Initial Experience with CD19/CD22 BiCistronic CAR T-Cells in Children and Young Adults with Recurrent or Refractory B-Cell Malignancies

Introduction: Single-antigen targeted CAR T-cells have great efficacy in B-cell acute lymphoblastic leukemia (B-ALL), but immune escape is a significant cause of relapse. Dual antigen targeting represents a potential mechanism to improve durable responses. While our prior bivalent CD19.CD22.BBζ CAR T-cells demonstrated good tolerability and anti-leukemia effect, expansion and persistence were limited. Additionally, suboptimal CD22 targeting limited dual antigen targeting. Our BiCistronic CD19.28ζ/CD22.BBζ (BiCis19/22) construct showed improved preclinical dual antigen targeting and cytokine profiling compared to the prior construct, prompting clinical translation. Herein, we report initial results from the first 6 patients treated in the phase I portion of a single-center trial of BiCis19/22 CAR T-cells for relapsed/refractory B-ALL (NCT05442515).

Methods: The primary objective was safety of autologous BiCis19/22 CAR T-cells in children, adolescents, and young adults with B-ALL or non-Hodgkin (NHL). Manufacturing was on the CliniMACS Prodigy. Initial trial design stratified patients by CAR T-cell exposure (A1: CAR naïve, B1: prior CAR) but was later amended to stratify by disease burden based on toxicity (A1b: bone marrow [BM] disease burden <25% without extramedullary disease [EMD], B1b: BM disease burden >25% or EMD). Consensus criteria were used to define cytokine release syndrome (CRS), immune effector cell associated neurotoxicity syndrome (ICANS), and immune effector cell associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS). Secondary objectives included anti-leukemia efficacy. Exploratory objectives included CAR T-cell expansion, persistence, and cytokine profiling. Eligible patients were 3-39y with relapsed/refractory CD19+ CD22+ B-ALL/NHL who were ineligible for, could not access, or recurred after stem cell transplant (SCT) or commercial CD19 CAR. Hyperleukocytosis, CNS3, or active CNS lymphoma were exclusionary. Patients received standard fludarabine/cyclophosphamide lymphodepletion. Patients received either 1x10⁶ transduced CAR T-cells/kg (1-4, dose level 1 [DL1]) or 3x10⁵ CAR T-cells/kg (5-6, DL-1). B-ALL disease assessment used standard BM morphology with minimal residual disease (MRD) detected by flow cytometry (FC) at a limit of < 0.01% of mononuclear cells (MNC). Data cutoff of July 31, 2024.

Results: The 6 patients infused (83.3% male, median age 25.5y [20.5-29.2y]) were heavily pretreated, receiving a median 5 prior lines of therapy (3-6, 66.6% post-SCT). All had successful BiCis19/22 CAR T-cell manufacturing and infusion.

Patients 1-4 were treated at DL1 with 1, 2 and 4 being CAR naïve (cohort A1). Patient 3 had prior CAR (cohort B1). Patients 1 & 2 had low disease burden (BM disease <6% of MNC by FC) while 3 & 4 had >90% BM disease by FC despite bridging therapy. Three developed grade (Gr) ≥3 CRS, 2 had Gr 4 ICANS including 1 case of myelitis, and 1 had Gr 2 IEC-HS. Due to the dose-limiting toxicity of severe ICANS in high-disease burden patients at DL1, the trial was amended to dose-reduce to DL-1 and stratify patients by disease burden. Two high disease burden patients were then treated at DL-1 (cohort B1b). Both developed CRS (Gr 1-2) with 1 Gr 1 IEC-HS that fully resolved. No ICANS was observed.

All 6 patients achieved MRD negative complete remission (CR) at day 28. Three proceeded to consolidative SCT, 2 remain in MRD negative CR without interval therapy >15 months post-infusion, and 1 is in early follow-up (57 days).

Correlative data from the 4 patients at DL1 demonstrate CAR T-cell expansion peaked at D+13 with persistence of CAR T-cells detected up to 6 months by FC (20d-6mo). Marked elevations were seen in IL-6, GM-CSF, and IL-2 compared to the earlier bivalent CAR (p=0.002, p=0.013, p=0.007, respectively).

Conclusions: This first experience in B-ALL with BiCis19/22 CAR T-cells demonstrates substantial efficacy (100% CR). This construct exhibits features of enhanced dual antigen targeting by virtue of its toxicity profile, including severe ICANS (not seen with CD22 CAR) and IEC-HS (less frequent with CD19 CAR), and more robust cytokine production that may be mitigated by dose reduction. Importantly, 2 patients with highly refractory disease remain in remission > 1 year without interval therapy, showing promise of dual antigen targeting. This CAR is also being tested in lymphoma and is reported separately (NCT05098613).