Pomalidomide, cyclophosphamide, and prednisone for relapsed/refractory multiple myeloma: a multicenter phase 1/2 open-label study

Multiple myeloma (MM) is characterized by a clonal proliferation of malignant plasma cells in the bone marrow and osteolytic lesions.¹  The introduction of novel agents, such as thalidomide, lenalidomide, and bortezomib, has considerably improved response rates and survival, both at diagnosis and at relapse.² 

However, MM remains incurable, and the majority of patients relapse and become refractory to available therapies. The outcome of these patients is very poor, with a median event-free survival of 5 months and overall survival (OS) of 9 months.³  Newer agents able to overcome drug resistance and achieve a sustained disease control are needed.

Several immunomodulatory derivatives were generated by introducing chemical modifications to the structural backbone of thalidomide. Pomalidomide, a closely related analog of thalidomide, showed potent activity against tumor necrosis factor α secretion and entered clinical studies.^4,5  Pomalidomide at 2 to 4 mg, combined with low-dose dexamethasone, has shown significant activity in pretreated patients refractory to lenalidomide and/or bortezomib.^6-10  Partial response (PR) rates were 32% to 35%, and median progression-free survival (PFS) was 4.6 to 6.3 months.^8-10  Neutropenia and thrombocytopenia were the most frequent adverse events.¹¹ 

The addition of alkylating agents to bortezomib and lenalidomide increased the response rates and, in some cases, prolonged disease-free interval.¹²  So far, no data on the role of cyclophosphamide added to pomalidomide are available.

These observations provided the rationale for this phase 1/2 trial. The primary aim of the study was to identify the most appropriate dose of pomalidomide in combination with cyclophosphamide-prednisone (PCP) and to determine its safety, tolerability, and efficacy in MM patients relapsed and/or refractory to lenalidomide.

Patients with MM who were 18 years of age or older, were relapsed or relapsed/refractory to lenalidomide, and who had received 1 to 3 prior lines of therapy were eligible. Relapse was defined as the reoccurrence of disease requiring the initiation of a salvage therapy, and refractory disease was defined as relapse while receiving salvage therapy or progression within 60 days of the most recent therapy. Patients were required to have measurable disease, a Karnofsky performance status 60% or higher, a platelet count 50 × 10⁹/L or higher, a neutrophil count of 1.00 × 10⁹/L or higher, a corrected serum calcium 3.5 mmol/L (14 mg/dL) or lower, serum hepatic aminotransferase levels 2.5-fold or less of the upper limit of normal, total bilirubin 1.5-fold or less of the upper limit of normal, and serum creatinine 2 mg/dL or less, and to agree to use contraception. Patients with clinically relevant active comorbid medical or psychiatric conditions or history of malignancy within the last 5 years were excluded. The institutional review board at each participating center approved the study in accordance with the Declaration of Helsinki. All patients provided written informed consent.

This trial was a phase 1/2, dose-escalating, open-label study. The primary endpoint of the dose-finding phase 1 was to identify the maximum tolerated dose (MTD) of pomalidomide, defined as the dose that achieved a dose-limiting toxicity (DLT) in 25% of patients. DLTs were defined as grade 4 neutropenia lasting more than 3 days, other grade 4 hematologic toxicity, any grade 3 or higher nonhematologic toxicity, febrile neutropenia, and/or infection requiring antibiotics occurring during the first cycle of therapy.

In phase 2, patients received the MTD of pomalidomide established in phase 1. The primary endpoint of phase 2 was the rates of complete response (CR) and very good PR (VGPR). Secondary endpoints were PFS and OS.

All adverse events were assessed during each cycle and graded according to the National Cancer Institute Common Toxicity Criteria (version 3.0).¹³ 

Responses were recorded during every cycle, according to the International Myeloma Working Group criteria.¹⁴  Responses among patients refractory to previous novel agents and at high risk were analyzed. High risk was defined as the presence of t(4;14), t(14;16), or del 17p13 at enrolment, detected by fluorescence in situ hybridization (FISH).

