Application of the Quality-by-Design (QbD) Approach for Erythropoietin Alpha Purification

This study was aimed at process characterization and improving quality of purification of erythropoietin α, a biopharmaceutical agent. In biopharmaceutical manufacturing, quality should always be targeted to ensure safety and efficacy. Design-of-experiments–based approaches have been explored to rapidly and efficiently achieve an optimized yield and an increased understanding of a product and process variables affecting the product's critical quality attributes in the biopharmaceutical industry; this system is known as the quality-by-design approach. Changes in three critical process parameters—buffer pH, flow rate, and loading amount—were evaluated. Process characterization was conducted on a scaled-down model previously validated by comparison with data from a large-scale production facility. Seven critical quality attributes—relative aggregate content, host cell protein, host cell deoxynucleotides, endotoxin, Z-value (N-glycan score), relative content of charge isomers, and step yield—were analyzed. Multivariate regression analysis was performed to establish statistical prediction models for performance indicators and quality attributes; accordingly, we constructed contour plots and conducted a Monte Carlo simulation to clarify the design space. As a result of the optimization analysis of the purification process, it was confirmed that proven acceptance ranges were optimized as follows: loading amount (mg/mL) 0.4–4.0, buffer pH 7.0–8.0, and flow rate (mL/min) 0.5–1.6.