Miguel Armenta Ochoa
BME PhD Defense Presentation

Date: 2025-06-09
Time: 1:00 pm
Location / Meeting Link: CHOA Conference Room EBB; https://gatech.zoom.us/j/94251446554?pwd=E2hMgR6qfDPVh7TTGWLSJInlJdkVRa.1

Committee Members:
Krishnendu Roy, Ph.D. (Advisor); Gabe Kwong, Ph.D.; Leslie Chan, Ph.D.; Erik Dreaden, Ph.D. ; Lawrence Boise, Ph.D.


Title: A Multi-Factor Characterization of Degradable Microscaffolds for Therapeutic T Cell Manufacturing

Abstract:
Cell therapies gained first approval in 2017 in the form of CD19-targeting Chimeric Antigen Receptor (CAR) T cells to treat certain B cell leukemias and lymphomas. Remarkable initial response rates and durable remission for several of the patients from the original trials have proven the curative potential of these treatments, with 7 CAR-T products now approved for a variety of liquid tumors. However, translation of these cells into the clinic has proven challenging due to cell manufacturing challenges and cost that drive down patient access. For T cell therapies to become more widespread, a deeper understanding of the culture process and the variables which most influence it is needed. This dissertation aimed to explore which factors are most influencing T cell culture outcomes using three essential tools: a biomaterial T cell activator platform, an automated perfusion bioreactor system, and a computational modeling software to help interpret complex multiparameter datasets. The activator platform was our Degradable Microscaffold (DMS) system which consists of commercially available collagen microcarriers functionalized with T cell activating antibodies. We developed a well characterized library of DMS lots to test the role of ligand density on T cell culture. The bioreactor in question was the Mobius® Breez (Millipore Sigma) consisting of 4 individually addressable culture pods and automated culture control. We used the platform to test environmental culture parameters such as agitation and hypoxia, while highlighting its utility as a T cell bioreactor. Finally, the Data Modeler software package (Evolved Analytics LLC) aided greatly in exploring the complex combinations of potential variables by building ensembles of high accuracy, low complexity non-linear computational models for any cell response of interest. We explored an array of culture manipulations such as cell isolation, activator removal timing, and culture vessel alongside testing our DMS library and commercial T cell activators. This research sought to improve the manufacturing of therapeutic T cells by detailing the importance of cell activator, cell starting material, and culture environment on cell expansion and phenotype.