Association of baseline corticosteroid treatment with outcomes for patients (pts) with BRAF-mutant melanoma brain metastases (MBMs) in COMBI-MB treated with dabrafenib and trametinib (DT).

Abstract

e21546 Background: The COMBI-MB trial showed that DT was safe for BRAF-mutant melanoma pts with MBMs. However, slightly lower intracranial response rates (ICRR) and markedly shorter intracranial progression-free survival (PFS) were observed than in previous trials in pts with extracranial metastases (EM) only. In order to understand the basis of this difference in outcomes, and to inform patient management, we performed post hoc analyses of baseline clinical features to identify associations with ICRR, PFS, and overall survival (OS). Methods: Analyses were performed on all patients (n = 125) enrolled in the COMBI-MB trial. Baseline features included age, gender, ECOG performance status (PS), BRAF genotype (V600E vs V600 other), number of MBM, sum of longest diameters (SLD) of MBMs, maximum MBM size, presence of EM, serum lactate dehydrogenase (LDH), baseline corticosteroid treatment, prior treatments (systemic, radiotherapy to brain, surgery to brain), and presence of uncontrolled MBM symptoms. Univariate (UV) and multivariate (MV, all variables entered) analysis utilized Cox proportional hazards modeling. Associations with ICRR were identified using logistic regression modelling. P values of > 0.05 were considered significant. Results: Corticosteroids treatment at baseline was the only factor associated with ICRR on either UV (p = 0.021) or MV (p = 0.048) analysis. Pts on corticosteroids (n = 13) had an ICRR of 39% versus 63% for patients not on steroids (n = 63) (OR 0.323, 95% CI 0.105-0.996 on MV). Pts receiving corticosteroids also had shorter PFS [median 4.3 versus 6.2 months (mos), HR 1.931, CI 1.061-3.513, P = 0.031 on MV]. BRAF V600E genotype (n = 104, median 5.9 months) was associated with improved PFS versus BRAF V600 other (n = 21, median 4.2 months, HR 0.565, CI 0.321-0.996, P =0.0483 in MV). On UV analysis, OS was associated with elevated LDH (median 10.6 versus 12.7 months, HR 1.595, CI 1.045- 2.434, P = 0.0305), baseline treatment with corticosteroids (median 9.3 versus 13.5 months, HR 1.642, CI 1.037- 2.598, P = 0.0343), and ECOG PS status of 0 (median 18.9 versus 8.2 months, HR 0.435, CI 0.285-0.664). Only ECOG status remained significant on MV of OS (HR 0.441, CI 0.249-0.779, P = 0.0048). Conclusions: Baseline corticosteroid treatment was independently associated with inferior clinical outcomes in MBM pts treated with D+T in COMBI-MB, similar to previous associations reported in immunotherapy trials for pts with MBMs. Likewise, superior outcomes in COMBI-MB patients with BRAFV600E genotype and ECOG status of zero builds on similar observations in COMB-D/V trials of extracranial disease only. The use of corticosteroids to treat symptoms is a major obstacle for the efficacy of systemic treatments for MBMs. Clinical trial information: NCT02039947.