Randomized phase II/III trial of paclitaxel (P) plus carboplatin with or without bevacizumab (NSC #704865) in patients with advanced non-squamous non-small cell lung cancer (NSCLC): An Eastern Cooperative Oncology Group (ECOG) Trial - E4599

Reviewer: S. Jack Wei, MD
University of Pennsylvania School of Medicine
Last Modified: May 15, 2005

Presenter: A.B. Sandler
Presenter's Affiliation: Easteran Cooperative Oncology Group
Type of Session: Plenary

  • The standard therapy for patients with advanced NSCLC is platinum-containing doublet chemotherapy.
  • Randomized trials of targeted therapies in chemotherapy-naïve patients with advanced NSCLC have all been negative, including trials targeting epidermal growth factor receptor, matrix metalloprotease, farnesyl transferase, PKC-alpha, and RXR.
  • Tumor growth is dependent on angiogenesis, which in turn is dependent on the vascular endothelial growth factor (VEGF) pathway .
  • VEGF is frequently overexpressed in NSCLC and is associated with poor prognosis.
  • Bevacizumab is a recombinant humanized monoclonal antibody to VEGF-A.
  • The combination of bevacizumab and chemotherapy has resulted in a survival advantage when compared to chemotherapy alone in patients with metastatic colorectal cancer.
  • A randomized phase II trial of bevacizumab and chemotherapy has been conducted in previously untreated patients with advanced NSCLC.
    • Patients who received chemotherapy + bevacizumab (15 mg/kg) showed higher response rates (RR), median time to progression (TTP), and median survival (MS) than chemotherapy + bevacizumab (7.5 mg/kg) and chemotherapy alone.
    • However, there were six life-threatening hemorrhages in this trial, four of which were fatal (overall rate=9%).
    • 31% of patients with squamous cell histology experienced life-threatening hemorrhage compared to 4% of patients with non-squamous histology.
    • Patients with baseline hemoptysis were also at high risk for life-threatening hemorrhage.
  • This trial was designed to determine if the addition of bevacizumab improved outcomes compared to chemotherapy alone in patients with non-squamous NSCLC.
Materials and Methods
  • Eligible patients had stage IIIB (pleural or pericardial effusion only) or IV NSCLC of non-squamous histology, ECOG performance status 0-1, no history of thrombotic or hemorrhagic disorders, INR <1.5 and PTT no greater than upper limits of normal, no history of hemoptysis, and no CNS metastases confirmed by radiographic imaging.
  • Patients were randomized to:
  1. PC: Paclitaxel (200 mg/m2) + carboplatin (AUC=6) q 3wk x 6 cycles
  2. PCB: PC x 6 cycles + bevacizumab (15 mg/kg) q3wks until disease progression
  • Patients were stratified by radiotherapy (RT) vs. no RT, stage IIIB vs. IV vs. recurrent, weight loss <5% vs. >=5%, and measurable vs. non-measurable disease.
  • Primary endpoint was overall survival (OS) with secondary endpoints of RR, TTP, and toxicity.
  • A planned sample size of 842 patients was designed to provide 80% power to detect a 25% improvement in MS with a 1-sided 2.5% type 1 error.
  • Interim analyses were planned at 286 and 455 deaths with final analysis planned at 650 deaths.
  • Between 7/01 and 4/04, 878 patients were enrolled with a total of 855 evaluated.
  • This report comes after the second interim analysis as of 2/9/05 after a total of 484 deaths.
  • Patient groups were well-balanced with regards to stage, measurable disease, prior weight loss, age, performance status, gender, and ethnicity.
  • Grade 4 neutropenia: PC 16.4% vs. PCB 24% (p=0.006)
  • Grade 4 thrombocytopenia: PC 0% vs. PCB 1.4% (p=0.01)
  • There was no difference in rates of anemia or febrile neutropenia.
  • Grade 3-5 hemmorhage was more common in the PCB arm (4.5% vs. 0.7%, p<0.001).
  • Grade 3-5 hypertension was more common in the PCB arm (6.0% vs. 0.7%, p<0.001).
  • 8 treatment-related deaths (5 hemoptysis, 2 GI bleed, 1 neutropenic fever) occured in the PCB arm.
  • 2 treatment-related deaths (1 GI bleed, 1 neutropenic fever) occured in the PC arm.
  • Overall RR: PC 10.0% vs. PCB 27.2% (p<0.0001).
  • 1 year PFS: PC 6.4% vs. PCB 14.6% (HR 0.62, p<0.0001).
  • 2 year OS: PC 16.9% vs. PCB 22.1% (HR 0.77, p=0.007).
  • Median survival: PC 10.2 mo vs. 12.5 mo.
  • The survival benefit was seen across all subgroups except for gender.
  • Hazard ratio for survival was 0.69 for men (p=0.003) and 0.96 for women (p=0.80).
  • PFS and RR was significantly better in the PCB regardless of gender.
  • Correlative studies of outcome with pre-treatment and post-2nd cycle VEGF, soluble E-selectin, soluble intercellular adhesion molecule-1 (ICAM), and basic fibroblast growth factor (bFGF) are ongoing.
Author's Conclusions
  • Bevacizumab improves survival when added to PC chemotherapy in patients with advanced non-squamous NSCLC.
  • Bevacizumab also improvces RR and PFS.
  • Bevacizumab is associated with small increases in serious bleeding events including 5 deaths from hemoptysis.
  • PCB is now the ECOG reference standard for the first-line treatment of advanced non-squamous cell NSCLC.
  • Future plans include combination of bevacizumab with chemotherapy and radiotherapy, combination with other targeted agents, and use of bevacizumab in the neo-adjuvant and adjuvant settings.

Clinical/Scientific Implications

This study shows significant improvement in OS, PFS, and RR with the addition of bevacizumab to standard chemotherapy in treatment-naive patients with non-squamous NSCLC. This study follows positive studies in colorectal cancer and breast cancer that showed an overall survival benefit with the addition of bevacizumab to standard chemotherapy in advanced disease. Interestingly, compared to these other trials, there appears to be a higher rate of severe, life-threatening toxicity in this trial. The 4.5% rate of grade 3-5 toxicity was consistent with the 4% rate of life-threatening hemorrhage seen in the phase II trial with bevacuzimab in patients with non-squamous histology. It is concerning that 8 patients died of treatment-related complications including 5 patients with severe hemorrhage in this study. Clearly, bevacizumab should be used with caution in these patients. Nevertheless, despite the higher rate of treatment related deaths in the PCB arm of this study, there was a survival advantage to the use of bevacizumab, and it is reasonable to use bevacizumab in patients with advanced NSCLC of non-squamous histology with no history of coagulopathies or hemoptysis.

It should be noted that although the benefit of bevacizumab was limited to males, the reason for this is unclear. It is possible that other causes, such as use of secondary therapy with tyrosine kinase inhibitors (which have been found to preferentially benefit women), account for these differences. Despite this interesting finding, it is not consistent with other studies of bevacuzimab which have not shown a preferential benefit in males, and in fact is contradictory to the pattern of NSCLC in general where women have improved outcomes compared to men. Bevacuzimab remains an interesting option ifor all patients n the growing list of targeted therapies that are being used in NSCLC.


Clinical Trials
by Bob Riter
July 5, 2012

Chemo and Radiation for Late Stage Oropharynx Cancer [Webinar]
by Carolyn Vachani, MSN, RN, AOCN
February 12, 2014

Stay informed with the latest information from OncoLink!   Subscribe to OncoLink eNews
View our newsletter archives