A phase 1/2 study of osimertinib and bevacizumab as initial treatment for patients with metastatic EGFR-mutant lung cancers.

Abstract

9086 Background: Osimertinib (osi) demonstrated improved progression-free survival (PFS) over erlotinib as initial treatment (trmt) for EGFR-mutant (EGFR+) lung cancers. The addition of bevacizumab (bev) to erlotinib as initial trmt resulted in improved PFS compared to erlotinib alone (16 vs 10 months, HR 0.41). The phase 1 study of osi and bev confirmed the ability to combine osi and bev at full doses. Methods: The phase 2 study is assessing safety and efficacy of the combination. The primary endpoint is PFS at 12 months; secondary endpoints include overall response (ORR), overall survival (OS), and CNS PFS. All pts had interval CNS as well as systemic imaging. Pre-treatment and post-progression (PD) tumor tissue and interval plasma samples are being collected to identify mechanisms of resistance (MOR) and for biomarker assessment. Results: From Nov 2016 to May 2018, 49 pts were enrolled, including 6 pts from the phase 1. Median age: 60; Women 69%; Never-smokers 65%. 13 pts had CNS metastases (9 untreated, 5 measurable) prior to study initiation. 49 pts are eligible for response; 34/49 pts had a confirmed partial response (PR)(ORR 69%). PFS at 12 months is 0.70 (95% CI: 0.57-0.84), with 8/49 pts on study without PD for less than 12 months. All pts with measurable CNS disease had a PR in the CNS; PD in the CNS was uncommon (17%). 24 pts remain on study without PD; 2 are being treated beyond PD. Reasons for study discontinuation include PD (n = 16), resection of all sites of disease (n = 3), withdrawal of consent (n = 3), unrelated death (n = 2), toxicity (n = 1). The most frequent trtmt-related adverse events (any grade) were thrombocytopenia (61%), diarrhea (57%), hypertension (55%), and rash (47%). 24% required a dose reduction of osi, 18% discontinued bev and median doses of bev was 17. 9 pts have paired pre-trtmt and post-progression tissue biopsies; MOR identified include squamous cell (n = 2) and small cell (n = 1) transformation, PTEN loss, and CCNE amplification. Conclusions: Combination osi and bev was well-tolerated and efficacy to date supports further evaluation. Results of secondary endpoints including PFS, mechanisms of resistance, cfDNA data are forthcoming. A randomized study of osi compared to osi and bev is planned (EA5182). Supported by AstraZeneca Clinical trial information: NCT02803203.