Key Words. Bevacizumab Avastin Nonsquamous Non-small cell lung cancer First-line Advanced/metastatic disease

Transcription

1 The Oncologist Regulatory Issues: FDA FDA Drug Approval Summary: Bevacizumab (Avastin ) Plus Carboplatin and Paclitaxel as First-Line Treatment of Advanced/Metastatic Recurrent Nonsquamous Non-Small Cell Lung Cancer MARTIN H. COHEN, JOE GOOTENBERG, PATRICIA KEEGAN, RICHARD PAZDUR Division of Biological Oncology Products, Office of Oncology Drug Products, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, Maryland, USA Key Words. Bevacizumab Avastin Nonsquamous Non-small cell lung cancer First-line Advanced/metastatic disease LEARNING OBJECTIVES After completing this course, the reader will be able to: 1. Describe the bevacizumab mechanism of action. 2. List the current bevacizumab indications. 3. Discuss the role of bevacizumab in the treatment of specific non-small cell lung cancer histologies. 4. Describe the clinical trial design leading to bevacizumab approval by the FDA. 5. Identify the major adverse events associated with bevacizumab treatment. CME Access and take the CME test online and receive 1 AMA PRA Category 1 Credit at CME.TheOncologist.com ABSTRACT On October 11, 2006, the U.S. Food and Drug Administration granted approval for bevacizumab (Avastin ; Genentech, Inc., South San Francisco, CA), administered in combination with carboplatin and paclitaxel, for the initial treatment of patients with unresectable, locally advanced, recurrent, or metastatic, nonsquamous, non-small cell lung cancer (NSCLC). Approval is based on a significant improvement in overall survival (OS). A randomized, open label, multicenter clinical trial, conducted by the Eastern Cooperative Oncology Group (ECOG), in chemotherapy-naïve patients with stage IIIB/IV nonsquamous NSCLC, evaluated bevacizumab plus carboplatin and paclitaxel (BV/CP, n 434) versus carboplatin and paclitaxel alone (CP, n 444). Exclusion of patients with squamous or predominantly squamous histology was based on life-threatening or fatal hemoptysis occurring in 4 of 13 patients with squamous histology who received a BV/CP regimen in a phase II study. Among the 878 randomized patients, the median age was 63, 46% were female, 76% had stage IV disease, 12% had stage IIIB disease with malignant pleural effusion, 11% had recurrent disease, and 40% had an ECOG performance status score of 0. OS was significantly longer in patients receiving BV/CP Correspondence: Martin H. Cohen, M.D., U.S. Food and Drug Administration, White Oak Campus, New Hampshire Avenue, Building 22, Room 2102, Silver Spring, Maryland , USA. Telephone: ; Fax: ; martin.cohen@fda.hhs.gov Received December 10, 2006; accepted for publication March 21, AlphaMed Press / 2007/$30.00/0 doi: /theoncologist The Oncologist 2007;12:

2 714 Drug Approval Summary Disclosure of potential conflicts of interest is found at the end of this article. INTRODUCTION Bevacizumab is a recombinant humanized monoclonal IgG 1 antibody that selectively binds to and neutralizes the biologic activity of human vascular endothelial growth factor (VEGF) [1 3]. Neutralization of the biologic activity of VEGF can result in the reduction of tumor vascularization and subsequent tumor growth. Bevacizumab (Avastin ; Genentech, Inc., South San Francisco, CA) was approved by the U.S. Food and Drug Administration (FDA) in February 2004 for use in combination with i.v. 5-fluorouracil (5-FU)-based chemotherapy for the first-line treatment of patients with metastatic carcinoma of the colon and rectum (CRC). Bevacizumab in combination with oxaliplatin and 5-FU was approved in June 2006 for the second-line treatment of patients with advanced or metastatic CRC that had received prior irinotecan and 5-FU containing regimen(s). While several platinum-based doublet chemotherapy regimens (vinorelbine/cisplatin, gemcitabine/cisplatin, paclitaxel/cisplatin, and docetaxel/cisplatin) have been approved by the FDA for the first-line treatment of advanced/ metastatic non-small cell lung cancer (NSCLC), these regimens demonstrate only a modest improvement in outcomes, with median survival durations of approximately 8 months, 1-year survival rates of about 33%, and response rates of approximately 20% [4, 5]. New first-line treatments for advanced/metastatic NSCLC are needed. BEVACIZUMAB PHARMACOKINETICS Based on a population pharmacokinetic analysis of 491 patients who received 1 20 mg/kg of bevacizumab weekly, every 2 weeks, or every 3 weeks, the estimated half-life of bevacizumab was approximately 20 days (range, than in