Selection of high-producing clones by a relative titer predictive model using image analysis

The commercial success of monoclonal antibodies (Mabs) has made biological therapeutics attractive to pharmaceutical companies. The priority of biopharmaceutical companies is to acquire and develop cell lines that enable them to manufacture biologics quickly, consistently, and economically. Clone selection is a critical process for cell line development. However, the traditional clone selection process requires the evaluation of large numbers of clones using cell growth rate, cell densities and titer, product quality, and so on.

Here, we present the first predictive model that can be used to estimate the relative productivity of Chinese hamster ovaries (CHO) clones during the cell line development. Additional experiments are currently in process to further improve the RT predictive model. Nevertheless, our current study will serve as a foundation for more prediction models for cell line development that can facilitate the selection of clones.

To improve efficiency of the clone selection strategies, we developed a relative titer (RT) prediction model by the quantitative information extracted from microscope images during the cell line development process. The performance of this RT prediction model was further evaluated with 50 clones from 5 different cell lines.

The RT prediction model was able to predict high producers from a given data set when the same host cells were used. Although inaccurate prediction occurred when different host cell was used, this RT prediction model may serve as an excellent proof of concept study that quantitative information from cell line development images provides valuable information to facilitate the cell line development process. Conclusions: Here, we present the first predictive model that can be used to estimate the relative productivity of Chinese hamster ovaries (CHO) clones during the cell line development. Additional experiments are currently in process to further improve the RT predictive model. Nevertheless, our current study will serve as a foundation for more prediction models for cell line development that can facilitate the selection of clones.