First-line osimertinib shows high response rates in EGFR-mutated NSCLC with active brain metastases.

INTRODUCTION: Osimertinib has documented CNS activity, but there is little data on the effect on untreated brain metastases (BM). We assessed the efficacy of osimertinib in patients with or without active BM and investigated whether circulating tumour DNA (ctDNA) at baseline provides prognostic information. METHODS: In this single-arm phase II clinical trial, patients with EGFR-mutated non-small cell lung cancer (NSCLC), with (cohort A) or without active BM (cohort B), received first-line treatment with osimertinib. The primary endpoint was objective response rate (ORR). Baseline plasma samples were analysed for the presence of ctDNA-mutations. RESULTS: One hundred patients were included: 46 in cohort A and 54 in cohort B. The ORR was 72.0% for the entire study population, and 69.6% and 74.1% in cohort A and B, respectively. Patients with measurable BM had an intracranial ORR of 81.8%. No significant differences in progression-free survival (PFS) or overall survival (OS) were observed between the two cohorts. Harbouring the L858R-mutation or uncommon EGFR-mutations was associated with a significantly shorter PFS (p = 0.010) and OS (p = 0.002) than for the exon19-deletion, irrespective of the presence of BM. ctDNA-mutations were detected in 83 of 97 available baseline plasma samples (85.6%). Absence of baseline ctDNA was associated with significantly improved PFS (p = 0.042) and OS (p = 0.028). CONCLUSIONS: First-line osimertinib treatment is effective in patients with active BM. The subtype of EGFR-mutations and the presence of ctDNA at baseline are all associated with poorer outcomes, and seem to be stronger predictors than the presence of BM.