those receiving CP alone (median OS, 12.3 versus 10.3 months; hazard ratio [HR], 0.80; p.013, stratified log rank test). Although a consistent effect was observed across most subgroups, in an exploratory analysis, evidence of a survival benefit was not observed in women (HR, 0.99; 95% confidence interval, ). Severe and life-threatening adverse events occurring more frequently in patients receiving BV/CP were neutropenia (27% versus 17%), fatigue (16% versus 13%), hypertension (8% versus 0.7%), infection without neutropenia (7% versus 3%), thrombosis/embolism (5% versus 3%), pneumonitis or pulmonary infiltrate (5% versus 3%), infection with grade 3 or 4 neutropenia (5% versus 2%), febrile neutropenia (5% versus 2%), hyponatremia (4% versus 1%), proteinuria (3% versus 0), and headache (3% versus 0.5%). Fatal, treatment-related adverse events in patients receiving bevacizumab were pulmonary hemorrhage (2.3% versus 0.5%), gastrointestinal hemorrhage, central nervous system infarction, gastrointestinal perforation, myocardial infarction, and neutropenic sepsis. The most serious, and sometimes fatal, bevacizumab toxicities are gastrointestinal perforation, wound healing complications, hemorrhage, arterial thromboembolic events, hypertensive crisis, nephrotic syndrome, congestive heart failure, and neutropenic sepsis. The most common adverse events in patients receiving bevacizumab are asthenia, pain, abdominal pain, headache, hypertension, diarrhea, nausea, vomiting, anorexia, stomatitis, constipation, upper respiratory infection, epistaxis, dyspnea, exfoliative dermatitis, and proteinuria. The Oncologist 2007;12: days). The predicted time to reach steady state was 100 days. The accumulation ratio following a dose of 10 mg/kg of bevacizumab every 2 weeks was 2.8. The clearance of bevacizumab varied by body weight, by gender, and by tumor burden. After correcting for body weight, men had a higher bevacizumab clearance rate (0.262 l/day versus l/day) and a larger clearance volume (3.25 l versus 2.66 l) than women. Patients with a higher tumor burden (at or above the median value of tumor surface area) had a higher bevacizumab clearance rate (0.249 l/day versus l/day) than patients with tumor burdens below the median. In a randomized study of 813 patients, there was no evidence of lesser efficacy (hazard ratio [HR] for overall survival [OS]) in men or in patients with higher tumor burden treated with bevacizumab than in women or in patients with a low tumor burden. The relationship between bevacizumab exposure and clinical outcomes has not been explored. Analyses of demographic data suggest that no dose adjustments are necessary for age or sex. No studies have been conducted to examine the pharmacokinetics of bevacizumab in patients with renal or hepatic impairment. PATIENTS AND METHODS The current submission consisted of two studies testing carboplatin/paclitaxel (CP) alone or with bevacizumab (BV/ CP) in previously untreated patients with advanced or metastatic NSCLC. A randomized, open-label, controlled study was conducted by the Eastern Cooperative Oncology Group (ECOG). Eligible patients were 18 years of age with histologically or cytologically confirmed advanced NSCLC

3 Cohen, Gootenberg, Keegan et al. 715 Table 1. Patient demographics CP BV/CP (n 444) (n 434) Age (years) Mean (SD) 62.0 (9.8) 62.3 (10.4) Median Range Age category (years) 40 4 (0.9%) 9 (2.1%) (55.4%) 240 (55.3%) (43.7%) 185 (42.6%) Sex Male 259 (58.3%) 219 (50.5%) Female 185 (41.7%) 215 (49.5%) Race Asian 3 (0.7%) 5 (1.2%) Black 24 (5.4%) 23 (5.3%) Filipino 1 (0.2%) 0 (0.0%) Hispanic 8 (1.8%) 9 (2.1%) Indian (Asian) 0 (0.0%) 1 (0.2%) Native American 2 (0.5%) 1 (0.2%) White 387 (87.2%) 366 (84.3%) Other 19 (4.3%) 29 (6.7%) Abbreviations: BV, bevacizumab; CP, carboplatin and paclitaxel; SD, standard deviation. (defined as stage IIIB with malignant pleural effusion, stage IV, or recurrent disease), except squamous cell carcinoma. Patients had to have adequate organ function and an ECOG performance status score of 0 or 1. Patients with gross hemoptysis ( 1/2 tsp red blood) and those receiving therapeutic anticoagulation were excluded. Study eligible patients were stratified by the presence of measurable disease, prior radiation therapy, degree of weight loss, and disease stage, and were randomized to BV/CP or to CP alone. Patients were treated on day 1 of each of six 3-week cycles. Chemotherapy included paclitaxel, 200 mg/m 2, i.v. infusion over 3 hours, and carboplatin, target area under the curve of 6.0 mg/ml min, i.v. infusion over minutes, immediately