Final report of an open-label phase II trial of bevacizumab plus docetaxel and gemcitabine in metastatic, previously untreated nonsquamous non-small cell lung cancer (NSCLC).

Abstract

e18046 Background: Platinum and non-platinum doublet chemotherapy has similar efficacy in advanced NSCLC patients (pts). Bevacizumab (B) improves outcomes when added to platinum doublets, but its safety and efficacy in combination with non-platinum doublets is unknown. This study was designed to test the combination of B, docetaxel (D), and gemcitabine (G) in first-line treatment of advanced NSCLC. Methods: Pts with metastatic, non-squamous NSCLC, PS 0-1, and measurable disease by RECIST were enrolled in this open-label, single arm phase II trial. Pts received D (75 mg/m2) on d1, G (900 mg/m2) on d1 & 8, and B (15 mg/kg) on d1 every 21d for up to 6 cycles, followed by B maintenance until progression or 12 mos total. Pts received growth factor d9. CT scans were performed every 6 wks. The primary endpoint was 1-yr progression-free survival (PFS), with secondary endpoints of safety, objective response rate (ORR), overall PFS, and overall survival (OS). Pts with tumor cavitation, untreated brain metastases, and hemoptysis were excluded. Planned enrollment was 46. Results: 13 pts were enrolled from 12/2009 to 4/2011. Pt characteristics: Median age 63 (35-69), 85% male, PS 0 (38%), PS 1 (62%). The median # of cycles of chemotherapy was 6 (1-6), median # cycles of B was 4 (1-15), with 2 pts coming off study prior to the first evaluation (1 grade 5 encephalopathy, 1 grade 4 febrile neutropenia). 5 pts (38%) had chemo dose reduction and 4 (31%) discontinued treatment for toxicity. 3 pts (23%) discontinued B prior to progression, 2 for tumor cavitation and 1 for grade 1 hemoptysis. The grade 3-5 non-hematologic toxicity rate was 69%, with 6 pts (46%) hospitalized with pneumonitis/pneumonia felt possibly related to study drugs. At this point enrollment was halted for safety concerns. The 1-yr PFS was 8%, and the median PFS was 6.9 mos (95% CI 2.0-NYR). 11 pts were evaluable for response, and 6 pts had partial responses for an ORR of 55%. The median OS was NYR with median follow up of 10.9 mos. Conclusions: The combination of B, D, and G was not tolerable at the doses and schedule used in this study. A formal phase I trial is needed if this combination is to be investigated further.