Accelerated cell line development and improved characterisation of lentiviral vector production through application of MALDI-ToF mass spectrometry and multivariate data analysis.

Abstract

Several cell and gene therapies will be commercially launched within the next few years using lentiviral vectors as the gene delivery vehicle. Oxford BioMedica’s Lentivector® platform is an advanced lentiviral-based gene delivery system designed for improved safety and efficacy. The growing interest in these vectors has created a strong demand for large scale production of lentiviral vectors as well as for development of packaging and producer cell lines. This EngD project used a combination of matrix assisted laser desorption ionisation time of flight mass spectrometry (MALDI-ToF MS) and multivariate data analysis (MVDA) to analyse cell and lentiviral vector samples. A comparison between mass spectra of samples produced across small and large scale in adherent and suspension culture was used to identify what aspects of the manufacturing process had the biggest impact on cell and vector variation. Principal component analysis was applied to compare different lentiviral vector production methods, assess data structure of the process parameters and examine whole cell and vector mass spectrometry data. This approach led to improved characterisation of lentiviral vectors and HEK293T cells. It demonstrated the capability to differentiate between adherent and suspension cells as well as cell lines of different levels of performance as defined by lentiviral vector infectious titre. Partial least squares discriminant analysis (PLS-DA) was used to calibrate and validate a predictive model of cell line performance based on mass spectrometry and viral vector titre data obtained from multiple HEK293T cell lines. PLS-DA model validation resulted in 87.5% accuracy in classification of cell lines as high or low producers based on a discrimination threshold determined by viral vector titre. The results of PLS-DA modelling indicated that this method can be used for accurate cell line performance prediction, accelerating cell line development by several weeks, improving cell selection and reducing campaign timelines.