MLTI-04. EVOLUTION OF TREATMENT PARADIGMS FOR PATIENTS WITH ≥1 BRAIN METASTASES FROM PRIMARY NON-SMALL-CELL LUNG CANCER – A SYSTEMATIC REVIEW

BACKGROUND: Brain metastases (BM) are common in non-small cell lung cancer (NSCLC), with approximately 10% of patients presenting with BM at the time of diagnosis. The aim of this systematic review was to critically evaluate the evolution of management paradigms for BM from NSCLC. METHODS: We searched MEDLINE, EMBASE, Web of Science, ClinicalTrials.gov, and CENTRAL for randomized controlled trials (RCTs) published until October 2018. Comparative RCTs based on ≥ 50 patients were selected. The primary outcomes of interest were overall survival (OS) and progression-free survival (PFS). RESULTS: Among 3188 abstracts, 14 RCTs (2494 patients) met inclusion criteria. Median sample size was 97 (range 59–538). Most trials were open-label, parallel, superiority trials. All included patients aged ≥18 with histologically proven NSCLC and ≥1 BM proven on CT/MRI. The majority of trials (11/14) excluded patients with non-favorable performance status (ECOG, KPS, or WHO scales), prior SRS or WBRT, and/or leptomeningeal metastases. Interventions assessed included WBRT (11/14), SRS (3/14), targeted therapies (e.g. EGFR inhibitors, 5/14), and various chemotherapeutic regimens (12/14). Most trials (12/13) reported no significant difference in OS between interventions. 4/10 trials reported a difference in PFS, two of which only included patients with EGFR-mutant NSCLC; these showed a significant increase in PFS in patients managed with EGFR inhibitors. The other two trials reported longer PFS with sodium glycididazole + WBRT vs. WBRT alone (p=0.038) and temozolomide + SRS vs. SRS alone (p=0.003). The incidence of adverse events was consistent across most treatment groups. CONCLUSIONS: Most trials showed no significant improvement in OS; however, improvement in PFS was seen in several trials, most notably in EGFR-positive patients treated with EGFR inhibitors. Given the long-standing merit of radiation-based therapies for BM management, these data support the need for an in-depth meta-analysis assessing the comparative efficacy of current management paradigms for specific patient populations.