VOLUME 27 ⅐ NUMBER 26 ⅐ SEPTEMBER 10 2009 JOURNAL OF CLINICAL ONCOLOGY ORIGINAL REPORT From the Department of Hematology/ Phase III Randomized Study of Bendamustine Compared Oncology, Onkologische Praxis, Frank- furt; Onkologische Gemeinschaftspraxis, With Chlorambucil in Previously Untreated Patients With Leipzig; Onkologische Schwerpunktpraxis, Minden; Department of Hematology, Chronic Lymphocytic Leukemia University Hospital, Jena; DSH Statistical Wolfgang U. Knauf, Toshko Lissichkov, Ali Aldaoud, Anna Liberati, Javier Loscertales, Raoul Herbrecht, Services, Rohrbach; Oncology Consult- Gunnar Juliusson, Gerhard Postner, Liana Gercheva, Stefan Goranov, Martin Becker, Hans-Joerg Fricke, ing, Miesbach, Germany; Hematology & Transfusion Medicine, National Hemato- Francoise Huguet, Ilaria Del Giudice, Peter Klein, Lothar Tremmel, Karlheinz Merkle, and Marco Montillo logical Center, Sofia; Department of Hematology, University Hospital, Varna; ABSTRACT Department of Hematology, University Hospital, Plovdiv, Bulgaria; Department of Purpose Oncology, Universita degli Studi, Perugia; This randomized, open-label, parallel-group, multicenter study was designed to compare the Ematologia, Ospedale Niguarda efficacy and safety of bendamustine and chlorambucil in previously untreated patients with Ca’Granda, Milano; Dip. Ematologia, advanced (Binet stage B or C) chronic lymphocytic leukemia (CLL). Universita “La Sapienza,” Roma, Italy; Patients and Methods Department of Hematology, Hopital de la 2 Princesa, Madrid, Spain; Hematology & Patients (Յ 75 years of age) were randomly assigned to receive bendamustine 100 mg/m /d Oncology, Hopital Universitaire Hautepi- intravenously on days 1 to 2, or chlorambucil 0.8 mg/kg (Broca’s normal weight) orally on days 1 erre, Strasbourg; Department of Hema- and 15; treatment cycles were repeated every 4 weeks for a maximum of six cycles. The response tology, Hopital Purpan, Toulouse, France; to treatment was assessed according to National Cancer Institute Working Group criteria, and the Department of Hematology, University final determination of response was made by a blinded independent review committee. Hospital, Lund, Sweden; Ludwig Boltz- mann Institute–Applied Cancer Research Results and Applied Cancer Research–Institute A total of 319 patients were randomly assigned (162 bendamustine, 157 chlorambucil). Complete for Translational Research VIEnna Kaiser or partial responses were achieved in 110 (68%) of 162 bendamustine-treated and 48 (31%) of 157 Franz Josef-Spital, Vienna, Austria; and chlorambucil-treated patients (P Ͻ .0001). More patients showed complete responses with the Cephalon Research Data Manage- bendamustine than with chlorambucil (31% v 2%). Median progression-free survival was 21.6 ment & Programming, Frazer, PA. months with bendamustine and 8.3 months with chlorambucil (P Ͻ .0001). Bendamustine was Submitted November 18, 2008; also associated with an improvement in duration of remission, compared with chlorambucil accepted May 6, 2009; published online (median, 21.8 v 8.0 months). Hematologic National Cancer Institute Common Toxicity Criteria ahead of print at www.jco.org on August 3, 2009. grade 3 to 4 adverse events were more common with bendamustine than with chlorambucil (occurring in 40% v 19% of patients). Severe infections (grade 3 to 4) occurred in 8% of Supported by grants from Ribosepharm bendamustine-treated patients and 3% of chlorambucil-treated patients. GmbH, Germany. and Mundipharma International, United Kingdom. Conclusion Presented in part at the Annual Meeting Bendamustine offers significantly greater efficacy than chlorambucil, and a manageable toxicity of the American Society of Hematology, profile, when used as first-line therapy in patients with advanced CLL. Atlanta, GA, December 6-9, 2008, and San Francisco, CA, December 8-11, 2007. J Clin Oncol 27:4378-4384. © 2009 by American Society of Clinical Oncology Authors’ disclosures of potential con- flicts of interest and author contribu- without affecting overall survival.4,5 However, there tions are found at the end of this INTRODUCTION article. remains a need for new treatment options in pa- Corresponding author: Wolfgang Knauf, Chronic lymphocytic leukemia (CLL) is the most tients with advanced CLL. MD, PhD, Onkologische Gemein- common form of adult leukemia in the Western Bendamustine is a novel agent, synthesized schaftspraxis, Frankfurter Diakonie world.1 Although patients with early-stage disease with the intent of combining the alkylating proper- Kliniken, Im Pruefling 17-19, 60389 Frankfurt, Germany; e-mail: have a life expectancy of longer than 10 years, those ties of mechlorethamine and the purine antimetab- 6,7 [email protected]. who progress or have advanced disease (Binet stage