Detection of lentiviral constructs for release testing of CAR- T cells using digital droplet PCR

Abstract

Background & Aim Viral vectors are commonly used to introduce chimeric antigen receptor (CAR) constructs into cell therapy products for the treatment of human disease. While viral vectors are efficient at gene delivery and able to stably integrate into the host genome for subsequent replication, they also pose a safety risk if replication competent lentivirus (RCL) remains in the final patient product. FDA guidelines require testing of CAR-T products to ensure absence of RCL before delivery to the patient. Current testing methods include cell based assays that require significant time to produce results or qPCR based assays that provide rapid results but lack the limit of detection for such low copies of target DNA. We describe here the development of digital droplet PCR (ddPCR) for detection of the vesicular stomatitis virus G glycoprotein (VSV-G) envelope sequence, a component required for the assembly of competent viral particles. Validation of this method provides several advantages over qPCR by improving limit of detection, reducing sources of human error by elimination of the standard curve and reducing hands on technical time. Methods, Results & Conclusion We used the BioRad QX200 ddPCR system to test samples of defined concentration over a broad range of conditions. Samples of limiting dilution were tested to prove assay linearity over a broad range of concentrations. Reproducibility was established by running the defined samples in multiple wells or over consecutive days to compare variability over time, within an assay and between assays. Specificity was demonstrated by running replicates of positive and negative samples to determine a false positive rate and limit of detection. Taken together, we have provided experimental and statistical evidence for the use of ddPCR as a reliable release test for the manufacture of CAR-T products. Our results demonstrate the reproducibility, linearity, specificity and sensitivity to detect RCL and assure the safety of patient products in a rapid manner. Future efforts will be focused on adapting this technology to detect other valuable targets throughout cell product manufacturing.