Safety and Antitumor Effects of CD19-Specific Autologous Chimeric Antigen Receptor-Modified T (CAR-T) Cells Expressing the Inducible Caspase 9 Safety Switch (iC9-CAR19 T Cells) in Adult Acute Lymphoblastic Leukemia (ALL)

Introduction: CAR-T cells targeting the CD19 antigen are approved to treat children and young adults with relapsed and refractory B-cell ALL, in whom response rates are >80%. Acute toxicities, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) complicate CAR-T therapy in most patients. Though most patients with CRS or ICANS experience toxicities that can be managed with supportive care including corticosteroids and tocilizumab, 46% experience grade 3-4 CRS and 13% experience grade 3 ICANS (Maude SL et al N Engl J Med 2018; 37:439). Thus, novel approaches to address the management of high grade toxicities are needed. Donor T lymphocytes engineered to express human caspase 9 fused to a modified human FK-binding protein that induces caspase-dependent apoptosis when exposed to the dimerizing drug rimiducid (Di Stasi A et al. N Engl J Med 2011; 365:1673). We hypothesized that the inducible caspase 9 safety switch (iC9) coupled with CAR19 could mitigate severe CRS or ICANS in patients treated with CAR-T cells. We initiated a phase I trial to test the safety and efficacy of autologous T lymphocytes, genetically modified to express both iC9 and CAR19 administered to patients with relapsed and refractory B-ALL.

Methods: Subjects with B-cell ALL in 2^nd or greater bone marrow (BM) relapse, relapse >100 days after allogeneic stem cell transplant, disease refractory to ≥2 induction therapies, or with measurable residual disease (MRD) persistence/recurrence were enrolled in a phase I dose escalation trial. Autologous T-lymphocytes were collected, and CAR-T cell products generated by gene modification with a γ-retroviral vector encoding for iC9, ΔNGFR (for selection and tracking purposes) and CAR.CD19 (encoding 4-1BB) genes (Diaconu I et al Mol Ther 2017; 25:580). Subjects underwent lymphodepletion with fludarabine and cyclophosphamide and CAR-T cells were subsequently infused at one of two dose levels (DL1: 5 x 10⁵ CAR-T cells/kg; DL2: 1 x 10⁶ CAR-T cells/kg). Toxicities were graded by CTCAE v5 or ASBMT consensus grading for CRS and ICANS. Dose limiting toxicities (DLT) were grade 3-4 CRS or ICANS lasting >7 days despite standard of care intervention or grade 3 or higher autoimmune or non-CRS/ICANS organ toxicity. CAR-T cell expansion in peripheral blood (PB) was determined by flow cytometry (FC) and Q-PCR. Leukemia response was determined by NCCN criteria at 4 and 8 weeks after CAR-T infusion.

Results: Nine products from 9 consecutive patients have been successfully manufactured in a median of 14 days (range 13-22), with transduction efficiency of 83 ± 6% after specific ΔNGFR selection. Six subjects have been enrolled to date, three in each cohort. Median age of subjects was 32 years (range 21-41). Median number of prior therapies was 3 (range 2-5). One subject had Philadelphia chromosome positive ALL. Maximum grade CRS was 2 in two subjects and 1 in three subjects. Median duration of any grade CRS was 4 days (range 2-7). One subject in DL1 experienced grade 1 ICANS for 2 days. All CRS/ICANS resolved with standard of care supportive measures, and no subject received rimiducid. No subject experienced DLT. CAR-T cells were found to be increased by PB Q-PCR, peaking at week 2 post infusion (7.9 x 10⁴ ± 3.4 x 10⁴ copies/μg of DNA). This trend paralleled the detection of CAR-T cells by FC. At 4 weeks post infusion, CAR-T cell signals were also detectable in BM samples, and BM B cells comprised <0.01% of BM cellularity in 5/6 subjects. All six subjects experienced response, with best response of CR (n=2), CRi (n=3) and morphologic leukemia-free state (n=1). At time of best response, all subjects were MRD negative by FC and 2 of 3 were MRD negative by molecular MRD assays.

Conclusions: iC9-CAR19 cells can be safely administered to adult patients with relapsed and refractory B-ALL. The use of iC9 as a safety switch has not been required due to the absence of patients with high-grade CRS/ICANS . Preliminary antileukemic activity was observed. The recommended phase 2 cell dose, 1 x 10⁶ transduced cells/kg is being tested in an expansion cohort. A dose finding study to determine the effects of rimiducid on CRS/ICANS grade, CAR-T cell persistence and cytokine levels is being conducted among expansion cohort patients who experience CRS/ICANS not responding to standard supportive care.

