Abstract A005: Preliminary results from STRIvE-01, Arm B: A phase I study of EGFR806 x CD19 CAR T cell immunotherapy for recurrent/refractory solid tumors in children and young adults

Abstract

The epidermal growth factor receptor (EGFR) is associated with aggressive neoplastic disease, chemotherapy resistance, and increased metastatic potential in solid neoplasms. Approximately 20-30% of pediatric solid tumors (ST) express EGFR, enriched for soft tissue sarcomas (STS). The unique EGFR monoclonal antibody (mAb) 806 selectively binds to an epitope that is conformationally hidden when EGFR is tethered but revealed when tethering is perturbed as occurs with EGFR overexpression or extra-cellular domain missense mutations. The use of bispecific CARs has been identified as an approach to enhance expansion and overcome tumor escape, such as engrafting bispecific tumor antigen-directed and CD19 CAR T cells to permit the expansion and retention of function of large numbers of CAR T cells outside the solid tumor that can then infiltrate solid tumors in large numbers. Children and young adults (CYA) with EGFR-expressing recurrent/refractory (R/R) ST were enrolled on a Phase 1 trial to examine the safety and feasibility of administering autologous EGFR806-directed (Arm A) or dual transduced EGFR806- and CD19-directed CAR T cells (Arm B) following lymphodepleting chemotherapy. The maximum tolerated dose (MTD) was determined based upon observed toxicity through day 28 from CAR-T infusion and using a 3+3 statistical design. Sixteen evaluable subjects (age range 11-28 yo) were enrolled on Arm B and received dose level (DL) 0 (0.5 x 10^6 CAR-T/kg, n=7), DL1 (1 x 10^6 CAR-T cells/kg, n=5) or DL2 (2.5 x 10^6 CAR-T cells/kg, n=4). CAR T cells were manufactured successfully for all subjects. The most common toxicities were cytokine release syndrome (n=4) and tumor-related pain (n=3). Dose limiting toxicity (DLT) included elevated transaminases grade 4 (n=1) and ICANS grade 3 (n=1) on DL0B and elevated bilirubin grade 3 (n=2), elevated transaminases grade 4 (n=1), hepatic failure grade 4 (n=1) and IEC-HS grade 4 (n=1) on DL2B. No DLT was observed on DL1B, which was determined to be the MTD. Median expansion was 4.36% of circulating lymphocytes for EGFR+CD19- CAR T cells and 2.70% for EGFR+CD19+ CAR T cells, vs a median of 0.3% on Arm A. Median persistence was 43 days (range 14-182) vs 28 days (range 0-90) on Arm A. There was not a significant differece in peak engraftment between dose levels but a trend toward increased toxicity with higher peak engraftment. The best response was a partial response in one synovial sarcoma patient treated on DL 2B. The most common response was stable disease (9/16). EGFR806 x CD19 CAR-T cells have an acceptable toxicity profile at DL1B (1 x 10^6 CAR-T cells/kg) in CYA with R/R ST and demonstrate anti-tumor activity in some patients. Engraftment and persistence are enhanced compared to EGFR806-directed CAR T cells alone.