after paclitaxel. Bevacizumab, 15 mg/kg i.v., was administered immediately after carboplatin. Upon completion of six cycles of therapy, patients in the CP alone arm were to discontinue treatment while BV/CP patients who did not have progressive disease (PD) continued to receive bevacizumab at a dose of 15 mg/kg i.v. every 3 weeks until PD or unacceptable toxicity. Patients who discontinued protocol therapy were to be followed every 3 months for up to 2 years from study entry, and then every 6 months for an additional 3 years. The primary efficacy outcome measure was survival duration. Survival analysis was based on an intent-to-treat population (all randomized patients regardless of whether or not the assigned treatment was actually received). Other endpoints were the progression-free survival (PFS) time and objective response rate (ORR). The primary safety analyses were based on all randomized patients who received study treatment, defined as at least one full or partial dose of any of the three study drugs. Two interim analyses Table 2. Disease characteristics CP BV/CP (n 444) (n 434) ECOG performance status score (baseline) n (39.5%) 171 (39.7%) (60.5%) 260 (60.3%) Disease stage n IIIb 56 (12.6%) 52 (12.0%) IV and recurrent 387 (87.4%) 381 (88.0%) IV 345 (77.9%) 324 (74.8%) Recurrent 42 (9.5%) 57 (13.2%) Histological type n Adenocarcinoma 302 (68.3%) 300 (69.3%) Squamous 2 (0.5%) 1 (0.2%) Large cell 30 (6.8%) 18 (4.2%) undifferentiated Bronchioloalveolar 11 (2.5%) 12 (2.8%) carcinoma NSCLC, NOS 86 (19.5%) 79 (18.2%) Other 11 (2.5%) 23 (5.3%) Measurable disease No 41 (9.2%) 36 (8.3%) Yes 403 (90.8%) 397 (91.7%) Body weight loss in previous 6 months n % 316 (71.3%) 308 (72.0%) 5% 127 (28.7%) 120 (28.0%) 5% to 10% 80 (18.1%) 75 (17.5%) 10% to 20% 37 (8.4%) 34 (7.9%) 20% 10 (2.3%) 11 (2.6%) Abbreviations: BV, bevacizumab; CP, carboplatin and paclitaxel; ECOG, Eastern Cooperative Oncology Group; NOS, not otherwise specified; NSCLC, non-small cell lung cancer.

4 716 Drug Approval Summary were specified within the protocol, the first when 286 deaths had occurred and the second when 455 deaths had occurred. The results of these interim analyses, as well as safety data, were monitored by the ECOG Data Monitoring Committee. A supporting randomized, open-label phase II study enrolled 98 patients who received BV/CP or CP alone, with BV at a dose of either 7.5 mg/kg or 15 mg/kg every 3 weeks [6]. Efficacy endpoints were time to disease progression, response rate, duration of response, and survival. That study identified squamous histology as a risk factor for lifethreatening hemoptysis. In that study, 66 patients received BV/CP therapy. Six of those patients experienced a life-threatening bleed that may have been caused by tumor-related hemorrhage from pulmonary tumors. The presumed pulmonary hemorrhage event was fatal in four cases. An analysis performed by the sponsor suggested that squamous histology and bevacizumab therapy were the most likely risk factors for fatal pulmonary bleeds. Based on the identification of a possible new safety signal involving massive pulmonary hemorrhage in patients treated with bevacizumab for squamous histology NSCLC, the ECOG trial excluded patients with predominant squamous histology. RESULTS Patient demographic characteristics are presented in Table 1 and disease characteristics are summarized in Table 2. Demographic characteristics were generally similar across the treatment arms except for sex, with the CP arm having a higher proportion of men (58.3%) than the BV/CP arm (50.5%). Prior therapy was generally similar between the treatment arms. Radiation therapy was administered to 9.1% of study patients, and 18.7% had prior surgery. One BV/CP patient received prior chemotherapy and an additional four BV/CP patients received prior systemic therapy, including vaccine therapy (one patient), gefitinib (two patients), and pamidronate (one patient). Table 3 summarizes poststudy therapy. Some type of treatment was given to 66% of CP-treated patients and 62% of BV/CP-treated patients. Chemotherapy was administered to approximately 50% of patients in each group. OS results are presented in Table 4 and Figure 1. Among the 878 randomized patients, 698 deaths occurred, 363 in the CP arm and 335 in the BV/CP arm. OS was statistically significantly longer among patients receiving BV/CP compared with those receiving CP alone; the median OS duration was 12.3 months versus 10.3 months (HR, 0.80; 95% confidence interval [CI], ; p-value.003, stratified log-rank test). In exploratory analyses across