Multi-objective optimization of monoclonal antibody production in bioreactor

Abstract

The increasing demand for monoclonal antibodies (mAbs) based therapeutics stipulates their efficient production. mAbs are predominantly produced in a fed-batch bioreactor with temperature and media composition as key control variables. In this paper, we have developed a mathematical model incorporating the effect of temperature, biomass, glucose, protein, and lactate concentration on mAb productivity. The model parameter values are estimated by minimizing the normalized error function using the particle swarm optimization (PSO) algorithm. The model is further used to find the optimum temperature and reactor input-output flow rate strategy for maximum mAb production in minimum reactor run time for a traditional sequential combination of the batch, fed-batch, and perfusion modes of operation. The non-dominated sorting genetic algorithm (NSGA-II) has been successfully implemented for bi-objective optimization, which estimated a 5% increase in mAb production with respect to the experimentally produced mAb following combined mode of operation, and the same reactor run time and nutrients’ consumption using the optimized operating conditions. Continuous production is becoming increasingly popular for the intensification of biopharmaceutical processes. Therefore, the simulation is further extended to continuous mAb production, and multiple flow strategies are compared to maximize daily mAb production while minimizing nutrients’ requirement, wastage, and reactor volume.