| With the rapid development and expanding production scale of monoclonal antibody(mAb)drugs,it is an urgent issue to improve the process efficiency and throughput.Continuous biomanufacturing showed obvious advantages of increasing process efficiency and product quality,promoting equipment miniaturization and process automation.The key bottleneck of mAbs continuous manufacturing is downstream separation and purification.Due to the complex operation modes of continuous chromatography,process development is difficult and systematic research is lacking.This thesis focused on the continuous capture process with Protein A affinity chromatography,and established the model-assisted process development and optimization methods.The effects of operating parameters,chromatographic resins and process conditions on the continuous capture process were investigated systematically to obtain a comprehensive and in-depth process understanding.In addition,a new model-assisted process characterization method was established for continuous capture to promote the practical application of continuous chromatography.Firstly,a model-assisted process development method was established for twincolumn CaptureSMB continuous capture process.The influence of operating parameters on process performance was analyzed systematically.Using MabSelect PrismA as a typical affinity resin,the model-based process optimization was carried out to increase productivity and resin capacity utilization.The optimal loading residence time was in the range of 1～2 min,the breakthrough percentage was between 0.6～0.75,and the optimal process productivity at different feed concentrations was between 14～47 g/L/h.The experimental results confirmed the accuracy and reliability of the model-based predictions.To explore the influence of affinity resins on CaptureSMB process,MabSelect PrismA,Praesto Jetted A50,MabSelect SuRe pcc and Unimab 50HC were compared by the model-assisted method.In the consideration of continuous process performance and basic resin characteristics,the maximum adsorption capacity and effective diffusion coefficient of resin were determined to be the key factors.Finally,the appropriate ranges of resin characteristics were determined by the model-assisted method,which would be useful for the design of next-generation affinity resins.The continuous capture process involves a large number of operating factors with complex interactions.Different continuous capture modes have different characteristics.Therefore,the model-assisted process development method was developed for threecolumn periodic counter-current chromatography(3C-PCC).The interactive effects of operating parameters,resin properties,process conditions and constraint factors on process performance were investigated.The results showed that the upper limit of the maximum productivity for 3C-PCC could be determined by the maximum adsorption capacity of resin and the recovery and regeneration(RR)time.With the increase of the maximum adsorption capacity or the decrease of feed concentration and the RR time,the residence time at the maximum productivity would be shortened.For short residence times,the maximum productivity would be limited by the resin capacity utilization and operating flow rate.It was found that there is a critical concentration that can determine whether the maximum productivity operation condition is affected by the constraint factors.Furthermore,a simplified process design strategy was proposed,which could use a set of breakthrough curves to determine the optimal operating conditions of 3C-PCC process.The results of experimental verification were good.For the complex multi-column continuous capture mode of BioSC process,the model-assisted process development method was established and verified by the multicolumn BioSC experiments.Many factors were evaluated,including operating parameters,resin properties,process conditions and constraint factors.It was found that there is a theoretical maximum value of process productivity for BioSC,and the operating conditions of the theoretical maximum productivity are mainly affected by the feed concentration.At low feed concentration,it is difficult to reach the theoretical maximum productivity.As the feed concentration increases,the actual process productivity could approach the optimal value.Based on the process understanding,it was proposed that the maxmum adsorption capacity,feed concentration and RR time could be used to determine whether the optimal process performance can be obtained under the constraint factors.Furthermore,the optimal productivity of CaptureSMB,3CPCC and BioSC under varying conditions was evaluated and compared systematically.The results indicated that BioSC can achieve the best process performance under most conditions.Based on the model calculations,a simple strategy was established for screening the appropriate continuous capture modes.Finally,based on the model-assisted process development on continuous capature,a new model-assisted process characterization method was established for continuous capture processes.The traditional experiment-based method requires numerous experiments,and is time-consuming and difficult to obtain the rational design space.Therefore,the continuous capture process model was introduced to construct a modelassisted process characterization method with risk assessment.The model was applied to exclude the non-operable areas and optimize the process performance,and then to establish a dynamic examination range of parameters according to the process conditions.The model-based DoE was carried out to characterize the influences of operating parameters on the process performance and determine the critical process parameters.According to the results of model-based DoE,the minimum number of DoE experiments were implemented to complete the experimental verification and parameter robustness analysis.For the typical 3C-PCC process,four critical process parameters and operating ranges were determined by the model-assisted method with 16 experiments.The feasibility and reliability of the model-assisted method were verified in the case of the 3C-PCC experiments.Compared with the traditional experiment-based method,the new model-assisted method can reduce the experimental workload and obtain more reasonable and reliable design space.In summary,this thesis proposed the model-assisted process development and optimization method for continuous capture processes.The effects of operating parameters,resin properties and operating modes on continuous capture process were studied,and a comprehensive process understanding was obtained.The results can guide the development of continuous process,design of new resins,selection of continuous capture modes,and provide a new method for continuous capture process characterization. |
PDF Full Text Request