(J-389) Degradable Microcarrier Scaffolds (DMSs) for Scalable CAR T Cell Manufacturing

Chimeric antigen receptor (CAR) T cells have emerged as an attractive form of cancer immunotherapy in recent years, causing remissions in previously relapsed and refractory hematologic malignancies. Despite the immense benefit of this personalized treatment, there remain several challenges associated with the ex vivo manufacturing process. Currently commercialized CAR T cell products are initially derived from patients’ own T cells—with samples being harvested, activated, transduced, and expanded to generate the autologous infusion. However, this process can take several weeks in highly time-sensitive scenarios. Owing to the small scale of the therapy, the complexity associated with manufacturing greatly increases the financial burden of this potentially lifesaving treatment modality, preventing widespread adoption. Thus, continued research into streamlining the expansion and redefining the “gold standard” of the process is required.

Here, we describe an approach for the activation of T cells in scaffold microenvironments by mimicking natural T cell activation in the lymph node. This system of antibody-coated, gelatin-based degradable microcarrier scaffolds (DMSs) will function as artificial antigen-presenting cells (APCs) and be used to develop functional CAR T cells. These CAR T cells will initially target B-cell maturation antigen (BCMA), expressed on the surface of multiple myeloma cells and already targeted by the FDA-approved cell product Abecma. We aim to mitigate any failures caused by variability resulting from the relatively limited quantity and disparate qualities of cells originating from patient samples to create a more effective downstream product with greater therapeutic benefit.

DMSs were fabricated from CultiSpher-S gelatin microcarriers (Sigma-Aldrich) and coated in biotinylated monoclonal anti-CD3 and anti-CD28 human antibodies.  The surface concentration of antibodies was determined using a bicinchoninic acid assay kit (Thermo Fisher). 0.05 mg/mL Collagenase D (Sigma-Aldrich) was added to degrade the DMSs overnight in the incubator. These were then used to culture isolated primary human T cells (Charles River or STEMCELL Technologies) for 14 days after the initial seeding and activation. Glucose readings were recorded using a GlucCell Glucose Monitoring System for each respective experimental group to monitor growth. Cell counts and viability were assessed every 2 days using Trypan Blue (Gibco) and an Invitrogen Countess 3 Automated Cell Counter (Thermo Fisher), Additional media consisting of recombinant human IL-2 (Peprotech) in serum-free medium (TexMACS, Miltenyi Biotec) was added as necessary. Using an IgG ELISA kit (Abcam), samples from the DMSs were taken to quantify the amount of antibody binding. Two distinct DMS groups, one group expanded using coupled magnetic beads (Dynabeads), and one group using the T Cell TransAct nanomatrix (Miltenyi Biotec) were compared. FACS buffer was then prepared for staining for multicolor flow cytometry corresponding to memory surface markers. The panel was constructed as follows to differentiate the stages of the memory population: CD62L, CCR7, CD4, CD8, CD45RO, and CD95.

Lymph node-mimicking DMSs resulted in robust expansion of T cells compared to paramagnetic beads, potentially owing to the increased T cell infiltration in the large pores of the microcarriers. DMSs can influence T cell memory profiles, with improvements in DMS fabrication and QC leading to less variability and better control during T cell growth. It has been previously established that T cells can be manufactured in preference of certain memory populations, the presence of which can influence clinical outcomes. Furthermore, evidence suggests that the starting T cell phenotype influences the ultimate CAR T cell product, so future comparisons of expanded T cells between healthy and multiple myeloma patients will be assessed. Cytotoxicity assays will be conducted to validate the efficacy of cells expanded from donor-derived samples. TSCM cells have remarkable proliferation potential and longevity leading to greater antitumor responses. Balanced CD4:CD8 ratios have also demonstrated improved CAR T cell efficiency, contributing to the therapeutic value of preselection. Flow cytometry will be ran on cells from both DMS batches to determine if the ultimate memory profile can be predicted from antibody concentration.

We would like to thank Miguel Armenta Ochoa and Dr. Krish Roy for their help and mentorship. This work was conducted at the NSF Engineering Research Center for Cell Manufacturing Technologies (CMaT), Award No. 1648035, and Marcus Center for Therapeutic Cell Characterization and Manufacturing (MC3M), which are dedicated to systematically realizing a reproducible approach for the cell therapy industry.

1. 
   

   Liu Y, Sperling AS, Smith EL, Mooney DJ. Optimizing the manufacturing and antitumour response of CAR T therapy. Nat Rev Bioeng. 2023;1(4):271-85. doi: 10.1038/s44222-023-00031-x.

   
   


2. 
   

   Odeh-Couvertier VY, Dwarshuis NJ, Colonna MB, et al. Predicting T-cell quality during manufacturing through an artificial intelligence-based integrative multiomics analytical platform. Bioeng Transl Med. 2022;7(2):e10282. Published 2022 Jan 4. doi:10.1002/btm2.10282.

   
   


3. Abou-El-Enein M, Elsallab M, Feldman SA, et al. Scalable Manufacturing of CAR T cells for Cancer Immunotherapy. Blood Cancer Discov. 2021;2(5):408-422. doi:10.1158/2643-3230.BCD-21-0084.