2nd Edition of Cell & Gene Therapy World Conference 2026

Abstract

Accurate detection of vesicular stomatitis virus glycoprotein (VSV-G) DNA is essential for monitoring replication-competent lentivirus (RCL) in gene therapy programs, where even low-level emergence of replication-competent virus poses significant safety concerns [1, 2]. To support this need, we validated an analytical workflow that integrates manual DNA extraction using QIAamp Blood Mini and Midi kits with a quantitative PCR (qPCR) assay optimized for VSV-G DNA detection in human whole blood and peripheral blood mononuclear cells (PBMC). The goal was to establish a robust, accurate, and CLIA-compliant method specifically designed for VSV-G material supporting internal qualifications and for use in regulated clinical studies [3]. The validation strategy encompassed a comprehensive set of analytical performance characteristics, including linearity, dynamic range, precision, limit of quantification (LOQ), effective concentration at 95% detection probability (EC95), specificity, accuracy, co-linearity, robustness, and extraction efficiency. Human matrices were spiked with VSV-G DNA isolated from HEK293 cells transduced with VSV-G–pseudotyped lentivirus, ensuring that assay performance reflected clinically relevant sample composition and potential matrix effects. The qPCR assay demonstrated strong quantitative performance across a validated range of 106 to 10 copies/5 µL. Probit analysis established an EC95 of 11.3 copies, defining the threshold below which results are reported as below the limit of detection (<LOD). The LLOQ was confirmed at 10 copies/5 µL, with consistent amplification efficiencies and reproducible cycle threshold values across multiple operators, instruments, and reagent lots. No undetermined replicates occurred at the LLOQ, supporting its reliability. These results demonstrate that the assay maintains sensitivity and precision at the low copy levels required for RCL surveillance. Specificity testing confirmed selective amplification of VSV-G sequences without cross-reactivity to non-target DNA. Accuracy assessments showed strong agreement between measured and nominal concentrations across the validated range, and co-linearity analysis demonstrated proportional response between expected and observed copy numbers. Robustness was evaluated through deliberate variations in operator, instrument platform, and reagent lots; all conditions produced consistent results, confirming that the workflow is resilient to routine laboratory variability. Extraction efficiency was evaluated using spiked whole blood and PBMC matrices. Recoveries ranged from 28% to 76%, reflecting acceptable performance for manual extraction workflows and confirming that the extraction step does not introduce significant bias or compromise downstream quantification. The positive control, VSV-G DNA (isolated from transduced HEK293 cells) performed equivalently to the linearized plasmid standard, validating its suitability as a reference material for ongoing assay qualification. No matrix interference was observed in either whole blood or PBMC, demonstrating that endogenous components do not inhibit amplification or reduce assay accuracy. Collectively, these results confirm that the combined extraction and qPCR workflow is robust and accurate, under the study conditions for detecting VSV-G DNA in support of RCL monitoring. The test method meets all acceptance criteria for sensitivity, specificity, precision, and reproducibility, supporting its use in both internal qualification studies and CLIA-regulated clinical workflows. This validated approach provides a dependable analytical foundation for supporting the safety of lentiviral gene therapy products through sensitive detection of VSV-G DNA.