Oral pomalidomide was administered at doses ranging from 1 to 2.5 mg/d in phase 1 and the MTD in phase 2, in combination with cyclophosphamide at 50 mg every other day and prednisone at 50 mg every other day on days 1 to 28, for 6 cycles of 28 days each (see supplemental Appendix 1 available on the Blood Web site). Maintenance therapy consisted of pomalidomide 1 mg/d and prednisone 25 mg every other day continuously until any sign of relapse or progression.

Pomalidomide dose reduction (from 2.5 to 2 to 1.5 to 1 to 0.5 mg/d) was allowed if toxicities occurred. Grade 4 neutropenia, or febrile neutropenia with any other hematologic toxicities, or any other grade 4 hematologic or any grade 3 nonhematologic toxicities required immediate interruption of treatment and subsequent dose reduction at the start of the following cycle. A new cycle could be started if the neutrophil count was 1.00 × 10⁹/L or higher, platelet count was 50 × 10⁹/L or higher, hemoglobin was 8 g/dL or higher, and nonhematologic adverse events were grade 2 or lower. If the start of a new cycle were delayed by 2 or more weeks, dose reductions were required. Aspirin 100 mg/day or low-molecular-weight heparin was recommended as prophylaxis, according to patient risk.¹⁵ 

In phase 1, the continual reassessment method was used as the dose allocation rule in the trial.^16,17  It is based on a mathematical modeling of dose–DLT relationship, iteratively updated using Bayes theorem. Before trial onset, prior opinions about DLT probability at each dose level were elicited from expert clinicians and were fixed at 0.15, 0.20, 0.30, and 0.45, respectively. A design with grouped inclusions of 4 patients per dose level was chosen; the starting dose was 1.5 mg. The dose level associated with an updated DLT probability close to 25% was administered to the next cohort. All this process was re-run until the fixed sample size (N = 24) was reached,¹⁸  using the BPCT software.¹⁹ 

In phase 2, we used a Simon optimal 2-stage design for the sample size calculation. A 5% response rate was considered not promising, a 20% rate was promising. The probability of both type I and type II errors was set at 0.05. Accordingly, 24 patients were planned in the first stage, and 31 (total = 55) were planned in the second stage. Our design required VGPR or better in at least 2 patients in the first stage to proceed to the second stage, and at least 5 patients for the treatment to be worth further consideration. Patients enrolled at the MTD in phase 1 were also included in phase 2.

All patients meeting the eligibility criteria who had received at least a single dose of pomalidomide were evaluated for response, toxicity, and survival. For responding patients, we measured the median time to response from the start of treatment to the date of the first response, as well as the duration of response from response to PD or death, censored at the date of last assessment for patients not progressing. We evaluated PFS and OS from the start of treatment until PD or death and until death, respectively.

Survival curves were estimated with the Kaplan-Meier method and compared with log-rank test.²⁰  The individual effects on PFS of age (>75 vs ≤75 years), FISH-defined risk (high vs standard), and achievement of at least PR (treated as a time-dependent variable) were evaluated using a Cox's model. Results are presented as hazard ratios (HRs) and 95% confidence intervals (95% CIs). The analyses were performed using SAS software, version 8.2 (SAS Institute). Data cutoff was October 16, 2012.

The pharmaceutical sponsor was not involved in the study design, collection, analysis, or interpretation of the data or the writing of the report. Celgene supplied pomalidomide free of charge. The corresponding author had full access to all the data and had final responsibility for the decision to submit this manuscript for publication.

Between August 2010 and May 2012, 69 patients were enrolled at 12 Italian centers. In the phase 1 study, 24 patients were accrued. In the phase 2 study, 12 patients who received the MTD during phase 1 and an additional 45 patients were enrolled. Two patients who rapidly developed PD and died were excluded from the analysis because they failed to start therapy.

Patient characteristics are listed in Table 1. The median age was 69 years (range, 41-84 years). The median number of previous treatments was 3 (range, 1-3 treatments). Nine percent, 29%, and 62% of patients had received 1, 2, and 3 prior regimens, respectively. All patients (100%) had previously received lenalidomide, 84% bortezomib, 20% thalidomide, 33% autologous transplantation, and 14% allogeneic stem-cell transplantation. Twenty-three patients were relapsed and 46 were refractory to lenalidomide; 22 were refractory to both lenalidomide and bortezomib. The median time from diagnosis to study entry was 53 months (range, 11-203 months). Eighteen patients (26%) were classified as high-risk by FISH.

