Randomized Phase III Study of Matrix Metalloproteinase Inhibitor BMS-275291 in Combination With Paclitaxel and Carboplatin in Advanced Non-S
http://www.100md.com
《临床肿瘤学》
the National Cancer Institute of Canada–Clinical Trials Group, Kingston
University Health Network/Princess Margaret Hospital, Toronto
The Margaret & Charles Juravinski Cancer Centre (formerly known as Hamilton Regional Cancer Centre), Hamilton, Ontario
Hospital Universitario Doce de Octubre, Madrid, Spain
Centre Rene Gauducheau, Nantes, France
Klinikum Rechts Der III Medizinische Klinik, Munich
Krankenhaus Grobhansdorf der LVA, Grobhansdorf, Germany
Hpital Jean Minjoz, Besanon, France
Istituto Clinico Humanitas, Rozzano, Italy
University of Bern, Inselspital, Institute of Medical Oncology, Bern, Switzerland
Medical University of Gdansk, Gdansk, Poland
Duke University Medical Center, Durham, NC
Bristol-Myers Squibb, Wallingford, CT
ABSTRACT
PURPOSE: To determine whether BMS-275291, a broad-spectrum matrix metalloproteinase inhibitor (MMPI), added to systemic chemotherapy improved survival in advanced non–small-cell lung cancer (NSCLC). In early phase studies, BMS- 275291 was not associated with dose-limiting joint toxicity seen with other MMPIs.
PATIENTS AND METHODS: Chemotherapy-naive patients with stage IIIB/IV NSCLC, performance status (PS) 0 to 2, and adequate organ function were eligible. All patients received paclitaxel 200 mg/m2 plus carboplatin (area under the curve, 6 mg/mL-min) intravenously every 21 days for up to 8 cycles, and were randomly assigned to receive BMS-275291, 1,200 mg orally daily, or placebo until disease progression. The primary study end point was survival (OS); secondary end points included progression-free survival (PFS), response rates (RR), toxicity, and quality of life.
RESULTS: From 2000 to 2002, 774 patients were randomly assigned. Pretreatment characteristics were well balanced between arms: median age, 61 years; male sex, 73%; stage IV, 79%; PS 0 to 1, 88%. Interim safety analysis revealed no survival advantage and increased toxicity in the experimental arm, and study treatment was stopped. Median OS, PFS and RR in the final analysis in the BMS-275291 arm were 8.6 months, 4.9 months, and 25.8% respectively, and in the control arm 9.2 months, 5.3 months, 33.7%. Toxicity was significantly higher in the BMS-275291 arm, including flu-like symptoms, rash, hypersensitivity reactions (8.6% v 2.4%), and febrile neutropenia (9.7% v 5.5%).
CONCLUSION: BMS-275291 added to chemotherapy increases toxicity and does not improve survival in advanced NSCLC.
INTRODUCTION
Lung cancer is the most common cause of cancer-related mortality in developed nations, with an estimated 1.2 million new cases diagnosed each year and 1 million deaths annually worldwide.1 The majority of patients with non–small-cell lung cancer (NSCLC) present with advanced disease, and first- and second-line chemotherapy yield modest improvements in survival and symptom control, with recent evidence to support similar benefits in the third-line setting in selected patients with epidermal growth factor tyrosine kinase inhibitors.2-4 However despite systemic therapy, the median survival in metastatic NSCLC remains less than 1 year. New strategies to improve the outcome of advanced NSCLC patients are required.
Matrix metalloproteinases (MMPs) belong to a family of enzymes that digest extracellular matrix and basement membrane components, and facilitate tumor growth, metastasis, and angiogenesis. Elevated expression of certain MMPs, including MMP-2 and -9, and tissue inhibitor of metalloproteinase (TIMP) -1 in tumor tissue or serum correlates with nodal metastasis, lymphovascular invasion, and poor prognosis in NSCLC.5-11 It was postulated, therefore that targeting these enzymes might slow tumor progression, and improve survival in NSCLC, particularly if MMP inhibitors (MMPIs) could be administered for prolonged periods during and after chemotherapy. Several synthetic inhibitors of MMPs have been developed and tested in advanced cancer, including five studies in advanced NSCLC and three in small-cell lung cancer (SCLC).12 To date, these studies have not demonstrated an improvement in outcome, but their results may have been limited by the inability to administer the MMPIs for prolonged periods in these trials. Marimastat, a broad-spectrum hydroxamate-based MMPI, has been studied as adjuvant treatment compared with placebo in patients with stage III NSCLC after combined-modality therapy, and as maintenance therapy in two large randomized trials of patients with SCLC who had responded to induction therapy.13 All three studies failed to demonstrate a survival difference, and up to 40% of patients stopped treatment with marimastat because of severe tendonitis and arthritis, a known dose-limiting toxicity of these agents. Prinomastat, a more selective nonpeptidomimetic MMPI, has been studied in advanced NSCLC in combination with platinum-based chemotherapy in previously untreated patients. In a study of prinomastat with paclitaxel and carboplatin, and another with gemcitabine and cisplatin, there was no improvement in survival seen with the addition of the MMPI.14,15 As in the marimastat studies, prolonged administration was problematic, with up to 37% of patients stopping prinomastat due to joint toxicity. BAY 12-9566, another nonpeptidomimetic MMPI that was not expected to cause joint toxicity, has been tested as maintenance therapy in both stage III NSCLC and SCLC after chemotherapy or chemoradiotherapy. These studies were closed when the interim analysis of the SCLC trial demonstrated inferior survival and unexpected thrombocytopenia in the BAY 12-9566 arm (P.A. Cyrus, personal communication, January 2001).
BMS-275291, a novel MMPI with a mercaptoacyl zinc-binding group, inhibits a broad range of MMPs in preclinical studies, including MMP-1, -2, -8, -9, -13, -14 and MMP-3 with lesser potency. However unlike hydroxamate-based MMPIs such as marimastat, BMS-275291 causes minimal inhibition of sheddase enzymes.16 Inhibition of sheddase activity is believed to mediate the musculoskeletal toxicity seen with other MMPIs.17 Thus it was postulated that BMS-275291 might be administered successfully as prolonged maintenance therapy without the joint toxicity seen with the older MMPIs.
In a phase I study, BMS-275291 was well tolerated for longer than 4 months at doses of 1,200 and 1,800 mg daily, and dose-limiting toxicities included hypersensitivity, (characterized by rash, fever, and dyspnea), as expected with the mercaptoacyl group, and elevations of serum ALT, but not musculoskeletal toxicity.18 Preclinically there was no interaction of BMS-275291 and chemotherapeutic agents including paclitaxel and carboplatin. It was anticipated that BMS-275291 could be administered safely with paclitaxel and carboplatin, and for a prolonged period following completion of chemotherapy without severe musculoskeletal toxicity. Given its expected tolerability, preclinical evidence of broad-spectrum inhibition of MMPs and the importance of MMP activity in NSCLC, the National Cancer Institute of Canada–Clinical Trials Group (NCIC CTG) initiated a randomized phase II/III study to determine the additional value of BMS-275291 in the treatment of advanced NSCLC. The phase II portion of the study showed that paclitaxel and carboplatin plus BMS-275291 1,200 mg orally daily resulted in similar response rates, acceptable toxicity profiles, and similar rates of arthrotoxicity compared to paclitaxel and carboplatin plus oral placebo.19 Therefore, the study was expanded to phase III, in which the primary goal was to determine whether adding BMS-275291 to platinum-based chemotherapy prolonged survival in patients with advanced NSCLC.
PATIENTS AND METHODS
All participating institutions received approval from their institutional ethics review boards to conduct the study.
