Efficacy of immune checkpoint inhibitors (ICI) in patients (pts) with central nervous system (CNS) metastases (mets) from solid tumors: A systematic review and meta-analysis.

Abstract

2023 Background: Brain metastases, a common complication of solid tumors, are associated with poor outcomes. The role of ICIs in this setting remains unclear. This meta-analysis aims to assess intracranial efficacy of ICI-based systemic treatment in pts with CNS mets from solid tumors. Methods: A systematic literature search of PubMed, Embase, CENTRAL, and conference proceedings (ESMO and ASCO) up to 15-Mar-24 (PROSPERO: CRD42021242755), was conducted to identify single-arm phase II or III, or randomized controlled trials of pts with CNS mets from solid tumors at baseline treated with ICI-based systemic treatment. The primary objective, CNS efficacy, was measured by pooled CNS objective response rate (CNS-ORR) and weighted median CNS progression-free survival (mCNS-PFS). Subgroup analyses evaluated the impact of disease and treatment characteristics. Overall effects were pooled using random-effects models. Results: Out of 1 690 records screened, a total of 1 224 pts enrolled in 32 clinical trials were included. Overall, ICI-based systemic therapy led to a CNS-ORR of 38.0% (95% confidence interval [CI] 31.8-45.5) and a mCNS-PFS of 9.1 months (mos). CNS efficacy was numerically higher in pts with non-small cell lung cancer (NSCLC, n=383, CNS-ORR 45.8% [34.4-60.9]; mCNS-PFS 9.6 mos) and melanoma (n=554, CNS-ORR 37.7% [30.6-46.5]; mCNS-PFS 11.4 mos) vs multi tumors (n=128, CNS-ORR 24.8% [7.7-80.1] and mCNS-PFS 2.8 mos). Efficacy was also greater in first-line therapy (n=553, CNS-ORR 45.2% [36.7-55.7]; mCNS-PFS 8.6 mos) vs second or later lines (n=190, CNS-ORR 14.9% [7.6-29.2]; and mCNS-PFS 2.6 mos). Dual ICI (n=279, CNS-ORR 43.9% [35.5-54.2]; mCNS-PFS 17.8 mos) and ICI plus non-ICI agents (n=432, CNS-ORR 48.7% [39.6-59.9]; mCNS-PFS 7.1 mos) were more effective than single ICI (n=419, CNS-ORR, 20.4% [11.9-34.9]; mCNS-PFS 4.8 mos). Subgroup analyses showed superior CNS outcomes in cerebral mets, treated lesions, asymptomatic pts, and low/no steroid use (table). Conclusions: ICI-based regimens demonstrate CNS efficacy in pts with solid tumors, particularly in pts with NSCLC and melanoma treated with first-line combination therapies. Subgroup analyses by CNS characteristics Category( n of pts for CNS-ORR) CNS-ORR, % (95% CI) mCNS-PFS, mos Type of CNS mets Cerebral (1063) 38.3 (31.9-45.9) 9.6 Leptomeningeal (51) 34.6 (15.9-75.2) 2.4 CNS local therapy Treated (197) 37.6 (26.8-52.8) 8.9 Untreated (307) 43.1 (35.0-53.0) 5.2 Symptoms Asymptomatic/mild (832) 40.4 (34.5-47.3) 10.3 Symptomatic (80) 31.3 (18.3-53.4) 2.5 Concomitant steroids No (391) 30.6 (21.6-43.4) 15.1 Low dose (434) 40.7 (31.5-52.6) 7.6 Any dose steroids (144) 41.6 (31.7-54.7) 3.3