patient subgroups, the impact of bevacizumab on OS was less robust in the following groups: women (HR, 0.99; 95% CI, ), age 65 years (HR, 0.91; 95% CI, ), patients with 5% weight loss at study entry (HR, 0.96; 95% CI, ), and patients with histology other than adenocarcinoma (HRs varying between 0.92 and 1.48). All hazard ratios are from unstratified models. Based on investigator assessment, which was not independently verified, patients were reported to have a longer PFS time and higher objective response rate with BV/CP than with CP alone. Results from a smaller, supportive, phase II trial similarly demonstrated an investigator-assessed significantly better time to progression (TTP) and response rate for patients receiving BV/CP. The TTP was 7.4 months versus 4.2 months and response rates were 31.5% and 18.8%, respectively [6]. Although the safety data were limited by the protocolmandated collection mechanisms, no new serious adverse Table 3. Poststudy therapy CP BV/CP (n 440) (n 429) n of patients who had an ECOG follow-up form 420 (95.5%) 381 (88.8%) Any antitumor therapy after progression No 152 (34.5%) 164 (38.2%) Yes 288 (65.5%) 265 (61.8%) Chemotherapy 220 (50.0%) 205 (47.8%) Gemcitabine 73 (16.6%) 67 (15.6%) Taxane 62 (14.1%) 77 (17.9%) Other 112 (25.5%) 101 (23.5%) Abbreviations: BV, bevacizumab; CP, carboplatin and paclitaxel; ECOG, Eastern Cooperative Oncology Group.

5 Cohen, Gootenberg, Keegan et al. 717 Table 4. Overall survival CP BV/CP (n 444) (n 434) n of patients alive (censored) 81 (18.2%) 99 (22.8%) Duration of survival (months) a Median % CI ( ) ( ) 25th 75th percentile Minimum maximum Stratified analysis b Hazard ratio (relative to 0.80 CP) 95% CI ( ) p-value (relative to CP) Log-rank.0030 Wilcoxon.0040 a Summaries of duration of survival (median, percentiles) were estimated from Kaplan-Meier curves. The 95% CI for the median was computed using the method of Brookmeyer and Crowley. b Hazard ratios were estimated by Cox regression. The strata are tumor measurability (yes versus no), prior radiotherapy (yes versus no), weight loss ( 5% versus 5%), and stage (IIIb versus IV or recurrent). Abbreviations: BV, bevacizumab; CI, confidence interval; CP, carboplatin and paclitaxel. event safety signal was identified from the data. National Cancer Institute Common Toxicity Criteria grade 3 5 nonhematologic and grade 4 and 5 hematologic adverse events in the phase III study occurring at a 2% higher incidence in BV/CP-treated patients than in control patients are listed in Table 5. Other serious adverse reactions in patients receiving bevacizumab include gastrointestinal perforations, wound healing complications, hemorrhage, arterial thromboembolic events, reversible posterior leukoencephalopathy syndrome, and congestive heart failure. Other common Table 5. Nonhematologic and hematologic adverse events occurring at a 2% higher incidence in bevacizumab-treated patients than in control patients n (%) of NSCLC patients NCI-CTC category term a CP BV/CP (n 441) (n 427) Any event 286 (65%) 334 (78%) Blood/bone marrow Neutropenia 76 (17%) 113 (27%) Constitutional symptoms Fatigue 57 (13%) 67 (16%) Cardiovascular (general) Hypertension 3 (0.7%) 33 (8%) Vascular Venous thrombus/embolism 14 (3%) 23 (5%) Infection/febrile neutropenia Infection without neutropenia 12 (3%) 30 (7%) Infection with NCI-CTC 9 (2%) 19 (4%) grade 3 or 4 neutropenia Febrile neutropenia 8 (2%) 23 (5%) Pulmonary/upper respiratory Pneumonitis/pulmonary 11 (3%) 21 (5%) infiltrates Metabolic/laboratory Hyponatremia 5 (1%) 16 (4%) Pain Headache 2 (0.5%) 13 (3%) Renal/genitourinary Proteinuria 0 (0%) 13 (3%) a Only NCI-CTC grade 3 5 nonhematologic and grade 4 5 hematologic adverse events were reported. The median duration of exposure to bevacizumab was 4.9 months. Abbreviations: BV, bevacizumab; CP, carboplatin and paclitaxel; NCI-CTC, National Cancer Institute Common Toxicity Criteria; NSCLC, non-small cell lung cancer. Figure 1. Overall survival of BV/CP and CP treated patients. adverse events in patients receiving bevacizumab were asthenia, pain, abdominal pain, headache, hypertension, diarrhea, nausea, vomiting, anorexia, stomatitis, constipation, upper respiratory infection, epistaxis, dyspnea, and exfoliative dermatitis. In the current study, the rate of pulmonary hemorrhage requiring medical intervention for the BV/CP arm was 2.3% (10 of 427) compared to 0.5% (2 of 441) for the CP alone arm. There were seven deaths due to pulmonary hemorrhage reported by investigators in the CP plus Avastin arm as compared to one in the CP alone arm.

