Optimizing bioreactor productivity for therapeutic immune cells

Recent clinical trials have shown the potential of adoptive transfer of antigen specific cytotoxic T-lymphocytes (CTL) to treat a wide range of viral and malignant disease. A key factor limiting the widespread adoption of this technology are the difficulties faced procuring clinical quantities of cells in a rapid and inexpensive manner. This collaborative work focuses on the generation and expansion of cytomegalovirus (CMV)-specific CTL for patients undergoing allogeneic haematopoietic stem cell transplantation. A two-stage protocol is employed. The first uses dendritic cells pulsed with the CMVpp65 matrix protein peptide to stimulate CTL from CMV seropositive donors while the second involves expansion of the cell culture to clinically significant numbers using high concentrations of the growth factor IL-2. Successful trials with 100 ml spinner flasks have prompted us to move to a 1-litre suspension bioreactor whose productivity is strongly influenced by IL-2 and lactate concentrations (the former having several possible source and sink terms). Attempts at modelling CTL expansion with a homogeneous unstructured model showed that its predictive capabilities are limited to about the first 48 h (an expansion may last up to 14 days) even when the kinetics of receptor-mediated IL-2 endocytosis are included. Further experimental data suggested that the down-regulation of activated CTL to quiescent cells in the absence of a stimulus has to be included in the model. The paper thus discusses a compartmentalised model incorporating individual phases of the cell-cycle. The latter is being used to look at the expansion kinetics in each phase of the cell cycle. The paper concludes with a discussion of this work as a generic approach to producing clinical quantities of CTL for a range of applications.