Patients age 18 years and older with histologic or cytologic confirmation of stage IIIB or IV NSCLC were eligible, although those with stage IIIB tumors who were candidates for potentially curative combined modality chemotherapy and radical radiation were excluded. Patients had to have an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0,1 or 2; life expectancy of at least 12 weeks; and adequate hematologic, renal, and hepatic function. Patients had to provide written informed consent and be willing to complete quality of life (QoL) questionnaires in countries where validated versions of the questionnaire were available. Patients were ineligible if they had brain metastases; a diagnosis of breast, melanoma, or kidney cancer at any time; or a history of other malignancies within the preceding 5 years. They were excluded if they had received prior chemotherapy, immunotherapy, or other biologic response modifiers; or if they had peripheral neuropathy or other significant comorbidities that would interfere with the ability to deliver protocol treatment.
Within 14 days before random assignment, patients underwent a complete history, physical examination, and routine hematology and biochemistry, and within 28 days imaging including computed tomography of the chest, abdomen and other scans as necessary to document all known disease, as well as toxicity evaluation and quality-of-life assessment.
Eligible patients were randomly assigned centrally by telephone, and stratified by stage (IIIB v IV), PS (0 and 1 v 2), and by center using the minimization method.20 The NCIC CTG maintained the trial database.
Treatment
Paclitaxel 200 mg/m2 was administered intravenously over 3 hours plus carboplatin, area under the curve (AUC) of 6 mg/mL per minute on day 1 every 21 days, with standard hypersensitivity and antiemetic prophylaxis. In the absence of disease progression or unacceptable toxicity, patients were to receive a maximum of eight cycles of systemic chemotherapy. In addition, patients were randomly assigned to receive either oral BMS-275291 1,200 mg daily or a matched placebo concurrent with the start of chemotherapy. Study tablets could be continued until disease progression or intolerable toxicity.
For patients who experienced grade 3 or 4 toxicity considered related to BMS-275291 or placebo, including diarrhea, nausea, other gastrointestinal symptoms, rash or symptoms indicating a serum sickness reaction, treatment with the investigational agent was discontinued. For other grade 3 or 4 toxicity possibly related to BMS-275291 or placebo, or grade 2 toxicities, treatment could be withheld for a maximum of 14 days and restarted at 50% of the dose (600 mg daily). If symptoms did not resolve or if they returned, study treatment was discontinued. Chemotherapy doses were reduced if patients experienced grade 4 neutropenia for 7 days, febrile neutropenia, neutropenia with infection, platelet count < 25 x 109/L or thrombocytopenic bleeding, grade 3 or 4 nonhematologic toxicity, or grade 2 sensory neuropathy, renal dysfunction, hearing loss, and symptomatic cardiac dysrhythmias. Granulocyte colony-stimulating factor use was permitted according to institutional practice in place of a single dose reduction for febrile neutropenia.
During chemotherapy treatment, patients were assessed with history and physical examination, biochemistry, and toxicity evaluation at the end of each cycle (once every 3 weeks) with weekly hematology, and repeat imaging every 6 weeks. After chemotherapy treatment, during treatment with BMS-275291 or placebo alone, patients were assessed with history and physical examination, biochemistry, and toxicity evaluation every 6 weeks, with hematology every 3 weeks and repeat imaging every 12 weeks for 2 years, then once every 6 months until progression. QoL was assessed using the European Organisation for the Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ-C30) and lung cancer module (QLQ-LC13), at baseline and every 6 weeks during chemotherapy, then every 12 weeks during administration of oral BMS-275291 or placebo alone until progression. Treatment with the oral medication was double-blinded, and patients were unblinded only if knowledge of treatment assignment was essential for management of toxicity.
Statistical Considerations
The study was designed as a phase II/III multicenter, randomized, double-blind, placebo-controlled trial. Patients randomly assigned during the phase II portion of the study (n = 75) were included in the phase III analysis (planned prospectively) on an intent-to-treat basis in the primary analyses of time to progression and overall survival. Management of all patients was identical for both the phase II and phase III portions of the study, and patient treatment assignment was not unblinded for the phase II analysis.
The primary objective of the phase III study was to assess the additional effect of BMS-275291 to paclitaxel plus carboplatin by comparing overall survival between the BMS-275291 and placebo arms among all randomly assigned patients. Secondary objectives included assessment of progression-free survival, response rate, response duration, QoL and toxicity. Overall survival and progression-free survival were measured from the date of randomization, until death or progression respectively.
To have 90% power to detect a 2.8-month difference (25% reduction in risk of death) in the median survival between the two treatment arms with a two-sided of 5%, assuming a median survival time of 8.5 months in the control arm and taking into account an interim analysis, 581 events (deaths) were required to be observed before the final analysis. It was estimated that 750 patients accrued in 26 months would be required to reach the necessary number of events after 14 months of follow-up for all patients.
An interim analysis was planned when a minimum of 193 events (33% of deaths) were observed from all patients, including patients randomly assigned in the phase II portion of the study. The study would be terminated if sufficient statistical evidence was reached to conclude that survival in the BMS-275291 arm was superior to the placebo arm based on the one-sided 0.025-level O’Brien-Fleming stopping boundary, or if the survival in the BMS-275291 arm was equal to or worse than the placebo arm based on one-sided 0.025-level Pocock stopping boundary. The interim analysis was performed in March 2003 and presented to the NCIC CTG Data and Safety Monitoring Committee (DSMC). After reviewing the results of the interim analysis, DSMC recommended the study be unblinded and the results be given to the trial committee and investigators. This recommendation was accepted; it was also decided that a final analysis would be performed when the required number of events (581) in the protocol had been recorded. The final analysis was performed after the database was locked in December 2003.
All comparisons between treatment arms were carried out using two-sided tests with an level of 0.05. Discrete and categoric variables were summarized, and compared using Fisher’s exact test and the Wilcoxon test respectively. Time to progression and survival were compared using the Kaplan-Meier method, with primary comparisons made using the stratified log-rank test. Primary estimates of treatment differences were obtained with hazard ratios and 95% CIs from stratified Cox regression models. Subset analyses of outcome variables were planned to examine stage IIIB versus IV, PS 0 and 1 versus 2, age younger than 65 versus older, and male versus female sex.
RESULTS
A total of 774 patients were accrued between April 2000 and May 2002 from 63 participating centers in 14 countries. Seventeen patients (2.2%) did not meet entry criteria, including incomplete baseline investigations (n = 7), brain metastases (n = 1), inappropriate disease stage (n = 1), nonlung cancer histology (n = 1) and other reasons (n = 7). All 774 patients were included in the intention-to-treat efficacy analyses. Of the 774 randomly assigned patients, seven did not receive treatment (five in the BMS-275291 arm, two in the placebo arm). One patient randomly assigned to the BMS-275291 arm received oral placebo instead. The safety and treatment exposure analyses were conducted on 767 patients who received treatment based on the actual treatment arms to which they were assigned (Fig 1). Since QoL data collection began after a protocol amendment, only a subset of centers received institutional ethics approval in time to proceed with QoL data collection. Therefore, only 205 patients completed baseline, and at least one additional measurement on study and would be included in the QoL analyses. Hence, no detailed results of QoL analysis will be presented in this article.
Patient Characteristics
Baseline characteristics for the 774 patients are shown in Table 1, and were well balanced between the two treatment arms. Three quarters of patients were male, with a median age of 61 years; two thirds of patients were younger than 65 years. Slightly more than half of the patients had experienced minimal weight loss in the preceding 6 months, and 12% had a PS of 2. The majority had stage IV disease at study entry, and the median time from initial diagnosis to random assignment was approximately 1 month.