6 718 Drug Approval Summary DISCUSSION BV/CP demonstrated efficacy in a randomized, multicenter trial as evidenced by a clinically meaningful prolongation of survival in patients with chemotherapy-naïve, unresectable, recurrent or metastatic nonsquamous NSCLC compared with patients who received CP alone. The secondary endpoints of PFS and ORR, although based on lesser quality data, support the primary efficacy endpoint. The lack of a survival benefit in an exploratory analysis of the subgroup of female patients is concerning. A Cox multivariate analysis suggests that a treatment effect of bevacizumab on survival prolongation was present in female patients despite the results of the univariate analysis. In addition to the multivariate analysis results, bevacizumab was also found to positively affect both PFS and ORR for female patients in favor of the BV/CP arm, although as previously indicated, these secondary endpoints are less reliable than the survival endpoint. Similar considerations might apply to other subgroups that did not demonstrate survival benefit with bevacizumab treatment, including patients 65 years of age or older and patients with histologic subtypes other than adenocarcinoma or large cell undifferentiated cancer. Possible imbalances in baseline demographic characteristics and poststudy therapy might account for some of the observed results. The small number of patients, especially in the survival outcome by histology results, is another confounding factor. REFERENCES 1 Jain RK, Duda DG, Clark JW et al. Lessons from phase III clinical trials on anti-vegf therapy for cancer. Nat Clin Pract Oncol 2006;3: Kerbel R, Folkman J. Clinical translation of angiogenesis inhibitors. Nat Rev Cancer 2002;2: Ferrara N, Hillan KJ, Gerber HP et al. Discovery and development of bevacizumab, an anti-vegf antibody for treating cancer. Nat Rev Drug Discov 2004;3: Schiller JH, Harrington D, Belani CP et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 2002; 346: Bunn PA Jr, Vokes EE, Langer CJ et al. An update on North American randomized studies in non-small cell lung cancer. Semin Oncol 1998;25(suppl 9): Johnson DH, Fehrenbacher L, Novotny WF et al. Randomized phase II trial Of interest, the BV/CP survival benefit was observed despite the administration of poststudy chemotherapy. Approximately 50% of patients received such treatment. Both erlotinib [7, 8] and docetaxel [9, 10] have been shown to increase survival when administered as second-line treatments. Because this study demonstrated clinical benefit (survival prolongation), regular approval was granted. There are no required phase IV commitments with regular approval. Phase IV requests included submission of data from ongoing first-line bevacizumab plus chemotherapy studies in patients with advanced or recurrent nonsquamous NSCLC, to update safety information on bevacizumab administered to patients with previously treated central nervous system metastases, and to conduct a substudy to address the impact of bevacizumab on the QT interval. ACKNOWLEDGMENTS The views expressed are the result of independent work and do not necessarily represent the views and findings of the U.S. Food and Drug Administration. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST The authors indicate no potential conflicts of interest. comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 2004;22: Cohen MH, Johnson JR, Chen YF et al. FDA drug approval summary: Erlotinib (Tarceva) tablets. The Oncologist 2005;10: Shepherd FA, Rodrigues Pereira J, Ciuleanu T et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353: 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 2000;18: Fossella FV, DeVore R, Kerr RN et al. Randomized phase III trial of docetaxel versus vinorelbine or ifosfamide in patients with advanced nonsmall-cell lung cancer previously treated with platinum-containing chemotherapy regimens. The TAX 320 Non-Small Cell Lung Cancer Study Group. J Clin Oncol 2000;18: