Optimization And Metabolic Characteristics Of CHO Cell Culture Process Using Quality By Design Principles

With the increasing clinical demand of monoclonal antibody,the manufacturers are urgent to improve the productivity and quality of therapeutic antibody produced by mammalian cells.The cell culture process is the key unit for the production of antibodies relating with the productivity and quality.Recently,more attention has been focused on process parameters optimization and their final impact on the process;performance,such as viable cell density and protein titer.The lack of research on their relationships impeded the development of therapeutic antibody.The traditional monitored parameters(i.e.physicochemical parameters,by-products and viable cell density)are limited to explain the mechanism connecting the input variables with output results during cell culture process.Therefore,this study combined metabolomics with multivariate data analysis under quality by design principles for the intense research of process data.The metabolites correlating with the process parameters,environmental physicochemical parameters and process performance were identified,which builds the basis of bioprocess understanding and in-process control strategy.The main innovation and academic contributions of this study are summarized as follows:1.The correlation between the physicochemical parameters with monoclonal antibody production in cell culture was studied.Process parameters in cell culture process were optimized by analyzing the process data,including physicochemical parameters and process performance.The correlation between physicochemical parameters was revealed,while the characteristic of physicochemical parameters relating with protein production under different culture conditions was identified with the aid of simple factor analysis and MVDA.The coefficients of process parameters by PLS regression were used for identifying the information relating with process performance.MVDA could effectively extract the physicochemical characteristics relating with monoclonal antibody production from process data that provide the direction for process parameters optimization and process monitoring.2.The feasibility of quality by design(QbD)in the development and optimization of a semi-quantitative method was evaluated for the CHO cell culture supernatant.Analytical design space of HPLC-MS/MS method to assay six representative metabolites in cell culture supernatant was defined.The impact of the HPLC-MS/MS process including chromatographic separation and mass spectrometric detection on the method performance was investigated.And the connection between the critical analytical parameters and the analytical quality of HPLC-MS/MS was further studied.The method developed following QbD distinctly improved the intensity of analytes by HPLC-MS/MS.This exploration provides a reference for developing a robust quantitative method by HPLC-MS/MS for metabolites in biological samples.3.The feasibility of QbD in optimization of cell culture process parameters was presented.The samples cultured in different levels of pH,DO and temperature were taken as research objects.The critical process parameters(CPPs)of cell culture were identified by partial least square(PLS).Principal component analysis(PCA)was applied to identify the metabolites significantly influenced by alerted CPPs,while Pearson correlation test was introduced for screening the metabolites connected with recombinant protein titer.The metabolites commonly identified by these two methods were the potential marker metabolites linking CPPs to mAbs yield.Their metabolic pathways were also shortlisted.It is demonstrated that combing with QbD principle,MVDA and metabolomics provide a guide tutorial to understand the functional relationships linking CPPs to critical quality attributes(CQAs).4.The metabolic characteristics in different scale bioreactors of CHO cell culture were compared.The correlation between environmental physicochemical parameters,metabolites and process performance in laboratory-and manufacturing-scale CHO cell culture was individually uncovered by hierarchical clustering.Meanwhile,the similarity and dissimilarity of the correlated metabolites and pathways in different scales were explained.According to the characteristic metabolites during manufacturing scale by PCA,the in-process control strategy was put forward as example.The current study shows that the understanding of the correlation between the environmental physicochemical parameters,metabolites and process performance will help a lot in process monitoring and process regulation.