Treatment Received
Patients received a median of four cycles of chemotherapy in the experimental arm and five in the placebo arm. Only 31.9% of patients in the BMS-275291 arm continued extended oral therapy postchemotherapy, compared with 44% of patients in the placebo arm. Patients receiving active BMS-275291 also had a shorter duration of extended oral treatment postchemotherapy (Table 2). Patients were more likely to discontinue treatment for toxicity with chemotherapy, oral medication or both, in the BMS-275291 arm (33.3%), compared with patients receiving placebo (23.5%). Patients in the BMS-275291 arm had more dose reductions or interruptions of oral medication during chemotherapy treatment than the placebo arm (50.9% v 34.6%, respectively), although the total doses of chemotherapy received did not differ significantly between the two arms.
Efficacy
Baseline predictors of survival and progression-free survival using Cox multivariate regression analysis included treatment arm (BMS-275291 v placebo), weight loss during preceding 6 months (< 5% v other), hemoglobin (Common Toxicity Criteria grade 0 v other), sex (male v female), histology (squamous v other), age (< 65 years v 65 years). Significant predictors of overall survival were baseline hemoglobin and weight loss. Pretreatment hemoglobin also predicted progression-free survival (Table 3).
Overall response rates to chemotherapy were 25.8% in the experimental arm and 33.7% in the control arm (P = .10). Survival and progression-free survival were similar in the two arms, (Table 4, Figs 2 and 3), 8.6 months for BMS-275291 versus 9.2 months for placebo for overall survival (P = .30), and 4.9 months versus 5.3 months for progression-free survival (P = .91). Preplanned subset analyses of outcomes examining stage, performance status, sex and age did not show any differences in outcome between subsets.
Toxicity
All patients who received protocol treatment were included in the safety analysis (n = 767), and toxicity data are summarized in Table 5. Patients treated with BMS-275291 had a significantly higher incidence of grade 3 flu-like symptoms, hypersensitivity reactions, febrile neutropenia, dermatologic symptoms, and rash (Fig 3) or desquamation. Other common toxicities that were similar between the two treatment arms included fatigue, gastrointestinal symptoms, infections, neurologic toxicity, and musculoskeletal pains.
DISCUSSION
Despite promising preclinical activity, BMS-275291 does not appear to improve outcome when added to systemic chemotherapy in advanced NSCLC. Although the drug did not cause the severe musculoskeletal toxicity that has been seen with other MMPIs, it did increase the rate of hypersensitivity reactions, rash and febrile neutropenia, and patients in the experimental arm were more likely to stop therapy because of toxicity. Response rate and progression-free and overall survival were lower in the BMS-275291 arm compared with placebo, although the differences were not statistically significant. The inferior outcome in the BMS-275291 arm is likely related to the added toxicity of the MMPI plus chemotherapy, leading to earlier discontinuation of chemotherapy rather than a potentially negative interaction between the two.
BMS-275291 has also been studied in breast and prostate cancer. In a small phase II study, 72 patients with early stage breast cancer who had received adjuvant therapy were randomly assigned to BMS-275291 1,200 mg per day or matched placebo for 1 year.21 Higher rates of grade 2 musculoskeletal toxicity, including development of tendon nodules, were seen in the experimental arm (36.2% v 16.7% in the placebo arm; P = NS). Patients receiving BMS-275291 also were more likely to develop grade 3 rash, (8.5% v 4.2%; P = NS), and one third of patients receiving BMS-275291 had to stop therapy prematurely, compared with 21% of placebo patients. Although these differences were not statistically significant, the DSMB recommended that the trial be stopped given the musculoskeletal toxicity seen, and a larger randomized trial was deemed not feasible in the adjuvant breast cancer population. In metastatic hormone-refractory prostate cancer, a randomized phase II study of two doses of BMS-275291 was conducted, in which patients were randomly assigned to receive either 1,200 mg daily or 2,400 mg daily.22 There were no responses, and no significant differences in progression-free or overall survival. Higher doses of the MMPI were associated with less rapid bone turnover, with an impact on osteopontin levels, a marker of malignant bone tumors. Toxicities were slightly more common in the higher dose arm, with grade 3 or 4 toxicities of 15% and 24% in the low- and high-dose arms, respectively.
Despite no clear increase in musculoskeletal toxicity, patients in our trial were less likely to continue receiving BMS-275291 than placebo. Instead, they discontinued treatment for other toxicities, most commonly hypersensitivity symptoms. Even though 8% of patients continued on BMS-275291 for 6 months or more, there appeared to be no beneficial impact on outcome. There are several possible explanations. First, it may be that chemotherapy treatment may mask the impact of the MMPI, and as many patients progress or die within the first year of advanced NSCLC therapy, only a small subset of patients was able to progress to the maintenance phase of the trial. Second, MMPs may not be important in tumorigenesis and metastasis in all cancers, but rather in only a small subset. In the absence of predictive markers of response to MMPI therapy, it will be difficult to identify a population enriched with patients more likely to benefit from MMP inhibition. Although MMPIs have been shown to have antiangiogenic properties, it may be that most cancers rely more on other pathways of angiogenesis (eg, vascular endothelial growth factor receptor activation). Third, the complex interaction of MMPs and TIMPs is not fully understood. Thus the MMPIs may be upsetting a delicate balance, which, rather than harming the tumor selectively, will also disadvantage the host and surrounding nonmalignant tissue. It may be that greater suppression of activity of MMPs or suppression of other components of angiogenesis is required for more effective, targeted anticancer effect.