Introduction: CAR-T cells targeting the CD19 antigen are approved to treat children and young adults with relapsed and refractory B-cell ALL, in whom response rates are >80%. Acute toxicities, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) complicate CAR-T therapy in most patients. Though most patients with CRS or ICANS experience toxicities that can be managed with supportive care including corticosteroids and tocilizumab, 46% experience grade 3-4 CRS and 13% experience grade 3 ICANS (Maude SL et al N Engl J Med 2018; 37:439). Thus, novel approaches to address the management of high grade toxicities are needed. Donor T lymphocytes engineered to express human caspase 9 fused to a modified human FK-binding protein that induces caspase-dependent apoptosis when exposed to the dimerizing drug rimiducid (Di Stasi A et al. N Engl J Med 2011; 365:1673). We hypothesized that the inducible caspase 9 safety switch (iC9) coupled with CAR19 could mitigate severe CRS or ICANS in patients treated with CAR-T cells. We initiated a phase I trial to test the safety and efficacy of autologous T lymphocytes, genetically modified to express both iC9 and CAR19 administered to patients with relapsed and refractory B-ALL.

Methods: Subjects with B-cell ALL in 2^nd or greater bone marrow (BM) relapse, relapse >100 days after allogeneic stem cell transplant, disease refractory to ≥2 induction therapies, or with measurable residual disease (MRD) persistence/recurrence were enrolled in a phase I dose escalation trial. Autologous T-lymphocytes were collected, and CAR-T cell products generated by gene modification with a γ-retroviral vector encoding for iC9, ΔNGFR (for selection and tracking purposes) and CAR.CD19 (encoding 4-1BB) genes (Diaconu I et al Mol Ther 2017; 25:580). Subjects underwent lymphodepletion with fludarabine and cyclophosphamide and CAR-T cells were subsequently infused at one of two dose levels (DL1: 5 x 10⁵ CAR-T cells/kg; DL2: 1 x 10⁶ CAR-T cells/kg). Toxicities were graded by CTCAE v5 or ASBMT consensus grading for CRS and ICANS. Dose limiting toxicities (DLT) were grade 3-4 CRS or ICANS lasting >7 days despite standard of care intervention or grade 3 or higher autoimmune or non-CRS/ICANS organ toxicity. CAR-T cell expansion in peripheral blood (PB) was determined by flow cytometry (FC) and Q-PCR. Leukemia response was determined by NCCN criteria at 4 and 8 weeks after CAR-T infusion.

Results: Nine products from 9 consecutive patients have been successfully manufactured in a median of 14 days (range 13-22), with transduction efficiency of 83 ± 6% after specific ΔNGFR selection. Six subjects have been enrolled to date, three in each cohort. Median age of subjects was 32 years (range 21-41). Median number of prior therapies was 3 (range 2-5). One subject had Philadelphia chromosome positive ALL. Maximum grade CRS was 2 in two subjects and 1 in three subjects. Median duration of any grade CRS was 4 days (range 2-7). One subject in DL1 experienced grade 1 ICANS for 2 days. All CRS/ICANS resolved with standard of care supportive measures, and no subject received rimiducid. No subject experienced DLT. CAR-T cells were found to be increased by PB Q-PCR, peaking at week 2 post infusion (7.9 x 10⁴ ± 3.4 x 10⁴ copies/μg of DNA). This trend paralleled the detection of CAR-T cells by FC. At 4 weeks post infusion, CAR-T cell signals were also detectable in BM samples, and BM B cells comprised <0.01% of BM cellularity in 5/6 subjects. All six subjects experienced response, with best response of CR (n=2), CRi (n=3) and morphologic leukemia-free state (n=1). At time of best response, all subjects were MRD negative by FC and 2 of 3 were MRD negative by molecular MRD assays.

Conclusions: iC9-CAR19 cells can be safely administered to adult patients with relapsed and refractory B-ALL. The use of iC9 as a safety switch has not been required due to the absence of patients with high-grade CRS/ICANS . Preliminary antileukemic activity was observed. The recommended phase 2 cell dose, 1 x 10⁶ transduced cells/kg is being tested in an expansion cohort. A dose finding study to determine the effects of rimiducid on CRS/ICANS grade, CAR-T cell persistence and cytokine levels is being conducted among expansion cohort patients who experience CRS/ICANS not responding to standard supportive care.