Table 2 lists the assigned pomalidomide dose levels and the observed DLTs. The dose level 2.5 mg/d was defined as the MTD, with an estimated probability of DLT of 0.258 (95% credibility interval, 0.101-0.468). DLTs were recorded in 4 patients: 1 grade 4 thrombocytopenia with 1.5 mg/d pomalidomide, 1 grade 3 peripheral neuropathy, 1 grade 3 hepatic toxicity, and 1 grade 4 thrombocytopenia, with 2.5 mg/d pomalidomide. The dose of pomalidomide maintenance was increased to the identified MTD.

Responses and time to event analysis during phase 1 are reported in Table 3 and Table 4.

Fifty-five patients treated at the MTD (2.5 mg/d) were evaluated. Patients received a median of 6 cycles (range, 1-6 cycles). Five patients did not complete the assigned 6 cycles for toxicity: grade 3 cutaneous rash and grade 2 pancreatitis (1 patient), grade 2 bradycardia and dyspnea (1 patient), grade 5 sepsis (1 patient), grade 4 deep vein thrombosis (1 patient), and grade 3 liver failure (1 patient). Eight patients did not complete salvage therapy for PD, 1 patient for a concomitant mesothelioma, and 1 patient skipped the last cycle and started maintenance for medical reasons. Thirty-four patients started maintenance treatment; 13 of them discontinued therapy for PD and 2 for toxicity, including pulmonary embolism and limbic encephalitis (1 patient each) (Figure 1).

At the end of the first stage of the phase 2 portion, 2 CRs and 3 VGPRs were observed, allowing us to proceed with the second stage.

At least PR was achieved in 28/55 patients (51%), at least VGPR was achieved in 13/55 patients (24%), and immunofixation-negative CR was achieved in 3/55 patients (5%). A high proportion of patients achieved a clinical benefit, with at least a minimal response (MR) in 39/55 patients (71%) and at least stable disease in 54/55 patients (98%) (Table 3). The median time to at least PR was 1.83 months (range, 0.65-6.4 months); at least PR was achieved in 19 patients after 2 cycles, in 20 patients after 4 cycles, and in 23 patients after 6 cycles. The median duration of response for the 28 responding patients has not been achieved yet.

The median follow-up from study entry was 14.8 months (range, 6-21 months). At the time of the analysis, 40 patients were alive, 26 had progressed, and 15 had died from PD (10 patients), pneumonia and respiratory failure (1 patient), sudden death (1 patient), sepsis (1 patient), liver failure (1 patient), and mesothelioma (1 patient). The 1-year PFS was 48% (95% CI, 33%-62%), with a median of 10.4 months (95% CI, 7.9-15.8 months; Figure 2A). The 1-year OS was 69% (95% CI, 54%-81%), and the median value was not reached (Figure 2B).

The 1-year PFS was 72% for patients relapsed after lenalidomide and 37% for those refractory to lenalidomide (P = .22; Figure 3A). The 1-year PFS was 68% in patients who achieved at least PR and 26% in those who achieved less than PR (P = .02; Figure 3B). The 1-year PFS was 47% in standard-risk and 35% in high-risk patients (P = .21; Figure 3C).

In a multivariable analysis, older age negatively affected PFS (HR, 2.65; P = .035), and the achievement of PR confirmed its positive effect on PFS (HR, 0.38; P = .059) (Figure 3B). No differences according to FISH-defined risk were noted (HR, 1.36; P = .48).