Despite promising preclinical data, it is clear that BMS-275291 does not add to the treatment of advanced NSCLC, and given the similar findings in other tumor sites and with other MMPIs, we anticipate that MMP inhibition will not be further pursued as an isolated anticancer strategy until there is a better understanding of the importance of MMP interactions and their inhibition in carcinogenesis. Ongoing studies of serum and tumor samples from patients on this study may provide further insight into how these agents act in cancer, and may identify a potential population that will benefit from these agents.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Jeffrey S. Humphrey, Pfizer Global R&D; Maurizio Voi, Bristol-Meyers Squibb; Susan Galbraith, Bristol-Meyers Squibb. Consultant/Advisory Role: Luis Paz-Ares, Bristol-Meyers Squibb; Jean-Yves Douillard, Bristol-Meyers Squibb. Stock Ownership: Maurizio Voi, Bristol-Meyers Squibb; Susan Galbraith, Bristol-Meyers Squibb. Honoraria: Jean-Yves Douillard, Bristol-Meyers Squibb; Frances A. Shepherd, Bristol-Meyers Squibb. Research Funding: Luis Paz-Ares, Bristol-Meyers Squibb; Jeffrey Crawford, Bristol-Meyers Squibb. For a detailed description of these categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We are indebted to the patients who participated in this study, the trial committee, investigators, pharmacists, clinical research associates, data safety and monitoring committee and central office staff of the National Cancer Institute of Canada Clinical Trials Group who contributed to the conduct of this trial. The following investigators also contributed to this study: Nathalie Caunes, MD, Toulouse, France; Piotr Koralewski, MD, Krakow, Poland; Maurizio Tonato, MD, Perugia, Italy; Etienne Lemarie, MD, Tours, France; Felipo De Marinis, MD, Rome, Italy; Karin Mattson, MD, Haartmaninkatu, Finland; Mary Davis, MD, Halifax, Nova Scotia; Tracy Dobbs, MD, Knoxville, TN; Giampietro Gasparini, MD, Rome, Italy; Eckhard Kaukel, MD, Hamburg, Germany; Alan Sandler, MD, Nashville, TN; Cezary Szczylik, MD, Warszawa, Poland; Jean-Franois Morere, MD, Bobigny, France; Francisco Robert, MD, Birmingham, AL; Rolf Stahel, MD, Zurich, Switzerland; Krystyna Folcik, MD, Mrozy, Poland; Haji Chalchal, MD, Regina, Canada; Andrew Maksymiuk, MD, Winnipeg, Canada; Hans-Nikol Macha, MD, Hemer, Germany; Larry Schlabach, MD, Chattanooga TN; Yee Ung, MD, Toronto, Canada; P. David Walde, MD, Sault Ste Marie, Canada; Peter Arrowsmith, MD, Chattanooga, TN; Leon Bosquee, MD, Liege, Belgium; Martin Wernli, MD, Aarau, Switzerland; Holger Baumgartner, MD, Vienna, Austria; Alan Cohen, MD, Nashville, TN; Howard Ozer, MD, Oklahoma City, OK; Wolfgang Schuette, MD, Halle, Germany; Marta Lopez Brea, MD, Santander, Spain; Violaine Frappat, MD, Grenoble, France; Michael Kosty, MD, La Jolla, CA; Elisabeth Quoix, MD, Strasbourg, France; David Stewart, MD, Ottawa, Canada; Zeenat Alam, MD, Windsor, Canada; Maria B. Parente, MD, Vila Nova de Gaia, Portugal; Todd Shuster, MD, Burlington, VT; Michael Smylie, MD, Edmonton, Canada; Mark Bower, MD, London, UK; Frederick M. Schnell, MD, Macon, GA; Luc Yves Dirix, MD, Wilrijk, Belgium; Pieter Joos, MD, Gent, Belgium; Anne-Marie Maddox, MD, Little Rock, AR; Amer Sami, MD, Saskatoon, Canada; Michael Thomas, MD, Muenster, Germany; Brian Findlay, MD, St. Catharines, Canada; Dwight Kaufman, MD, Jackson TN; Bryn Pressnail, MD, Barrie, Canada; Brian Murphy, MD, Franklin TN; Kendrith Rowland Jr., MD, Urbana, IL; Guenther Steger, Vienna, Austria; G.P.M. ten Velde, MD, Maastricht, Netherlands.
NOTES
Supported in part by a grant from Bristol-Myers Squibb, Wallingford, CT.
Presented in part at the 40th Annual Meeting of the American Society of Clinical Oncology, June 5-8, 2004, New Orleans, LA.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Parkin DM, Bray F, Ferlay J, et al: Estimating the world cancer burden: Globocan 2000. Int J Cancer 94:153-156, 2001
Non-Small Cell Lung Cancer Collaborative Group: Chemotherapy for non-small cell lung cancer [Cochrane Database System Review]. Oxford, England, Cochrane Library, CD002139, issue 2, 2000
Shepherd FA, Dancey J, Ramlau R, et al: Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095-2103, 2000
Shepherd FA, Pereira J, Ciuleanu T, et al: A randomized placebo-controlled study of erlotinib (OSI-774, Tarceva) versus placebo in patients with incurable non-small cell lung cancer who have failed standard therapy for advanced or metastatic disease. Proc Am Soc Clin Oncol 22:622S, 2004 (abstr 7022)
Janecki-Delebecq T, Porte H, Zerimech F, et al: Overexpression level of stromelysin 3 is related to the lymph node involvement in non-small cell lung cancer. Clin Cancer Res 6:1086-1092, 2000
Passlick B, Srenel W, Seen-Hibler R, et al: Overexpression of matrix metalloproteinases predicts unfavorable outcome in early-stage non-small cell lung cancer. Clin Cancer Res 6:3944-3948, 2000
Brown PD, Bloxidge RE, Stuart NS, et al: Association between expression of activated 72-kilodalton gelatinase and tumor spread in non-small-cell lung carcinoma. J Natl Cancer Inst 85:574-578, 1993
Garbisa S, Scagliotti G, Masiero L, et al: Correlation of serum metalloproteinase levels with lung cancer metastasis and response to therapy. Cancer Res 52:4548-4549, 1992
Thomas P, Khokha R, Shepherd FA, et al: Differential expression of matrix metalloproteinases and their inhibitors in non-small cell lung cancer. J Pathol 190:150-156, 2000
Simi L, Andreani M, Davini F, et al: Simultaneous measurement of MMP9 and TIMP1 mRNA in human non small cell lung cancers by multiplex real time RT-PCR. Lung Cancer 45:171-179, 2004
Aljada IS, Ramnath N, Donohue K, et al: Upregulation of the Tissue Inhibitor of Metalloproteinase-1 Protein is associated with progression of human non-small cell lung cancer. J Clin Oncol 22:3218-3229, 2004
Shepherd FA, Sridhar SS: Angiogenesis inhibitors under study for the treatment of lung cancer. Lung Cancer 41:S63-S72, 2003
Shepherd FA, Giaccone G, Seymour L, et al: Prospective, randomized, double-blind, placebo-controlled trial of marimastat after response to first-line chemotherapy in patients with small-cell lung cancer: A trial of the National Cancer Institute of Canada–Clinical Trials Group and the European Organization for Research and Treatment of Cancer. J Clin Oncol 20:4434-4439, 2002
Bissett D, O’Byrne K, von Pawel J, et al: Phase III study of the matrix metalloprotease (MMP) inhibitor prinomastat in combination with gemcitabine and cisplatin in non-small cell lung cancer. Proc Am Soc Clin Oncol 21:296a, 2002 (abstr 1183)
Douillard JY, Peschel C, Shepherd F, et al: Randomized phase II feasibility study of combining the matrix metalloproteinase inhibitor BMS-275291 with paclitaxel plus carboplatin in advanced non-small cell lung cancer. Lung Cancer 46:361-368, 2004
Bristol-Myers Squibb Pharmaceutical Research Institute: Investigator Brochure—Matrix metalloproteinase inhibitor BMS-275291, version 4. Bristol-Myers Squibb, Wallingford, CT, 2002
Lombard MA, Wallace TL, Kubicek MF, et al: Synthetic matrix metalloproteinase inhibitors and tissue inhibitor of metalloproteinase (TIMP)-2, but not TIMP-1, inhibit shedding of tumor necrosis factor-alpha receptors in a human colon adenocarcinoma (Colo 205) cell line. Cancer Res 58:4001-4007, 1998
Rizvi NA, Humphrey JS, Ness EA, et al: A phase I study of oral BMS-275291, a novel nonhydroxamate sheddase-sparing matrix metalloproteinase inhibitor, in patients with advanced or metastatic cancer. Clin Cancer Res 10:1963-1970, 2004
Douillard JY, Peschel C, Shepherd F, et al: Randomized phase II study of combining the matrix metalloproteinase inhibitor BMS-275291 with paclitaxel plus carboplatin in advanced non-small cell lung cancer. Lung Cancer 46:361-368, 2004
Tu D: Minimization Procedure, in Chow SC (ed) Encyclopaedia of Biopharmaceutical Statistics. New York: Marcel Dekker, Inc, 2003, pp 614-618
Miller KD, Saphner TJ, Waerhouse DM, et al: A randomized phase II feasibility trial of BMS-275291 in patients with early stage breast cancer. Clin Cancer Res 10:1971-1975, 2004
Lara PN, Longmate J, Stadler W, et al: Angiogenesis inhibition in metastatic hormone refractory prostate cancer (HRPC): A randomized phase II trial of two doses of the matrix metalloproteinase inhibitor (MMPI) BMS-275291. Proc Am Soc Clin Oncol 23:417, 2004 (abstr 4647)(Natasha B. Leighl, Luis P)
University Health Network/Princess Margaret Hospital, Toronto
The Margaret & Charles Juravinski Cancer Centre (formerly known as Hamilton Regional Cancer Centre), Hamilton, Ontario
Hospital Universitario Doce de Octubre, Madrid, Spain
Centre Rene Gauducheau, Nantes, France
Klinikum Rechts Der III Medizinische Klinik, Munich
Krankenhaus Grobhansdorf der LVA, Grobhansdorf, Germany
Hpital Jean Minjoz, Besanon, France
Istituto Clinico Humanitas, Rozzano, Italy
University of Bern, Inselspital, Institute of Medical Oncology, Bern, Switzerland
Medical University of Gdansk, Gdansk, Poland
Duke University Medical Center, Durham, NC
Bristol-Myers Squibb, Wallingford, CT
ABSTRACT
PURPOSE: To determine whether BMS-275291, a broad-spectrum matrix metalloproteinase inhibitor (MMPI), added to systemic chemotherapy improved survival in advanced non–small-cell lung cancer (NSCLC). In early phase studies, BMS- 275291 was not associated with dose-limiting joint toxicity seen with other MMPIs.
