The first-in-human ENCIT01 trial comparing second- versus third-generation L1CAM-specific CAR T cells in patients with primary refractory or relapsed neuroblastoma.

Abstract

Outcomes for children with relapsed and refractory neuroblastoma are dismal. ENCIT-01 (NCT02311621) was a first-in-human clinical trial for patients with relapsed and refractory neuroblastoma using chimeric antigen receptor (CAR) T cells targeting L1CAM, an adhesion molecule that is overexpressed in neuroblastoma with limited normal tissue expression. This trial evaluated three different CAR constructs: a short spacer second-generation 4-1BB CAR (2GS, Arm A), a short spacer third-generation 4-1BB+CD28 CAR (3GS, Arm B) and a long spacer second-generation 4-1BB CAR (2GL, Arm C). Thirty-six patients were enrolled and 22 were treated (Arm A/2GS n=11, Arm B/3GS n=8 and Arm C/2GL n=3). Thirty-four of 36 patients had a CAR T cell product successfully manufactured. Cytokine release syndrome, skin rash and hyponatremia were common ≥ Grade 2 toxicities. Hyponatremia was dose-limiting in 3 patients (DL5 Arm A/2GS, DL3 Arm B/3GS and DL2 Arm C/2GL). Patterns of toxicity appeared at lower dose levels on Arm B and Arm C compared to Arm A, suggesting differential potency of the third generation and long spacer products. No objective responses were seen. Correlative analyses demonstrated CAR T cell presence in tumor and skin, with evidence of macrophage tumor infiltration. While feasible to manufacture in a heavily pretreated population, L1CAM may not be an appropriate target in neuroblastoma. Additional engineering strategies may be needed to prevent toxicity and provide durable anti-tumor effects.