Patients in this study were enrolled in an open-label phase 1 dose-escalation study of CD19-CAR T cells in children and young adults with ALL or non-Hodgkin lymphoma, NCT01593696, or an open-label phase 1 dose-escalation study of GD2-CAR T cell in children and young adults with GD2 expressing osteosarcoma or neuroblastoma, NCT02107963. Clinical results of the first 21 of the patients receiving CD19-CAR T cell therapy have previously been reported . The results of CD19-CAR T cell manufacturing from the first 43 patients and GD2-CAR T cells from the first 11 patients have previous been reported . This study reports the results of manufacturing CD19-CAR T cells from 8 additional patients and GD2-CAR T cells from 5 additional patients. Among the 8 patients treated with CD19-CAR T cells 7 had ALL and one had diffuse large B cell lymphoma (DLBCL). All 5 patients treated with GD2-CAR T cells all had osteosarcoma. All 8 patients receiving CD19-CAR T cells were given a dose of 1 × 106 cells/kg and 2 patients receiving GD2-CAR T cells were given a dose of 3 × 106 CAR T cells/kg and 3 patients were given a dose of 1 × 107 CAR T cells/kg. All subjects were enrolled in protocols approved by the National Cancer Institute (NCI) Institutional Review Board and inform consent was obtained.
We also compared the results of manufacturing CD19- and GD2-CAR T cells in this study with the previously reported results of manufacturing CD19- and GD2-CAR T cells using PBMC concentrates enriched for T cells by selection with anti-CD3/CD28 beads with and without plastic adherence .
Manufacturing CAR T cells
Peripheral blood mononuclear cells concentrates were collected using a blood cell separator (Cobe Spectra, Terumo BCT, Lakewood, CO, USA) and 10–15 L of blood was processed. CD19-CAR T cells were manufactured from PBMC concentrates using a modification of the method we previously described [11, 13]. When PBMCs were enriched for lymphocytes by anti-CD3/CD28 bead selection, on day 0, a fresh or cryopreserved PBMC concentrate containing 600 × 106 CD3+ cells were enriched for CD3+ cells using anti-CD3/CD28 antibodies bound to paramagnetic beads (Dynabeads ClinExVivo CD3/CD28, Invitrogen, Camarillo, CA) at a ratio of 3:1 (beads:cells). The cells and beads were co-incubated for 2 h at room temperature and CD3+ cell enrichment was performed using Dynal ClinExVIVO MPC magnet (Invitrogen, Camarillo, CA). A total of 100 × 106 cells in the CD3+ fraction were resuspended at a concentration of 1 × 106 cells/mL in PermaLife bags (OriGen Biomedical, Austin, TX) at 37 °C in 5% CO2 in AIM V medium (Gibco, Grand Island, NY), supplemented with 5% heat-inactivated human AB Serum (Valley Biomedical, Winchester, VA), 1% Gluta-Max (Gibco, Grand Island, NY), 40 IU/mL IL-2 (Novartis Vaccines and Diagnostics, Inc. Emeryville, CA).
For manufacturing CD19-CAR T cells, the lymphocyte enriched cells were transduced twice with clinical grade MSGV-FMC63-28Z recombinant retroviral vector supernatant, once on day 2 and once on day 3, in retronectin-coated bags. The cells were maintained in culture for 7–9 days. The cell concentration was maintained at 0.4 × 106 cells/mL by adding fresh media every other day. On the day of harvest the anti-CD3/CD28 paramagnetic beads were removed using the Dynal ClinExVIVO MPC magnet (Invitrogen, Camarillo, CA), washed, concentrated and quality control assessment was performed.
For manufacturing GD2-CAR T cells, the same lymphocyte enrichment processes was used and a similar process was used to transduce and expand the cells. Cells were transduced once on day 2 with anti-GD2.28.z.OX40.ICD9 retroviral vector supernatant and were harvested after 10 or 11 days in culture.
When PBMC concentrates were enriched for lymphocytes by anti-CD3/CD28 bead selection plus plastic adherence the PBMC concentrates were incubated with anti-CD3/CD28 beads for 2 h in T flasks rather than in bags. At the end of the 2 h incubation period the non-adherent cells were collected and the cells were processed as described above.
When the PBMC concentrates were enriched for lymphocytes by elutriation they were subject to elutriation using a semi-automatic counter-flow elutriation instrument (Elutra Cell Separation System, version 1.1, Terumo BCT) using a user defined profile which collects cells in 5 fractions. The chamber rotation speed was maintained at 2400 RPMs for fractions 1 through 4 and the media flow rate was maintained at 60 mL/min for fraction 1, 120 mL/min for fraction 2, 122 mL/min for fraction 3 and 124 mL/min for fraction 4. Fraction 5 was the cells remaining in the chamber and they were collected with the rotor turned off. Lymphocytes were found in fractions 1 and 2 and monocytes and granulocytes were in fraction 5. Fraction 1 and 2 also were enriched with platelets and RBCs .
The lymphocyte fraction was depleted of RBCs by lysis. The cells were pelleted by centrifugation, incubated for 7–10 min with ACK Lysing Buffer (Lonza, Walkersville, MD) and then washed and resuspended in 0.9% saline (B.Braun Medical Inc., Irvine, CA) with 0.3% trisodium citrate (TriCitrasol anticoagulant sodium citrate concentrate 46.7%, Citra Labs, Braintree, MA).
Cell counts and flow cytometry
Blood counts were measured using automated hematology analyzer (Cell-Dyn 3700). Flow cytometry was performed with a FACSCanto II (BeckinDickinson) using CD3, CD4, CD8, CD14, CD15, CD19, CD45 and CD56 antibodies (BD Biosciences, San Jose, CA). The expression of CD19-CAR and GD2-CAR was assessed by flow cytometry with anti-idiotype antibodies.
The values shown are mean ± 1 standard deviation unless otherwise indicated. Groups were compared using t-tests (Microsoft Excel, Microsoft Inc., Redmond, WA).