CD19-Targeted 19-28z CAR Modified Autologous T Cells Induce High Rates of Complete Remission and Durable Responses in Adult Patients with Relapsed, Refractory B-Cell ALL

Background: Relapsed adult acute lymphoblastic leukemia (ALL) is associated with high reinduction mortality, chemotherapy resistance, and dismal prognosis with a median overall survival (OS) < 6 months and 5-year OS ≤10%. We have previously reported a high anti-tumor activity of autologous T cells genetically modified to express 19-28z chimeric antigen receptor (19-28z CAR) targeting CD19 in adult patients with CLL and ALL (Brentjens R et al. Blood 2011; Davila M et al. Sci Transl Med2014). Herein, for the first time, we further report the long-term outcome of our phase I clinical trial in adults with relapsed/refractory (R/R) ALL (NCT01044069) with analysis on potential predictive markers of response and neurological toxicities.

Patients and Methods: Adult patients with R/R B-ALL were enrolled. Eligible patients underwent leukapheresis, and T cells were transduced with a retrovirus encoding a CAR construct composed of anti-CD19 scFV linked to CD28 and CD3ζ signaling domains (19-28z). All patients received lymphodepleting chemotherapy followed 2 days later by 1x10⁶ – 3x10⁶19-28z CAR T cells/kg. The primary objective of the study was to evaluate the safety and anti-tumor activity of 19-28z CAR T cells in ALL. Post-treatment minimal residual disease (MRD) was assessed at day 14-28 by multiparameter flow cytometry and deep sequencing in the bone marrow (BM) samples (Adaptive Biotech Corp.)

Results: 24 patients have been treated. The median age was 56 years (range, 23-74). 6 patients (25%) had Ph+ B-ALL (T315I mutation in 2 patients), 6 patients (25%) had prior allogeneic hematopoietic stem cell transplant (allo-HSCT), and 11 patients (46%) had 3 or more prior lines of ALL therapy before receiving the 19-28z CAR T cell therapy.

Of the 24 patients, 22 patients were evaluable for response. At the time of 19-28z CAR T cell infusion, 12 of 22 patients had morphologic disease (6 to 97% blasts in the BM) and the remaining 10 patients had MRD. Twenty out of 22 patients (91%) were in complete remission (CR) after 19-28z CAR T-cell infusion, and 18 of these 20 patients (90%) achieved an MRD-negative CR. Ten of the 13 transplant eligible patients (77%) successfully underwent allo-HSCT following the 19-28z CAR T cell therapy. As of July 1, 2014, the median follow-up was 7.4 months (range 1-34), with 13 patients having at least 6 months of follow-up. Responses appear durable with 6 patients remaining disease-free beyond 1 year (range 12.6 – 34 months). Median overall survial (OS) is 9 months. 5 patients relapsed during the follow-up, including 1 patient with CD19 negative relapse. Three of the relapsed patients were treated again with the 19-28z CAR T cells, and two patients achieved a second CR.

Comparing responders to non-responders, no association was observed between response and age (<60 vs. ≥60), prior allo-HSCT, number of prior therapies, or pre-treatment blast percentage.

While none of the 10 patients with MRD at the time of T cell infusion developed cytokine release syndrome (CRS), 9 of 13 patients with morphologic disease at the time of the T cell infusion developed CRS with or without neurological symptoms that required intervention with an IL-6R antagonist or corticosteroid. A detailed analysis of serum cytokines demonstrated a consistent peak of IL-6 (22.2 to 553-fold increase) immediately prior to the development of neurological toxicities. Based on these data, we have developed a multi-disciplinary CRS management algorithm for patients at high risk in order to reduce the severity of CRS and improve safety of the 19-28z CAR T cell therapy.

Conclusions: While longer follow-up is needed to confirm the durability of the observed responses, the potent induction of MRD-negative responses and successful long-term outcomes, including subsequent allo-HSCT without apparent additional post-transplant toxicities, strongly support the use of 19-28z CAR T cells in adult patients with B-ALL. A temporal relationship between serum IL-6 levels and neurological toxicities indicates that early intervention with IL-6 directed therapy may be more effective in ameliorating neurological toxicities in patients with morphologic disease at the time of T-cell infusion. These findings will need to be evaluated systematically and confirmed in a larger phase 2 trial.