At the MTD, the most frequent grade 3 to 4 adverse events were neutropenia (42%), thrombocytopenia (11%), anemia (9%), neurologic (7%), dermatologic reactions (7%), and infections (5%). Two grade 5 infections were reported (Table 5). Grade 3 peripheral neuropathy was detected in 2 patients. Dermatologic events were mild to moderate and were manageable with pomalidomide dose reduction and corticosteroids. Grade 4 deep vein thrombosis was reported in 1 patient, despite low-molecular-weight heparin prophylaxis. Pomalidomide dose was reduced in 17 patients for grade 4 hematologic toxicity (4 patients), grade 1 to 2 nonhematologic toxicity (2 patients), grade 3 to 4 nonhematologic toxicity (7 patients), and unknown causes (4 patients). During maintenance treatment, the frequency of new grade 3 to 4 adverse events was low and included grade 4 neutropenia (2 patients), pulmonary embolism (1 patient, while receiving aspirin), and grade 3 neurologic toxicity (vertigo, peripheral neuropathy, and limbic encephalitis, 1 patient each).

In this study, we evaluated dosing, the safety profile, and the efficacy of the combination of PCP in patients who were relapsed/refractory to multiple lines of treatment, including lenalidomide. At the MTD (2.5 mg/d pomalidomide), the at least PR rate was 51%, and the median PFS was 10.4 months. Adverse events were mainly hematologic: the rate of grade 4 neutropenia was 16%, and the rate of grade 4 thrombocytopenia was 5%. A Bayesian adaptive design for dose finding was implemented. This approach is expected to replace classic dose-finding schemes because it enables more patients be treated at near-optimal doses while controlling excessive toxicities.

Patients who experience multiple relapses and become refractory to current salvage treatments have virtually no treatment options. Responses after relapse are generally short-lived, and outcomes can be affected by comorbidities, adverse chromosomal abnormalities, and the toxicity of the previous treatments.^21-23  A recent survey on 286 relapsed myeloma patients who were refractory to bortezomib and relapsed/refractory to an immunomodulatory drug reported at least MR in 44% of patients, including 32% PR, and median event-free survival and OS of 5 and 9 months, respectively.³ 

Initial reports demonstrated encouraging activity with pomalidomide alone or with low-dose dexamethasone in relapsed/refractory myeloma.^6-12,24,25  In a phase 3 study comparing pomalidomide (4 mg/d for 21 days in a 28-day cycle) plus low-dose dexamethasone (160 mg monthly) with high-dose dexamethasone (480 mg monthly) in multirelapsed/refractory myeloma, median PFS was 3.9 vs 2 months (P < .001), and median OS was not reached vs 8.5 months (P < .001), in the pomalidomide and the high-dose dexamethasone groups, respectively.²⁶ 

In a previous study, pomalidomide-dexamethasone induced at least PR rate of 32%, at least MR rate of 47%, and median PFS of 4.8 months in patients refractory to lenalidomide.⁸  In another phase 2 trial, the at least PR rate was 40% in patients refractory to lenalidomide and 60% in those refractory to bortezomib.⁶  In our trial, the PCP regimen induced an at least PR rate of 51% and a clinical benefit, with an at least MR rate of 71%. These results are clinically meaningful in patients who had received most of the available therapies. Furthermore, patients were mostly refractory to lenalidomide (67%) or bortezomib (39%), and a subgroup was refractory to both these agents (29%). At least PR was reported in 46% of patients refractory to lenalidomide and in 50% of those refractory to bortezomib-lenalidomide. The addition of an alkylating agent to novel drugs has demonstrated an additive positive effect.^12,27,28  In our study, median PFS was 10.4 months and median OS was not reached, supporting the hypothesis that cyclophosphamide increased the clinical efficacy of pomalidomide.

The results obtained with PCP are impressive considering the dose and schedule used compared with the most recent trials.^9,10,12,29  In our study, the monthly dose of cyclophosphamide was 700 mg, which is inferior to the most commonly used doses. The monthly dose of pomalidomide (2.5 mg daily, 70 mg in a 28-day cycle) is slightly inferior to the standard schedule (4 mg/d for 21 days in a 28-day cycle, 84 mg monthly). This suggests that the positive results obtained with PCP are mainly related to the synergistic activity of the combination. Because PCP was well-tolerated, increasing dose intensity of pomalidomide to 4 mg for 21 days might further improve its efficacy without a major increase in toxicity.^9,10,12,29  We planned 6 cycles of PCP, but prolonging the treatment up to 9 cycles might also improve response rate and outcome. Future phase 3 trials should investigate higher doses of pomalidomide (4 mg/day) in a less-intensive 21-day schedule to minimize both the acute and the cumulative toxicity.

