Noaks, Elsa; (2021) Enhancing the consistency of donor derived peripheral blood mononuclear cells for CAR T-cell manufacturing. Doctoral thesis (Eng.D), UCL (University College London).

Preview

Text Full thesis_final version_no signature.pdf Download (11MB) | Preview

Abstract

The majority of adoptive T-cell products are manufactured using an autologous process. Although using a patient’s own cells reduces risks of rejection, it introduces variability and the presence of cell populations, such as monocytes, that might negatively affect the success of CAR T-cell production. Furthermore, the current method of collecting T-cells, leukapheresis, requires specialist equipment and trained operators, limiting patient accessibility. This project explored the effect of donor starting material composition on CAR T-cell manufacture. To achieve this, the study was divided into three phases, with the first reviewing an alternative leukapheresis enrichment method. Bead based magnetic separation is the current gold standard for T-cell purification. However, due to differences in adherency, T-cells can also be enriched by capturing unwanted cell types, such as monocytes, on a surface. A range of commercial surface coatings were trialled in static and dynamic systems. Microfluidic platforms were explored but suffered issues with consistent manufacture and cell recovery. Although only recovering ∽20% of CD3+ cells, the most successful enrichment arose from agitated microcarrier cultures, reducing monocyte populations by ~75% and enhancing T-cell activation by up to 33%. While it was possible to enrich T-cells using surface capture, monocytes were never completely removed from culture with ~20% of the starting population remaining. It was determined that microcarrier protocols would require development to make them a viable option for CAR T-cell processing. Having established the ability to deplete monocytes, subsequent work planned to examine the relationship between donor material composition and the success of CAR T processing stages. The impact of monocytes on the level of activation, growth and transduction efficiency was monitored across well-plate and culture bag platforms using healthy donor apheresis. Removal of monocytes from leukapheresis improved the level of activation 2-fold, achieving the same level of activation as when initiating the process with a purified T-cell starting material. Two activation reagents were tested in well-plate cultures, revealing differing sensitivities to starting material composition. Monocyte depletion in culture bag systems had a significant impact on transduction efficiency, improving consistency and increasing the level of CAR expression by up to 64% compared to leukapheresis. Cytotoxicity assays revealed that CAR T-cell products produced from donor material depleted of monocytes and isolated T-cells consistently outperformed those made from unsorted leukapheresis. Analysis of memory phenotypes and gene expression indicated that CAR T-cells produced using depleted starting material displayed a more rested and naïve state, potentially contributing to their enhanced cytotoxic performance. The final phase of this project explored the potential of using whole blood collections as an alternative starting material to leukapheresis for CAR T-cell manufacture. To test its applicability in CAR T-cell processing, healthy whole blood donations were processed to recover leukocytes using density gradients (Ficoll, Sepax) and less conventional filtration techniques (Imugard, Leukotrap and Hematrate). It was thought that blood filters could provide a rapid methods for WBC purification without the need for additional reagents. A complication with using whole blood as a starting material is its high level of red blood cells (RBCs). Density gradients were able to completely isolate white blood cells (WBCs) however, filters retained approximately a sixth of the RBCs present in the starting whole blood, even with refined operation. Lymphocytes derived from an automated density gradient or newly established blood filtration processes were activated. CD4+ T-cells were stimulated to a similar level as leukapheresis from unrelated donors, achieving 43-55% CD25+CD69+%. Conversely, CD8+ T-cells exhibited a significantly lower level of CD25+CD69+% than leukapheresis, at approximately half. Retroviral transduction was poor in filtered material, achieving an efficiency of ∽7% compared to ∽56% by Sepax samples. Addition of an RBC lysis inducing freeze thaw to the process alleviated this issue, with filtered whole blood able to yield transduction efficiencies of 64 – 88%. Furthermore, CAR T-cells derived from density gradients and filtered whole blood consisted of >50% early central memory cells. Although whole blood filtration can produce CAR T-cells, a higher level of RBC depletion and review of processing techniques would be necessary to achieve higher retroviral transduction.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item