CLINICAL STUDY Survival outcomes and safety of carmustine wafers in the treatment of high-grade gliomas: a meta-analysis

Abstract Carmustine wafers (CW; Gliadel wafers) are approved to treat newly-diagnosed high-grade glioma (HGG) and recurrent glioblastoma. Widespread use has been limited for several reasons, including concern that their use may preclude enrollment in subsequent clinical trials due to uncertainty about confounding of results and potential toxicities. This meta-analysis estimated survival following treatment with CW for HGG. A literature search identified relevant studies. Overall survival (OS), median survival, and adverse events (AEs) were summarized. Analysis of variance evaluated effects of treatment (CW vs non-CW) and diagnosis (new vs recurrent) on median survival. The analysis included 62 publications, which reported data for 60 studies (CW: n = 3,162; non-CW: n = 1,736). For newlydiagnosed HGG, 1-year OS was 67 % with CW and 48 % without; 2-year OS was 26 and 15 %, respectively; median survival was 16.4 ± 21.6 months and 13.1 ± 29.9 months, respectively. For recurrent HGG, 1-year OS was 37 % with CW and 34 % without; 2-year OS was 15 and 12 %, respectively; median survival was 9.7 ± 20.9 months and 8.6 ± 22.6 months, respectively. Effects of treatment (longer median survival with CW than without; P = 0.043) and diagnosis (longer median survival for newly-diagnosed HGG than recurrent; P \ 0.001) on median survival were significant, with no significant treatment-by-diagnosis interaction (P = 0.620). The most common AE associated with wafer removal was surgical site infection (SSI); the most common AEs for repeat surgery were mass effect, SSI, hydrocephalus, cysts in resection cavity, acute hematoma, wound healing complications, and brain necrosis. These data may be useful in the context of utilizing CW in HGG management, and in designing future clinical trials to allow CW-treated patients to participate in experimental protocols.