A dose-response relationship of pomalidomide is difficult to establish because the trial was not powered to evaluate it. Although the number of patients enrolled at different dose levels was quite limited, 2.5 mg/d pomalidomide induced better responses and a significant prolonged PFS (P = .05; data not shown) compared with lower doses.

The quality of response and the amount of cytoreduction seem to be predictive factors of longer remission duration. In our study, patients with at least PR showed a significantly prolonged PFS compared with patients achieving less than PR, suggesting a potential clinical benefit of a more intense cytoreduction in fit patients. Older age negatively affects PFS, demonstrating that dose adjustments are needed in vulnerable patients. Chromosomal abnormalities are the major prognostic factors for MM. A quarter of the patients in this study were classified as high risk at enrolment. PFS was not significantly different between standard- and high-risk patients, but the numbers are too limited and larger series are needed to confirm these preliminary findings.

The most common toxicities were hematologic, with grade 3 to 4 toxicities occurring in 25 patients (45%) at the MTD. Neutropenia was seen mainly in the first cycles, suggesting a concomitant role of the disease and the toxicity of the regimen. Hematologic toxicities were consistent with previous studies in which pomalidomide-dexamethasone induced a rate of grade 3 to 4 adverse events ranging from 38% to 53%.^6.8 Similarly, hematologic toxicity reported with PCP was comparable to the rates reported with lenalidomide-dexamethasone (52%).³⁰  These data suggest that cyclophosphamide at 50 mg every other day does not significantly increase hematologic toxicity.

The most frequent grade 3 to 4 nonhematologic toxicities were neurologic events, infections, and dermatologic reactions. The incidence of treatment-related peripheral neuropathy was low (4%), particularly when compared with bortezomib or thalidomide. In our study, grade 3 to 5 infections were noted in 9% of patients. In these patients, antibiotic prophylaxis may be recommended. A careful management of fever and neutropenia with the prompt institution of antibiotics is also suggested to reduce the incidence of infections. The incidence of grade 3 to 4 thromboembolic events was low (4%), supporting the need for anticoagulant prophylaxis.

This is the first study to establish that a novel combination, PCP, induces encouraging responses and outcomes in refractory MM. Data from this phase 1/2 study justify further exploration of this combination.

The authors thank the patients who took part in the study; the nurses, D’Ambrosio Luisella and Mastropasqua Gianmauro; the data managers, Antonella Fiorillo and Jessica Mastrovito; and the editorial assistant, Giorgio Schirripa.

The study was supported by Fondazione Neoplasie Sangue Onlus and Celgene.

Contribution: A.P., M.B., and P.C. designed the study, supervised its conduct and data analysis; A.L. and A.P. wrote the manuscript; A.L., V.M., S.B., D.R., C.C., R.M., M.G., M.M., G.L.V., N.G., V.M., T.G., D.R-S., P.O., A.S., A.M.C., and P.C. provided study material or recruited patients; I.B. performed statistical analysis; and all the authors had access to, commented on, and approved the final manuscript.

Conflict-of-interest disclosure: A.L. has received honoraria from Celgene and Janssen-Cilag. S.B. has received honoraria from Celgene, Janssen-Cilag, and Novartis and served on the advisory committee for Merck Sharp & Dohme. M.B. has received research support from and served on the scientific advisory board for Celgene and Janssen-Cilag. A.P. has received honoraria and consultancy fees from Celgene, Janssen-Cilag, Bristol-Myers Squibb, Millennium, Amgen, and Onyx. The remaining authors declare no competing financial interests.

Correspondence: Antonio Palumbo, Myeloma Unit, Division of Hematology, Azienda Ospedaliera Città della Salute e della Scienza di Torino, Via Genova 3, 10126, Torino, Italy; e-mail: appalumbo@yahoo.com.