olite properties of benzimidazole. This agent, The Acknowledgment and Appendix BorCorRaistageIItoIV)haveamediansurvivalof alone or in combination with other chemotherapeu- are included in the full-text version approximately 2 to 7 years.2,3 First-line treatment is tic agents, has been shown to produce good clinical of this article; they are available efficacy and acceptable tolerability in patients with online at www.jco.org. They are frequently conducted with chlorambucil, fludara- 8,9 not included in the PDF version bine, or fludarabine plus cyclophosphamide, either non-Hodgkin’s lymphoma and multiple myelo- (via Adobe® Reader®). alone or in combination with rituximab. Fludara- ma.10 In phase I/II trials in patients with advanced © 2009 by American Society of Clinical bine has been reported to produce higher response relapsed or refractory CLL, bendamustine has Oncology rates, a longer duration of remission, and longer been shown to produce overall response rates 0732-183X/09/2726-4378/$20.00 progression-free survival than chlorambucil in pre- (ORR) similar to or higher than those achieved with 11-14 DOI: 10.1200/JCO.2008.20.8389 viously untreated younger patients with CLL, but chlorambucil. Therefore, a phase III trial was 4378 © 2009 by American Society of Clinical Oncology Downloaded from jco.ascopubs.org by JOSE CRUZ on October 8, 2009 from 70.238.206.169. Copyright © 2009 by the American Society of Clinical Oncology. All rights reserved. Bendamustine v Chlorambucil in CLL undertaken to compare the efficacy and tolerability of bendamustine investigator to the same maximum of six cycles. Patients with progressive with that of chlorambucil in previously untreated patients with CLL. disease were withdrawn. After the last treatment cycle, patients were moni- tored for response and survival at 3-month intervals. Final assessment of best response was performed in a blinded fashion by an Independent Committee PATIENTS AND METHODS for Response Assessment (ICRA) and classified as CR, PR, PR with nodular involvement, stable disease, or progressive disease based on the National Cancer Institute Working Group criteria.15 The study was a randomized, open-label, parallel-group, phase III trial con- Primary end points were the overall response rate and progression-free ducted at 45 centers in Austria, Bulgaria, France, Germany, Italy, Spain, Swe- survival. Secondary end points included time to progression, duration of den, and the United Kingdom. The protocol was approved by local ethics remission, and overall survival. Safety end points were infection rates and committees at all participating centers, and the study was conducted in accor- adverse events. dance with the International Conference on Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki. Statistical Methods and Sample Size Calculation The statistical analysis was performed on the intention-to-treat (ITT) Patients patient population. The safety population consisted of all patients who re- Previously untreated patients up to 75 years of age with Binet stage B (ie, ceived at least one dose of study medication. Ն 3lymphnoderegionsinvolvedincludinghepatomegalyandsplenomegaly) Statistical analysis of the primary end points was performed by means of or Binet stage C (ie, anemia and/or thrombocytopenia regardless of the num- an a priori–sequenced hypothesis testing and an adaptive group sequential test ber of lymph node regions) CLL confirmed by demonstration of coexpression 15,16 procedure. Overall remission rate was analyzed by means of Fisher’s exact test, of CD5, CD23, and either CD19, CD20, or both, and in need for treatment stratified by Binet stage; progression-free survival was analyzed by log-rank were included. All patients were required to have a WHO performance status test, stratified by Binet stage. All tests were two tailed with a multiple signifi- of 0 to 2 and a life expectancy of at least 3 months. Women of childbearing cance level of ␣ ϭ 5%. potential were required to use adequate contraception for at least 6 months Afive-stageadaptivegroupsequentialprocedurewithPocockcut-offsof after treatment. Patients with a second malignancy other than cured basal cell ␣i ϭ .016 was used, with a maximum of four planned interim analyses, of carcinoma or cured cervical cancer were excluded, as were patients with which three were performed (first analysis after treated 85 patients with a manifest immune hemolysis or thrombocytopenia that could be treated with follow-up of at least 5 months; second analysis after 158 patients; third analysis corticosteroids alone, and patients with Richter’s syndrome or transformation after 264 patients). In each interim analysis, ORR was tested first, while to prolymphocytic leukemia. Other exclusion criteria were hepatic