First-in-human trial of E-SYNC T-cells, an autologous anti-EGFRvIII SynNotch receptor-induced anti-EphA2/IL13Ra2 CAR T-cells, in adult patients with EGFRvIII+ glioblastoma

Background: Developing safe and effective CAR-T therapy for glioblastoma (GBM) has been hindered by antigen heterogeneity, on-target off-tumor toxicity, and CAR-T cell exhaustion. We have developed a novel synthetic Notch ( syn Notch) receptor-based “prime-and-kill” dual antigen recognition T cell circuit. E GFRvIII, which is expressed by a subset of GBM cells, primes the T-cells to express a C AR that recognizes IL-13Ralpha2 and EphA2, which are broadly expressed in GBM, thereby eradicating GBM cells expressing either EphA2 or IL-13Ralpha2 (E-SYNC T-cells). We developed a Phase I clinical trial to evaluate the safety of intravenously infused E-SYNC T-cells. Methods: The study has 2 sequential patient cohorts. Cohort 1 is a 2-level dose-escalation cohort, targeting newly diagnosed patients with EGFRvIII-mutant, MGMT-unmethylated GBM. The E-SYNC cells are manufactured by ex vivo transduction of autologous T-cells with a lentiviral E-SYNC vector, and then intravenously infused after lymphodepleting chemotherapy. Cohort 2 will be a tissue analysis cohort. At the time of recurrence, participants will receive an infusion of E-SYNC T-cells at the maximum tolerated/recommended dose from Cohort 1 prior to a planned clinical surgical resection. The primary objective is safety; secondary objectives are feasibility and (for Cohort 2 only) to determine the priming status of E-SYNC T-cells in the GBM tissues and peripheral blood. We are currently enrolling patients in cohort 1. Results: Autologous E-SYNC cells have been successfully manufactured for 9 patients. To date, we have treated six patients at dose level 1 (DL1: 5x10 7 CAR + cells) and two patients at dose level 2 (DL2: 1.5x10 8 CAR + cells), with no serious adverse events or dose-limiting toxicities. Two of six patients at DL1 remain progression-free for 20 months and 15 months since the initial resection. Digital PCR-based evaluation of the peripheral blood demonstrated the post-infusion persistence of E-SYNC T-cells for at least 40 weeks. Serum analyses show post-infusion increases in CCL2, CCL5, CXCL9, and CCL22. In two patients who recurred within two months following the infusion, the recurrent tumor tissues showed the loss of EGFRvIII, despite the infiltration of E-SYNC cells. Conclusions: We have safely treated eight patients with EGFRvIII+ GBM with autologous E-SYNC cells, with no significant toxicity thus far. Considering the poor prognosis of MGMT-unmethylated GBM patients, the progression-free survival in two patients is promising. To overcome the loss of EGFRvIII as the priming signal, we are also developing a second-generation synNotch-CAR study using Brevican, a brain-specific extracellular matrix protein, to induce the CAR expression (Simic, Watchmaker et al., Science, 386, 2024).