P017 | QUANTITATIVE IMAGING BASED ON ADVANCED MICROSCOPY TECHNIQUES TO MEASURE CAR-T ACTIVITY

Developing new effective methods to characterize CAR-T activation and cytotoxic activity is crucial to improve the understanding of this cell therapy product action and to optimize their manufacturing and administration. At this purpose, we developed a time-lapse confocal microscopy-based approach to quantify several parameters describing CAR-T killing activity. GFP positive CAR-T cells were generated through transduction with a third generation anti CD19.CAR-T eGFP tagged lentiviral vector from primary T cells obtained from 5 healthy donors (HD) and 4 patients affected by B cell lymphoma with the therapeutic indication for CAR-T cell therapy at the advanced cell therapy unit of the IRCCS AOU of Bologna (NCT04892433_CE 319/2021/sper/AOUBO). After the generation procedure, CAR-T killing potency was tested in co-culture both using fresh cells and after a freezing step. K562 cell line, engineered to express CD19 cDNA and a mCherry tag, was used as target cells in the co-cultures. Cells were seeded at a 1:1 CAR-T to target ratio and placed under a Nikon A1R laser scanning confocal microscope equipped with a stage incubator. Imaging was performed every 3 minutes for at least 12 hours. Collected images have been analysed to obtain quantitative data of the interaction between CAR-T and target cells in the different conditions. The measure of co-localization of GFP (CAR-T) and mCherry (target) signal, the dimension and the number of clusters generated by cells interactions have been considered as the most representative variable correlated with CAR-T cell killing activity. Through statistical modelling of these time series data, significant differences have been observed in co-localization trends across the different conditions. HD derived CAR-T display a significantly increased slope if compared with patients derived CAR-T and, most interestingly, the ratio between dimension and number of clusters revealed a four times increased slope in the HD fresh CAR-T condition compared with the patient-derived fresh ones. These data suggest that time lapse microscopy represents a valid approach to deeply characterize CAR-T cell activity giving novel perspectives and insights to the quantification of their efficacy. The comparison between different T cells sources and manufacturing conditions allowed us to stratify donor-derived CAR-T from patients ones with a significant better performance of the first group according to the evaluated parameters.