CTIM-27. DOC1021 CELL-BASED VACCINATION AS ADJUVANT THERAPY FOR GLIOBLASTOMA: PHASE I CLINICAL TRIAL ANALYSIS

Abstract Glioblastoma is an aggressive tumor for which median survival remains 14-18 months despite aggressive standard of care. Cellular immunotherapy approaches have recently shown promise. Homologous antigenic loading is an ex-vivo technique that leverages p38MAPK and mTORC1 signaling cascades to initiate powerful cDC1-like skewing in monocyte-derived dendritic cells, leading to downstream induction of tissue-infiltrating, cytolytic effector memory T-cells. This open-label phase I clinical trial evaluated the autologous dendritic cell vaccine DOC1021 prepared through homologous antigenic loading and administered bilaterally near the deep cervical lymph nodes. Three courses of vaccine were administered every 2 weeks after completion of chemoradiation, adjuvanted concurrently with six weeks of weekly type I interferon. Four dose levels from 3.5x106 to 3.6x107 total vaccine cells were tested. Tumor progression prior to vaccination was not an exclusion criterion. Sixteen newly diagnosed IDH-WT patients completed treatment, median age 63 years and 94% MGMTp unmethylated. The most common related AEs were grade 1 injection-site reactions and grade 1-3 fatigue and urticaria. No DLTs were observed. Immunohistochemistry of tumors derived from early post-vaccination second resections showed enhanced CD8+ T-cell infiltration and pathologic findings consistent with residual rather than relapsed GBM in 2/3 patients. Analysis of post-vaccination PBMC in dose level cohorts 2-4 indicated expansion of both CD4+ (13/13) and CD8+ (11/13) central memory T-cell compartments (p<0.002 and p<0.05, respectively) as well as expansion of CD8+CD127+ T-cells (12/13; p<0.001). Among three patients for whom tissue was obtained pre- and post-vaccination, clinical outcomes correlated with Visium spatial transcriptomic analysis. Median survival for the MGMTp unmethylated cohort (n=15) has not yet been reached but is statistically greater (p<0.05) than that of matched historic controls. The results indicate that DOC1021 is safe, can be effectively integrated within existing standards of care and is potentially efficacious in a challenging patient population.