Speakers’ Bureau and Advisory Board for Millennium. All other authors have nothing to disclose.

Manufacturer

Bortezomib (Velcade®) is produced by Millennium Pharmaceuticals, Inc., Cambridge, MA.

Copyright information

© Springer Science+Business Media, LLC 2010