Optimization techniques: Studies in cell culture

With the present large worldwide demand for vaccines and cellular by products, a number of anchorage-dependent cell mass culture systems have been developed. To obtain the maximum yield of product per run, the number of cells per system volume must be maximized and yet maintain efficiency. There is a large number of independent variables that could affect the cell yield. During the course of an experimental program to increase the cell yield, the usefulness of an existing computer optimization program was evaluated.A series of nine statistically designed experiments was done using multi-disk propagators for the mass culture of primary chick embryo cells. Cell input and time between medium refeeds were chosen as the independent variables from the many variables that could affect cell yield. The primary dependent variable was cell yield. The results indicated a substantial increase in cell yield as cell input was increased (57 to 85 percent increase by doubling the cell input) but a lesser increase in cell yield with an increasing number of refeeds.The data were subjected to regression analysis by the existing computer program. The second order polynomial equation (cell output as a function of cell input and time between refeeds) aptly described the experimental data. Three additional experiments were done with 50 percent more cell input than the maximum cell input in the experimental design. The predicted cell yields from the polynomial equation were approximately 10% greater than the cell yields obtained from the three experiments.