Successful Manufacturing of CAR T-Cells with Small Volume Peripheral Blood from Healthy Donors Using the Clinimacs Prodigy Device

Introduction

In recent years, CAR T-cell therapy has emerged as a potentially curative intent treatment for some patients with relapsed, refractory hematologic malignancies. Despite the exciting results, not all patients are able to receive CAR T-cells due to manufacturing failures. T-cells for CAR products are typically autologous and isolated from heavily pre-treated patients, which might account for some of the manufacturing failures and suboptimal clinical efficacy. T-cells collected either early into cancer diagnosis or prior to diagnosis may improve CAR T-cell expansion and limit manufacturing failure. We evaluated the feasibility of generating a CAR T-cell product manufactured from 50 ml of healthy donor blood.

Methods

Collaborators at Cell Vault collected 50 ml of whole blood from 3 healthy donors, isolated peripheral blood mononuclear cells (PBMCs), and cryopreserved the cells in cryovials at 5e6/vial (1.05-1.35e8 total cells). The vials were shipped to the Medical College of Wisconsin and stored frozen in liquid nitrogen until use. All PBMC vials for a given donor were thawed and pooled. Thawed PBMCs (0.93-1.17e8 cells) were loaded onto a CliniMACS Prodigy device, CD4 and CD8 T cells enriched by immunomagnetic sorting, and T cells placed in the culture chamber with IL-7, IL-15 and TransAct reagent to induce proliferation. On the second day of manufacturing, T cells were transduced with a lentiviral CAR vector encoding anti-CD19, 4-1BB and CD3z. Final CAR T-cell products for these pre-clinical studies were harvested on day 8 of manufacture.

Results

Starting enriched T-cell numbers from the 3 healthy donors ranged from 4.0-4.8e7 cells, the cells were 74-79% CD4/8+, and the average CD4/CD8 ratio was 1.4. On the day of CAR T harvest (day 8), total cells in the chamber had expanded to 3.6-4.6e9 cells (74-115 fold expansion), the cells were >99% CD3+, and the average CD4/CD8 ratio was 2.9 (Table 1). Final cell numbers were similar to what previously published CAR T manufacturing runs on the CliniMACS Prodigy (Zhu et al., Cytotherapy, 2018), that started with 1x108 enriched T-cells obtained from apheresed mononuclear cells. Cell surface CD19 CAR expression on the final cell products varied from 19.2-48.1%. While more than 50% of the starting T cells had a naïve (CD62L+ CD45RO-) phenotype, the final cell products contained greater than 80% central-memory (CD62L+ CD45RO+) T cells. Finally, the number of CD19 CAR T cells obtained from these pre-clinical manufacturing runs ranged from 7.82e8 to 2.21e9 cells.

Conclusions

50 ml of cryopreserved PBMCs was adequate to manufacture clinically relevant CAR T-cell therapy doses from healthy donors not previously exposed to chemotherapy. Sufficient numbers of CAR T-cells were obtained to dose an 80 kg individual with at least 9e6 cells/kg which is greater than prescribed commercial doses of CD19 CAR T-cells. Further studies are indicated to determine if T-cells collected prior to disease modifying chemotherapies result in an improved product. These results demonstrate feasibility for generating CAR T cells from small volumes of whole blood collected at a time point before a cancer patient has been treated with multiple lines of therapy that could negatively impact starting T cell numbers and function.