VEGF Antibody in the Maintenance Setting
Bevacizumab is a recombinant, humanized monoclonal antibody against vascular endothelial growth factor (VEGF), a critical factor in tumor angiogenesis.1 The antiangiogenic mechanism of action of bevacizumab supports its continued use until progressive disease (PD). In preclinical lung cancer models2,3 and in clinical settings such as colorectal4,5 and ovarian6 cancers, continuous VEGF suppression with bevacizumab has been shown to be key to tumor control. In the Eastern Cooperative Oncology Group (ECOG) 4599 study,7 patients with previously untreated advanced, metastatic, or recurrent non–small cell lung cancer (NSCLC) were randomized to 6 cycles of carboplatin (C) + paclitaxel (P) with or without bevacizumab. Patients without PD after induction therapy in the CP+bev arm continued to receive bevacizumab until PD or unacceptable toxicity. Median progression-free survival (PFS) was significantly improved in the CP+bev arm (6.2 vs 4.5 months; hazard ratio [HR] = 0.66; P <.0001). Median overall survival (OS), the primary endpoint, was also significantly increased in the CP+bev arm (12.3 vs 10.3 months; HR = 0.79; P = .003). Because current data on single-agent bevacizumab maintenance in NSCLC are limited, we conducted and present here a retrospective analysis of safety and efficacy outcomes for patients who received maintenance bevacizumab after induction treatment and the maintenance-eligible population of the control arm in ECOG 4599.
Landmark analyses were conducted in patients in both the CP and CP+bev groups who were alive and progression-free through the completion of 6 cycles + 21 days. The landmark date was chosen to approximate the end of the induction period for both patient groups and the beginning of the seventh cycle of therapy for patients receiving maintenance bevacizumab. The bevacizumab maintenance population consisted of patients in the CP+bev arm without PD before the start of maintenance (maintenance non-progressor population) CP non-progressors were those patients in the CP-alone arm without PD after 6 cycles of CP + 21 days. Response rates, PFS, OS, and 1-year survival rates were estimated using Kaplan-Meier methods. HRs were based on a Cox model that adjusted for baseline factors, including age, sex, race, ECOG PS, stage of disease, weight loss, disease histology, and best response to induction therapy.
Patient Disposition and Demographics
Between July 2001 and April 2004, 869 patients were randomized to treatment. Among the 440 patients randomized to CP, 194 (44%) completed 6 cycles of induction therapy and 134 (30%) were alive and progression-free at the landmark date. Among the 429 patients randomized to CP+bev, 258 (60%) completed 6 cycles of induction therapy and 217 (51%) were alive and progression-free at the landmark date. Patient and disease characteristics were similar for the CP non-progressor and bevacizumab maintenance populations (Table 1).
Table 1. Patient and Disease Characteristics at Baseline
(n = 134)
(n = 217)
|Analysis population (n = 351)||P valueα|
|Age <70 years||102 (76)||165 (76)||267 (76)||.986|
|Female sex||62 (46)||107 (49)||169 (48)||.580|
|White, race||122 (91)||183 (84)||305 (87)||.070|
|ECOG PS 0 vs 1||69 (51)||97 (45)||166 (47)||.216|
|Stage IIIB vs IV||16 (12)||33 (15)||49 (14)||.391|
|Weight loss <5%||105 (78)||175 (81)||280 (80)||.604|
|Adenocarcinoma||99 (74)||161 (74)||260 (74)||.948|
|Best induction response
Efficacy outcomes for patients in the CP non-progressor and bevacizumab maintenance groups are presented in Table 2. 162 of 217 patients in the bevacizumab maintenance group (75%) and 92 of 134 patients in the CP non-progressor group (69%) were alive and on-study 1 year after the start of induction therapy. The 2-year OS and 1-year PFS rates were also higher for the Bevacizumab maintenance population compared with CP non-progressors. Post-induction PFS was significantly longer in the bevacizumab maintenance group relative to CP non-progressors (4.4 vs 2.8 months; HR 0.64; P<.001). Median post induction OS was also significantly longer for the bevacizumab maintenance group compared with CP non-progressors (12.8 vs 11.4 months; HR 0.75; P=.030).
Table 2. Summary of Efficacy Outcomes by Treatment Group for Patients in ECOG 4599 Who Were Alive and Progression-Free at 6 Cycles + 21 Days
|Median postinduction PFS,a mo
|2.8 . . . . . . . . . . . . . . . . . . . . . .4.4
|Median PFS,b mo||7.2||8.7|
|Median postinduction OS,a mo
|11.4 . . . . . . . . . . . . . . . . . . . . . .12.8
|Median OS,b mo||15.8||17.0|
|1-year PFS rate, % (95% CI; n at risk)||17 (11-25; 21)||32 (26-39; 62)|
|1-year OS rate, % (95% CI; n at risk)||69 (61-77; 92)||75 (69-81; 162)|
|2-year OS rate, % (95% CI; n at risk)||25 (18-34; 17)||34 (28-42; 46)|
In this retrospective analysis of patients in the ECOG 4599 study who were alive, progression-free, and on-study 21 days after 6 cycles of induction therapy, significant reductions in HRs for progression (0.64, P<.001) and survival (0.75, P=.03) were associated with bevacizumab treatment during induction and maintenance compared with CP induction therapy alone. These results suggest possible benefit due to bevacizumab maintenance therapy in patients with advanced NSCLC.