Biotechnology is a strategic emerging industry in the 21st century. Vigorously developing the biopharmaceutical industry with biotechnology as the core is not only conducive to promoting the transformation of an innovative economy, but also the only way to promote the construction of a healthy China and improve people's health standards. In recent years, with the reform and improvement of the drug review and approval system, the number and speed of new drug approvals have increased, which has also brought new challenges to biopharmaceutical companies, that is, how to quickly scale up complex biopharmaceutical production processes from small trials to production, establish a reliable and stable production control strategy, and continue to produce safe and effective drugs.
Faced with this problem, large pharmaceutical companies at home and abroad generally follow the relevant ICH guiding principles, adopt the core concept of Quality by Design (QbD), and widely use Design of Experiment (DOE) to solve the problem of biological drugs. There are many process parameters and complex interactions in the production process. On the basis of small-scale detailed DOE experiments, the correlation between production process parameters and quality attributes is established, and the characteristics of the production process are fully understood to effectively ensure the smooth implementation of large-scale production. See Figure 1 for details.
Figure 1. Flow chart for drug development and production according to ICH guidelines
The DOE method can be used to study the relationship between multiple parameters with fewer experiments. The traditional DOE implementation method first conducts a low-resolution screening experimental design for a large number of parameters. Depending on the experimental results, a second round of amplification design or reaction surface experimental design may be required. There are many parameters in the biopharmaceutical production process, which often include high-order effects that require response surface design to clearly distinguish. In addition, they also face the pressure of experimental material cost and time constraints. These practical difficulties determine that it is very challenging to use traditional DOE experimental design methods to develop biological drugs. The Definitive Screening Design (DSD) unique to JMP software has the advantages of fewer experiments, high resolution, and time saving, which provides great convenience for solving this problem. The following is explained through specific examples.
The production process of monoclonal antibodies includes three parts: upstream cell culture, downstream purification and product filling. Cation exchange chromatography is an important step in fine purification in downstream purification production. Impurities such as aggregates, charge variants, host cell protein residues, and exogenous DNA residues can be effectively removed through cation exchange chromatography. These impurities are critical quality attributes. (Critical Quality Attribute, CQA). Through experimental design methods, the impact of cation exchange chromatography process parameters on CQA was studied and appropriate control strategies were formulated. The reaction indicators investigated by cation exchange chromatography are shown in Table 1.
Table 1. Reaction indicators investigated by cation exchange chromatography
The cation exchange chromatography process includes nearly 20 process parameters. Based on the knowledge and experience of other projects, 8 process parameters such as chromatography flow rate and loading capacity that may have an impact on the purification results were selected for further experimental investigation. Develop research plans through JMP software DSD experimental design. All process parameters are continuous variables and do not need to be divided into blocks. The JMP software is preset to generate a set of 21 experimental runs. The R&D personnel added one more repeated experiment under three conditions to examine the systematic error in the experimental process. A total of 24 experiments were conducted. The plan is detailed in Table 2.
Table 2. Cation exchange chromatography experimental scheme
Randomly run experiments according to the experimental plan and collect test data. Since DSD uses a small number of experiments to examine a large number of factors and effects, it is impossible to directly perform standard least squares analysis on DSD data. You can use the unique "fitting certainty screening" method of JMP software, or through the "fitting model" The "stepwise regression" fitting method in . The R&D personnel successfully established regression models for five responses, including yield and SEC-HPLC purity, through a step-by-step fitting method. The prediction profiler results are shown in Figure 2.
figure 2. Cation Exchange Chromatography Experimental Design Prediction Profiler Results
According to the above model, the cationic process parameters X6, X7 and X8 have a greater impact on various critical quality attributes and yields. They are critical process parameters (CPP) and the parameter range must be strictly controlled. Other parameters have little impact on various critical quality attributes and yields. The risk of CQA exceeding the range caused by changes in process parameters is low, so they are all non-critical process parameters (non CPP). Using the deterministic screening experimental method of JMP software, we successfully studied the process characteristics of cation exchange chromatography through only one round of 24 experiments, screened out key process parameters, and formulated targeted control strategies to ensure the safety and reliability of the final product. If you are interested in JMP deterministic screening DSD, you are welcome to download it. JMP 30-day free trial .