PATIENTS AND METHODS: Chemotherapy-naive patients with stage IIIB/IV NSCLC, performance status (PS) 0 to 2, and adequate organ function were eligible. All patients received paclitaxel 200 mg/m2 plus carboplatin (area under the curve, 6 mg/mL-min) intravenously every 21 days for up to 8 cycles, and were randomly assigned to receive BMS-275291, 1,200 mg orally daily, or placebo until disease progression. The primary study end point was survival (OS); secondary end points included progression-free survival (PFS), response rates (RR), toxicity, and quality of life.
RESULTS: From 2000 to 2002, 774 patients were randomly assigned. Pretreatment characteristics were well balanced between arms: median age, 61 years; male sex, 73%; stage IV, 79%; PS 0 to 1, 88%. Interim safety analysis revealed no survival advantage and increased toxicity in the experimental arm, and study treatment was stopped. Median OS, PFS and RR in the final analysis in the BMS-275291 arm were 8.6 months, 4.9 months, and 25.8% respectively, and in the control arm 9.2 months, 5.3 months, 33.7%. Toxicity was significantly higher in the BMS-275291 arm, including flu-like symptoms, rash, hypersensitivity reactions (8.6% v 2.4%), and febrile neutropenia (9.7% v 5.5%).
CONCLUSION: BMS-275291 added to chemotherapy increases toxicity and does not improve survival in advanced NSCLC.
INTRODUCTION
Lung cancer is the most common cause of cancer-related mortality in developed nations, with an estimated 1.2 million new cases diagnosed each year and 1 million deaths annually worldwide.1 The majority of patients with non–small-cell lung cancer (NSCLC) present with advanced disease, and first- and second-line chemotherapy yield modest improvements in survival and symptom control, with recent evidence to support similar benefits in the third-line setting in selected patients with epidermal growth factor tyrosine kinase inhibitors.2-4 However despite systemic therapy, the median survival in metastatic NSCLC remains less than 1 year. New strategies to improve the outcome of advanced NSCLC patients are required.
Matrix metalloproteinases (MMPs) belong to a family of enzymes that digest extracellular matrix and basement membrane components, and facilitate tumor growth, metastasis, and angiogenesis. Elevated expression of certain MMPs, including MMP-2 and -9, and tissue inhibitor of metalloproteinase (TIMP) -1 in tumor tissue or serum correlates with nodal metastasis, lymphovascular invasion, and poor prognosis in NSCLC.5-11 It was postulated, therefore that targeting these enzymes might slow tumor progression, and improve survival in NSCLC, particularly if MMP inhibitors (MMPIs) could be administered for prolonged periods during and after chemotherapy. Several synthetic inhibitors of MMPs have been developed and tested in advanced cancer, including five studies in advanced NSCLC and three in small-cell lung cancer (SCLC).12 To date, these studies have not demonstrated an improvement in outcome, but their results may have been limited by the inability to administer the MMPIs for prolonged periods in these trials. Marimastat, a broad-spectrum hydroxamate-based MMPI, has been studied as adjuvant treatment compared with placebo in patients with stage III NSCLC after combined-modality therapy, and as maintenance therapy in two large randomized trials of patients with SCLC who had responded to induction therapy.13 All three studies failed to demonstrate a survival difference, and up to 40% of patients stopped treatment with marimastat because of severe tendonitis and arthritis, a known dose-limiting toxicity of these agents. Prinomastat, a more selective nonpeptidomimetic MMPI, has been studied in advanced NSCLC in combination with platinum-based chemotherapy in previously untreated patients. In a study of prinomastat with paclitaxel and carboplatin, and another with gemcitabine and cisplatin, there was no improvement in survival seen with the addition of the MMPI.14,15 As in the marimastat studies, prolonged administration was problematic, with up to 37% of patients stopping prinomastat due to joint toxicity. BAY 12-9566, another nonpeptidomimetic MMPI that was not expected to cause joint toxicity, has been tested as maintenance therapy in both stage III NSCLC and SCLC after chemotherapy or chemoradiotherapy. These studies were closed when the interim analysis of the SCLC trial demonstrated inferior survival and unexpected thrombocytopenia in the BAY 12-9566 arm (P.A. Cyrus, personal communication, January 2001).
BMS-275291, a novel MMPI with a mercaptoacyl zinc-binding group, inhibits a broad range of MMPs in preclinical studies, including MMP-1, -2, -8, -9, -13, -14 and MMP-3 with lesser potency. However unlike hydroxamate-based MMPIs such as marimastat, BMS-275291 causes minimal inhibition of sheddase enzymes.16 Inhibition of sheddase activity is believed to mediate the musculoskeletal toxicity seen with other MMPIs.17 Thus it was postulated that BMS-275291 might be administered successfully as prolonged maintenance therapy without the joint toxicity seen with the older MMPIs.
In a phase I study, BMS-275291 was well tolerated for longer than 4 months at doses of 1,200 and 1,800 mg daily, and dose-limiting toxicities included hypersensitivity, (characterized by rash, fever, and dyspnea), as expected with the mercaptoacyl group, and elevations of serum ALT, but not musculoskeletal toxicity.18 Preclinically there was no interaction of BMS-275291 and chemotherapeutic agents including paclitaxel and carboplatin. It was anticipated that BMS-275291 could be administered safely with paclitaxel and carboplatin, and for a prolonged period following completion of chemotherapy without severe musculoskeletal toxicity. Given its expected tolerability, preclinical evidence of broad-spectrum inhibition of MMPs and the importance of MMP activity in NSCLC, the National Cancer Institute of Canada–Clinical Trials Group (NCIC CTG) initiated a randomized phase II/III study to determine the additional value of BMS-275291 in the treatment of advanced NSCLC. The phase II portion of the study showed that paclitaxel and carboplatin plus BMS-275291 1,200 mg orally daily resulted in similar response rates, acceptable toxicity profiles, and similar rates of arthrotoxicity compared to paclitaxel and carboplatin plus oral placebo.19 Therefore, the study was expanded to phase III, in which the primary goal was to determine whether adding BMS-275291 to platinum-based chemotherapy prolonged survival in patients with advanced NSCLC.
PATIENTS AND METHODS
All participating institutions received approval from their institutional ethics review boards to conduct the study.
