Manufacturing chimeric antigen receptor T cells from cryopreserved peripheral blood cells: time for a collect-and-freeze model?

BACKGROUND AIMS

Chimeric antigen receptor (CAR)-modified T-cell therapy has revolutionized outcomes for patients with relapsed/refractory B-cell malignancies. Despite the exciting results, several clinical and logistical challenges limit its wide applicability. First, the apheresis requirement restricts accessibility to institutions with the resources to collect and process peripheral blood mononuclear cells (PBMCs). Second, even when utilizing an apheresis product, failure to manufacture CAR T cells is a well-established problem in a significant subset. In heavily pre-treated patients, prior chemotherapy may impact T-cell quality and function, limiting the ability to manufacture a potent CAR T-cell product. Isolation and storage of T cells shortly after initial cancer diagnosis or earlier in life while an individual is still healthy are an alternative to using T cells from heavily pre-treated patients. The goal of this study was to determine if a CAR T-cell product could be manufactured from a small volume (50 mL) of healthy donor blood.

METHODS

Collaborators at Cell Vault collected 50 mL of whole peripheral venous blood from three healthy donors. PBMCs were isolated, cryopreserved and shipped to the Medical College of Wisconsin. PBMCs for each individual donor were thawed, and CAR T cells were manufactured using an 8-day process on the CliniMACS Prodigy device with a CD19 lentiviral vector.

RESULTS

Starting doses of enriched T-cell numbers ranged from 4.0 × 10 cells to 4.8 × 10 cells, with a CD4/CD8 purity of 74-79% and an average CD4:CD8 ratio of 1.4. On the day of harvest, total CD3 cells in the culture expanded to 3.6-4.6 × 10 cells, resulting in a 74- to 115-fold expansion, an average CD4:CD8 ratio of 2.9 and a CD3 frequency of greater than 99%. Resulting CD19 CAR expression varied from 19.2% to 48.1%, with corresponding final CD19+ CAR T-cell counts ranging from 7.82 × 10 cells to 2.21 × 10 cells. The final CAR T-cell products were phenotypically activated and non-exhausted and contained a differentiated population consisting of stem cell-like memory T cells.

CONCLUSIONS

Overall, these data demonstrate the ability to successfully generate CAR T-cell products in just 8 days using cryopreserved healthy donor PBMCs isolated from only 50 mL of blood. Notably, numbers of CAR T cells were more than adequate for infusion of an 80-kg patient at dose levels used for products currently approved by the Food and Drug Administration. The authors offer proof of principle that cryopreservation of limited volumes of venous blood with an adequate starting T-cell count allows later successful manufacture of CAR T-cell therapy.