Local Control of Brain Metastases with Osimertinib Alone in Patients with EGFR-Mutant Non-Small Cell Lung Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Osimertinib has emerged as a new standard of care in the treatment of EGFR-mutant metastatic non-small cell lung cancer (NSCLC). However, data on lesion-level control and durability of intracranial response to osimertinib is limited. We aim here to describe longitudinal lesion-level outcomes of brain metastases treated with osimertinib alone. <h3>Materials/Methods</h3> We retrospectively reviewed charts of patients with EGFR-mutant NSCLC with at least one untreated brain metastasis at the time of initiation of osimertinib at our institution from 2006-2021. An FDA-approved brain tumor segmentation tool, VBrain, was used to identify, track, and measure brain metastases larger than 5 mm on serial MRI brain scans. Patients with leptomeningeal disease were excluded. Cumulative incidence of local failure (LF) was calculated with competing risk analysis with death as a competing risk and censored at the last brain MRI follow up. LF was defined as a ≥20% increase in the maximum diameter of a brain lesion using linear measurements on serial imaging. LF was calculated on a per-lesion basis, and univariable and multivariable analysis was conducted with competing risks regression used to identify predictors for LF. Overall survival (OS) was analyzed using Kaplan-Meier. <h3>Results</h3> A total of 73 patients with 284 untreated brain metastases larger than 5 mm were identified. Median follow-up time was 17 months (range 1.3-50.6 months). Median age at brain metastases diagnosis was 60 years old (range 30-90); 68.5% were female. Median number of prior systemic therapies received was 0 (range 0-6), with 32.9%, 27.4%, and 15.1% of patients receiving prior tyrosine kinase inhibitors, chemotherapy, and immunotherapy, respectively. The median size of brain metastases was 6.9 mm (range 5.0-33.6 mm) and the median number of brain metastases larger than 5 mm per patient was 1 (range 0-43). Median OS was 24.4 months (95% confidence interval [CI] 17.6-34.9). The 1-year cumulative incidence of LF was 12.1% (95% CI 8.8-16.1). On univariable analysis, uncontrolled primary tumor (hazard ratio [HR] 3.78, 95% CI 1.87-7.66; <i>p</i><0.001), increasing number of prior systemic therapies before osimertinib (HR 2.12, 95% CI 1.49-3.04; <i>p</i><0.001), and higher ECOG score (HR 7.8, 95% CI 1.99-31.81; <i>p</i>=0.003) was associated with LF. These factors remained significant in multivariable analysis. Median duration on osimertinib was 14.8 months (range 0.2-48.3 months), and 26% of patients experienced toxicities from osimertinib that required either a change in systemic therapy or a dose reduction. 9.6% and 18.6% of patients underwent salvage whole brain radiation therapy and salvage stereotactic radiosurgery, respectively. <h3>Conclusion</h3> Although NSCLC patients have excellent response of untreated intracranial metastases on osimertinib alone, over 10% of metastases will progress locally at 1 year. Further studies are needed to identify the brain metastases at risk for local failure to guide optimal intracranial management in these patients.