Patients age 18 years and older with histologic or cytologic confirmation of stage IIIB or IV NSCLC were eligible, although those with stage IIIB tumors who were candidates for potentially curative combined modality chemotherapy and radical radiation were excluded. Patients had to have an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0,1 or 2; life expectancy of at least 12 weeks; and adequate hematologic, renal, and hepatic function. Patients had to provide written informed consent and be willing to complete quality of life (QoL) questionnaires in countries where validated versions of the questionnaire were available. Patients were ineligible if they had brain metastases; a diagnosis of breast, melanoma, or kidney cancer at any time; or a history of other malignancies within the preceding 5 years. They were excluded if they had received prior chemotherapy, immunotherapy, or other biologic response modifiers; or if they had peripheral neuropathy or other significant comorbidities that would interfere with the ability to deliver protocol treatment.
Within 14 days before random assignment, patients underwent a complete history, physical examination, and routine hematology and biochemistry, and within 28 days imaging including computed tomography of the chest, abdomen and other scans as necessary to document all known disease, as well as toxicity evaluation and quality-of-life assessment.
Eligible patients were randomly assigned centrally by telephone, and stratified by stage (IIIB v IV), PS (0 and 1 v 2), and by center using the minimization method.20 The NCIC CTG maintained the trial database.
Treatment
Paclitaxel 200 mg/m2 was administered intravenously over 3 hours plus carboplatin, area under the curve (AUC) of 6 mg/mL per minute on day 1 every 21 days, with standard hypersensitivity and antiemetic prophylaxis. In the absence of disease progression or unacceptable toxicity, patients were to receive a maximum of eight cycles of systemic chemotherapy. In addition, patients were randomly assigned to receive either oral BMS-275291 1,200 mg daily or a matched placebo concurrent with the start of chemotherapy. Study tablets could be continued until disease progression or intolerable toxicity.
For patients who experienced grade 3 or 4 toxicity considered related to BMS-275291 or placebo, including diarrhea, nausea, other gastrointestinal symptoms, rash or symptoms indicating a serum sickness reaction, treatment with the investigational agent was discontinued. For other grade 3 or 4 toxicity possibly related to BMS-275291 or placebo, or grade 2 toxicities, treatment could be withheld for a maximum of 14 days and restarted at 50% of the dose (600 mg daily). If symptoms did not resolve or if they returned, study treatment was discontinued. Chemotherapy doses were reduced if patients experienced grade 4 neutropenia for 7 days, febrile neutropenia, neutropenia with infection, platelet count < 25 x 109/L or thrombocytopenic bleeding, grade 3 or 4 nonhematologic toxicity, or grade 2 sensory neuropathy, renal dysfunction, hearing loss, and symptomatic cardiac dysrhythmias. Granulocyte colony-stimulating factor use was permitted according to institutional practice in place of a single dose reduction for febrile neutropenia.
During chemotherapy treatment, patients were assessed with history and physical examination, biochemistry, and toxicity evaluation at the end of each cycle (once every 3 weeks) with weekly hematology, and repeat imaging every 6 weeks. After chemotherapy treatment, during treatment with BMS-275291 or placebo alone, patients were assessed with history and physical examination, biochemistry, and toxicity evaluation every 6 weeks, with hematology every 3 weeks and repeat imaging every 12 weeks for 2 years, then once every 6 months until progression. QoL was assessed using the European Organisation for the Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ-C30) and lung cancer module (QLQ-LC13), at baseline and every 6 weeks during chemotherapy, then every 12 weeks during administration of oral BMS-275291 or placebo alone until progression. Treatment with the oral medication was double-blinded, and patients were unblinded only if knowledge of treatment assignment was essential for management of toxicity.
Statistical Considerations
The study was designed as a phase II/III multicenter, randomized, double-blind, placebo-controlled trial. Patients randomly assigned during the phase II portion of the study (n = 75) were included in the phase III analysis (planned prospectively) on an intent-to-treat basis in the primary analyses of time to progression and overall survival. Management of all patients was identical for both the phase II and phase III portions of the study, and patient treatment assignment was not unblinded for the phase II analysis.
The primary objective of the phase III study was to assess the additional effect of BMS-275291 to paclitaxel plus carboplatin by comparing overall survival between the BMS-275291 and placebo arms among all randomly assigned patients. Secondary objectives included assessment of progression-free survival, response rate, response duration, QoL and toxicity. Overall survival and progression-free survival were measured from the date of randomization, until death or progression respectively.
To have 90% power to detect a 2.8-month difference (25% reduction in risk of death) in the median survival between the two treatment arms with a two-sided of 5%, assuming a median survival time of 8.5 months in the control arm and taking into account an interim analysis, 581 events (deaths) were required to be observed before the final analysis. It was estimated that 750 patients accrued in 26 months would be required to reach the necessary number of events after 14 months of follow-up for all patients.
An interim analysis was planned when a minimum of 193 events (33% of deaths) were observed from all patients, including patients randomly assigned in the phase II portion of the study. The study would be terminated if sufficient statistical evidence was reached to conclude that survival in the BMS-275291 arm was superior to the placebo arm based on the one-sided 0.025-level O’Brien-Fleming stopping boundary, or if the survival in the BMS-275291 arm was equal to or worse than the placebo arm based on one-sided 0.025-level Pocock stopping boundary. The interim analysis was performed in March 2003 and presented to the NCIC CTG Data and Safety Monitoring Committee (DSMC). After reviewing the results of the interim analysis, DSMC recommended the study be unblinded and the results be given to the trial committee and investigators. This recommendation was accepted; it was also decided that a final analysis would be performed when the required number of events (581) in the protocol had been recorded. The final analysis was performed after the database was locked in December 2003.
All comparisons between treatment arms were carried out using two-sided tests with an level of 0.05. Discrete and categoric variables were summarized, and compared using Fisher’s exact test and the Wilcoxon test respectively. Time to progression and survival were compared using the Kaplan-Meier method, with primary comparisons made using the stratified log-rank test. Primary estimates of treatment differences were obtained with hazard ratios and 95% CIs from stratified Cox regression models. Subset analyses of outcome variables were planned to examine stage IIIB versus IV, PS 0 and 1 versus 2, age younger than 65 versus older, and male versus female sex.
RESULTS
A total of 774 patients were accrued between April 2000 and May 2002 from 63 participating centers in 14 countries. Seventeen patients (2.2%) did not meet entry criteria, including incomplete baseline investigations (n = 7), brain metastases (n = 1), inappropriate disease stage (n = 1), nonlung cancer histology (n = 1) and other reasons (n = 7). All 774 patients were included in the intention-to-treat efficacy analyses. Of the 774 randomly assigned patients, seven did not receive treatment (five in the BMS-275291 arm, two in the placebo arm). One patient randomly assigned to the BMS-275291 arm received oral placebo instead. The safety and treatment exposure analyses were conducted on 767 patients who received treatment based on the actual treatment arms to which they were assigned (Fig 1). Since QoL data collection began after a protocol amendment, only a subset of centers received institutional ethics approval in time to proceed with QoL data collection. Therefore, only 205 patients completed baseline, and at least one additional measurement on study and would be included in the QoL analyses. Hence, no detailed results of QoL analysis will be presented in this article.
Patient Characteristics
Baseline characteristics for the 774 patients are shown in Table 1, and were well balanced between the two treatment arms. Three quarters of patients were male, with a median age of 61 years; two thirds of patients were younger than 65 years. Slightly more than half of the patients had experienced minimal weight loss in the preceding 6 months, and 12% had a PS of 2. The majority had stage IV disease at study entry, and the median time from initial diagnosis to random assignment was approximately 1 month.
