Abstract 1161
Background
A retrospective subset analysis from a phase 3 trial, and preclinical data supported prospective study of elrotinib and bevacizumab in patients (pts) with EGFR mutant NSCLC.
Methods
Pts were required to have an EGFR exon 19 deletion or exon 21 L858R mutation based on local testing. Pts were assigned to erlotinib 150 mg daily (E) or erlotinib 150 mg daily and bevacizumab 15 mg/kg IV every 3 weeks (EB). The primary objective was progression-free survival (PFS); secondary objectives were objective response rate (ORR), and overall survival (OS). cfDNA samples were collected at baseline, at time of imaging, and disease progression. The study was designed to detect a hazard ratio (HR) of 0.667 in favor of EB, with a power of 81% at one-sided significance level of 0.20; under exponential hazards. HR corresponds to an improvement in median PFS from 10 to 15 months.
Results
From 11/2012 to 8/2016 88 pts were enrolled. The median age was 63 years (range 31 to 84), the majority were women (70%), had a history of never smoking (55%), performance status (PS) of 1 (51%), and EGFR exon 19 deletion (67%). With a median follow-up of 23 months, 69 PFS events have been observed. A statistically significant difference in PFS in pts assigned to EB compared to E was not observed (HR of 0.87, 95 CI: 0.54-1.43; p = 0.59; median PFS 17.9 and 13.5 months, respectively). The ORR in the EB and E arms were 83% vs 81% (p = 1.0). 33 OS events have been observed. The OS analysis of pts assigned to EB compared to E revealed a HR of 1.54, 95% CI: 0.74-3.19, p = 0.25; median OS 29.9 months and not evaluable, respectively. Grade ≥ 3 adverse events (rate ≥ 10%) in the EB and E arms were: rash (26% and 18%), diarrhea (9% and 13%), hypertension (40% and 22%), and proteinuria (12% and 0%). In EB and E arms 23 and 21 pts received subsequent therapies; in the EB and E arms osimertinib was a subsequent therapy in 10 and 13 pts, respectively. cfDNA were available for 36/69 pts with progressive disease; exon 19 deletion or exon 21 L858R detected in 12 samples and T790M in 5 samples.
Conclusions
Treatment with EB compared to E did not result in a statistically significant improvement in PFS in pts with EGFR mutant NSCLC. OS data are immature. A review of subsequent therapies and additional cfDNA analyses are ongoing.
Clinical trial identification
NCT01532089.
Legal entity responsible for the study
Academic and Community Cancer Research United.
Funding
Genentech.
Editorial Acknowledgement
Disclosure
T.E. Stinchcombe: Advisory board related to immunotherapy: Genentech; Advisory board: AstraZeneca; Research funding to institution for clinical trials: Genentech/Roche and AstraZeneca. P.A. Janne: Consultant: BI, Pfizer, Roche/Genentech, Chugai, Ariad, AstraZeneca, Merrimack, Acea Bioscience; Research funding: Astellas, AstraZeneca, BI; Post-marketing royalties: Dana Farber Cancer Institute for intellectual property on EGFR mutations licensed to Lab Corp. E.M. Bertino: Advisory board: BI. E.E. Vokes: Consultant/advisory roles: AstraZeneca, Genentech. All other authors have declared no conflicts of interest.