Treatment Received
Patients received a median of four cycles of chemotherapy in the experimental arm and five in the placebo arm. Only 31.9% of patients in the BMS-275291 arm continued extended oral therapy postchemotherapy, compared with 44% of patients in the placebo arm. Patients receiving active BMS-275291 also had a shorter duration of extended oral treatment postchemotherapy (Table 2). Patients were more likely to discontinue treatment for toxicity with chemotherapy, oral medication or both, in the BMS-275291 arm (33.3%), compared with patients receiving placebo (23.5%). Patients in the BMS-275291 arm had more dose reductions or interruptions of oral medication during chemotherapy treatment than the placebo arm (50.9% v 34.6%, respectively), although the total doses of chemotherapy received did not differ significantly between the two arms.
Efficacy
Baseline predictors of survival and progression-free survival using Cox multivariate regression analysis included treatment arm (BMS-275291 v placebo), weight loss during preceding 6 months (< 5% v other), hemoglobin (Common Toxicity Criteria grade 0 v other), sex (male v female), histology (squamous v other), age (< 65 years v 65 years). Significant predictors of overall survival were baseline hemoglobin and weight loss. Pretreatment hemoglobin also predicted progression-free survival (Table 3).
Overall response rates to chemotherapy were 25.8% in the experimental arm and 33.7% in the control arm (P = .10). Survival and progression-free survival were similar in the two arms, (Table 4, Figs 2 and 3), 8.6 months for BMS-275291 versus 9.2 months for placebo for overall survival (P = .30), and 4.9 months versus 5.3 months for progression-free survival (P = .91). Preplanned subset analyses of outcomes examining stage, performance status, sex and age did not show any differences in outcome between subsets.
Toxicity
All patients who received protocol treatment were included in the safety analysis (n = 767), and toxicity data are summarized in Table 5. Patients treated with BMS-275291 had a significantly higher incidence of grade 3 flu-like symptoms, hypersensitivity reactions, febrile neutropenia, dermatologic symptoms, and rash (Fig 3) or desquamation. Other common toxicities that were similar between the two treatment arms included fatigue, gastrointestinal symptoms, infections, neurologic toxicity, and musculoskeletal pains.
DISCUSSION
Despite promising preclinical activity, BMS-275291 does not appear to improve outcome when added to systemic chemotherapy in advanced NSCLC. Although the drug did not cause the severe musculoskeletal toxicity that has been seen with other MMPIs, it did increase the rate of hypersensitivity reactions, rash and febrile neutropenia, and patients in the experimental arm were more likely to stop therapy because of toxicity. Response rate and progression-free and overall survival were lower in the BMS-275291 arm compared with placebo, although the differences were not statistically significant. The inferior outcome in the BMS-275291 arm is likely related to the added toxicity of the MMPI plus chemotherapy, leading to earlier discontinuation of chemotherapy rather than a potentially negative interaction between the two.
BMS-275291 has also been studied in breast and prostate cancer. In a small phase II study, 72 patients with early stage breast cancer who had received adjuvant therapy were randomly assigned to BMS-275291 1,200 mg per day or matched placebo for 1 year.21 Higher rates of grade 2 musculoskeletal toxicity, including development of tendon nodules, were seen in the experimental arm (36.2% v 16.7% in the placebo arm; P = NS). Patients receiving BMS-275291 also were more likely to develop grade 3 rash, (8.5% v 4.2%; P = NS), and one third of patients receiving BMS-275291 had to stop therapy prematurely, compared with 21% of placebo patients. Although these differences were not statistically significant, the DSMB recommended that the trial be stopped given the musculoskeletal toxicity seen, and a larger randomized trial was deemed not feasible in the adjuvant breast cancer population. In metastatic hormone-refractory prostate cancer, a randomized phase II study of two doses of BMS-275291 was conducted, in which patients were randomly assigned to receive either 1,200 mg daily or 2,400 mg daily.22 There were no responses, and no significant differences in progression-free or overall survival. Higher doses of the MMPI were associated with less rapid bone turnover, with an impact on osteopontin levels, a marker of malignant bone tumors. Toxicities were slightly more common in the higher dose arm, with grade 3 or 4 toxicities of 15% and 24% in the low- and high-dose arms, respectively.
Despite no clear increase in musculoskeletal toxicity, patients in our trial were less likely to continue receiving BMS-275291 than placebo. Instead, they discontinued treatment for other toxicities, most commonly hypersensitivity symptoms. Even though 8% of patients continued on BMS-275291 for 6 months or more, there appeared to be no beneficial impact on outcome. There are several possible explanations. First, it may be that chemotherapy treatment may mask the impact of the MMPI, and as many patients progress or die within the first year of advanced NSCLC therapy, only a small subset of patients was able to progress to the maintenance phase of the trial. Second, MMPs may not be important in tumorigenesis and metastasis in all cancers, but rather in only a small subset. In the absence of predictive markers of response to MMPI therapy, it will be difficult to identify a population enriched with patients more likely to benefit from MMP inhibition. Although MMPIs have been shown to have antiangiogenic properties, it may be that most cancers rely more on other pathways of angiogenesis (eg, vascular endothelial growth factor receptor activation). Third, the complex interaction of MMPs and TIMPs is not fully understood. Thus the MMPIs may be upsetting a delicate balance, which, rather than harming the tumor selectively, will also disadvantage the host and surrounding nonmalignant tissue. It may be that greater suppression of activity of MMPs or suppression of other components of angiogenesis is required for more effective, targeted anticancer effect.
Despite promising preclinical data, it is clear that BMS-275291 does not add to the treatment of advanced NSCLC, and given the similar findings in other tumor sites and with other MMPIs, we anticipate that MMP inhibition will not be further pursued as an isolated anticancer strategy until there is a better understanding of the importance of MMP interactions and their inhibition in carcinogenesis. Ongoing studies of serum and tumor samples from patients on this study may provide further insight into how these agents act in cancer, and may identify a potential population that will benefit from these agents.
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Jeffrey S. Humphrey, Pfizer Global R&D; Maurizio Voi, Bristol-Meyers Squibb; Susan Galbraith, Bristol-Meyers Squibb. Consultant/Advisory Role: Luis Paz-Ares, Bristol-Meyers Squibb; Jean-Yves Douillard, Bristol-Meyers Squibb. Stock Ownership: Maurizio Voi, Bristol-Meyers Squibb; Susan Galbraith, Bristol-Meyers Squibb. Honoraria: Jean-Yves Douillard, Bristol-Meyers Squibb; Frances A. Shepherd, Bristol-Meyers Squibb. Research Funding: Luis Paz-Ares, Bristol-Meyers Squibb; Jeffrey Crawford, Bristol-Meyers Squibb. For a detailed description of these categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We are indebted to the patients who participated in this study, the trial committee, investigators, pharmacists, clinical research associates, data safety and monitoring committee and central office staff of the National Cancer Institute of Canada Clinical Trials Group who contributed to the conduct of this trial. The following investigators also contributed to this study: Nathalie Caunes, MD, Toulouse, France; Piotr Koralewski, MD, Krakow, Poland; Maurizio Tonato, MD, Perugia, Italy; Etienne Lemarie, MD, Tours, France; Felipo De Marinis, MD, Rome, Italy; Karin Mattson, MD, Haartmaninkatu, Finland; Mary Davis, MD, Halifax, Nova Scotia; Tracy Dobbs, MD, Knoxville, TN; Giampietro Gasparini, MD, Rome, Italy; Eckhard Kaukel, MD, Hamburg, Germany; Alan Sandler, MD, Nashville, TN; Cezary Szczylik, MD, Warszawa, Poland; Jean-Franois Morere, MD, Bobigny, France; Francisco Robert, MD, Birmingham, AL; Rolf Stahel, MD, Zurich, Switzerland; Krystyna Folcik, MD, Mrozy, Poland; Haji Chalchal, MD, Regina, Canada; Andrew Maksymiuk, MD, Winnipeg, Canada; Hans-Nikol Macha, MD, Hemer, Germany; Larry Schlabach, MD, Chattanooga TN; Yee Ung, MD, Toronto, Canada; P. David Walde, MD, Sault Ste Marie, Canada; Peter Arrowsmith, MD, Chattanooga, TN; Leon Bosquee, MD, Liege, Belgium; Martin Wernli, MD, Aarau, Switzerland; Holger Baumgartner, MD, Vienna, Austria; Alan Cohen, MD, Nashville, TN; Howard Ozer, MD, Oklahoma City, OK; Wolfgang Schuette, MD, Halle, Germany; Marta Lopez Brea, MD, Santander, Spain; Violaine Frappat, MD, Grenoble, France; Michael Kosty, MD, La Jolla, CA; Elisabeth Quoix, MD, Strasbourg, France; David Stewart, MD, Ottawa, Canada; Zeenat Alam, MD, Windsor, Canada; Maria B. Parente, MD, Vila Nova de Gaia, Portugal; Todd Shuster, MD, Burlington, VT; Michael Smylie, MD, Edmonton, Canada; Mark Bower, MD, London, UK; Frederick M. Schnell, MD, Macon, GA; Luc Yves Dirix, MD, Wilrijk, Belgium; Pieter Joos, MD, Gent, Belgium; Anne-Marie Maddox, MD, Little Rock, AR; Amer Sami, MD, Saskatoon, Canada; Michael Thomas, MD, Muenster, Germany; Brian Findlay, MD, St. Catharines, Canada; Dwight Kaufman, MD, Jackson TN; Bryn Pressnail, MD, Barrie, Canada; Brian Murphy, MD, Franklin TN; Kendrith Rowland Jr., MD, Urbana, IL; Guenther Steger, Vienna, Austria; G.P.M. ten Velde, MD, Maastricht, Netherlands.
NOTES
Supported in part by a grant from Bristol-Myers Squibb, Wallingford, CT.
Presented in part at the 40th Annual Meeting of the American Society of Clinical Oncology, June 5-8, 2004, New Orleans, LA.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Parkin DM, Bray F, Ferlay J, et al: Estimating the world cancer burden: Globocan 2000. Int J Cancer 94:153-156, 2001
Non-Small Cell Lung Cancer Collaborative Group: Chemotherapy for non-small cell lung cancer [Cochrane Database System Review]. Oxford, England, Cochrane Library, CD002139, issue 2, 2000
Shepherd FA, Dancey J, Ramlau R, et al: Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 18:2095-2103, 2000
Shepherd FA, Pereira J, Ciuleanu T, et al: A randomized placebo-controlled study of erlotinib (OSI-774, Tarceva) versus placebo in patients with incurable non-small cell lung cancer who have failed standard therapy for advanced or metastatic disease. Proc Am Soc Clin Oncol 22:622S, 2004 (abstr 7022)
Janecki-Delebecq T, Porte H, Zerimech F, et al: Overexpression level of stromelysin 3 is related to the lymph node involvement in non-small cell lung cancer. Clin Cancer Res 6:1086-1092, 2000
Passlick B, Srenel W, Seen-Hibler R, et al: Overexpression of matrix metalloproteinases predicts unfavorable outcome in early-stage non-small cell lung cancer. Clin Cancer Res 6:3944-3948, 2000
Brown PD, Bloxidge RE, Stuart NS, et al: Association between expression of activated 72-kilodalton gelatinase and tumor spread in non-small-cell lung carcinoma. J Natl Cancer Inst 85:574-578, 1993
Garbisa S, Scagliotti G, Masiero L, et al: Correlation of serum metalloproteinase levels with lung cancer metastasis and response to therapy. Cancer Res 52:4548-4549, 1992
Thomas P, Khokha R, Shepherd FA, et al: Differential expression of matrix metalloproteinases and their inhibitors in non-small cell lung cancer. J Pathol 190:150-156, 2000
Simi L, Andreani M, Davini F, et al: Simultaneous measurement of MMP9 and TIMP1 mRNA in human non small cell lung cancers by multiplex real time RT-PCR. Lung Cancer 45:171-179, 2004
Aljada IS, Ramnath N, Donohue K, et al: Upregulation of the Tissue Inhibitor of Metalloproteinase-1 Protein is associated with progression of human non-small cell lung cancer. J Clin Oncol 22:3218-3229, 2004
Shepherd FA, Sridhar SS: Angiogenesis inhibitors under study for the treatment of lung cancer. Lung Cancer 41:S63-S72, 2003
Shepherd FA, Giaccone G, Seymour L, et al: Prospective, randomized, double-blind, placebo-controlled trial of marimastat after response to first-line chemotherapy in patients with small-cell lung cancer: A trial of the National Cancer Institute of Canada–Clinical Trials Group and the European Organization for Research and Treatment of Cancer. J Clin Oncol 20:4434-4439, 2002
Bissett D, O’Byrne K, von Pawel J, et al: Phase III study of the matrix metalloprotease (MMP) inhibitor prinomastat in combination with gemcitabine and cisplatin in non-small cell lung cancer. Proc Am Soc Clin Oncol 21:296a, 2002 (abstr 1183)
Douillard JY, Peschel C, Shepherd F, et al: Randomized phase II feasibility study of combining the matrix metalloproteinase inhibitor BMS-275291 with paclitaxel plus carboplatin in advanced non-small cell lung cancer. Lung Cancer 46:361-368, 2004
Bristol-Myers Squibb Pharmaceutical Research Institute: Investigator Brochure—Matrix metalloproteinase inhibitor BMS-275291, version 4. Bristol-Myers Squibb, Wallingford, CT, 2002
Lombard MA, Wallace TL, Kubicek MF, et al: Synthetic matrix metalloproteinase inhibitors and tissue inhibitor of metalloproteinase (TIMP)-2, but not TIMP-1, inhibit shedding of tumor necrosis factor-alpha receptors in a human colon adenocarcinoma (Colo 205) cell line. Cancer Res 58:4001-4007, 1998
Rizvi NA, Humphrey JS, Ness EA, et al: A phase I study of oral BMS-275291, a novel nonhydroxamate sheddase-sparing matrix metalloproteinase inhibitor, in patients with advanced or metastatic cancer. Clin Cancer Res 10:1963-1970, 2004
Douillard JY, Peschel C, Shepherd F, et al: Randomized phase II study of combining the matrix metalloproteinase inhibitor BMS-275291 with paclitaxel plus carboplatin in advanced non-small cell lung cancer. Lung Cancer 46:361-368, 2004
Tu D: Minimization Procedure, in Chow SC (ed) Encyclopaedia of Biopharmaceutical Statistics. New York: Marcel Dekker, Inc, 2003, pp 614-618
Miller KD, Saphner TJ, Waerhouse DM, et al: A randomized phase II feasibility trial of BMS-275291 in patients with early stage breast cancer. Clin Cancer Res 10:1971-1975, 2004
Lara PN, Longmate J, Stadler W, et al: Angiogenesis inhibition in metastatic hormone refractory prostate cancer (HRPC): A randomized phase II trial of two doses of the matrix metalloproteinase inhibitor (MMPI) BMS-275291. Proc Am Soc Clin Oncol 23:417, 2004 (abstr 4647)(Natasha B. Leighl, Luis P)