Purpose: • Develop a formulation of a humanized IgG1 monoclonal antibody suitable for commercial manufacturing using a multivariate statistical Design of Experiment (DoE)
• Conduct study to optimize the key formulation parameters with respect to chemical and physical stability properties
• Explore the design space surrounding the optimal points identified for the variables
Methods: Selection of the variables to be investigated was based on preformulation studies and previous knowledge of formulation parameters critical to the design of the DoE study.
The variables were mAb concentration, pH and polysorbate-80 concentration
The identification of the possible optimal centre point of mAb concentration, pH, and polysorbate-80 concentration for DoE was performed.
The number of experiment was determined using JMP SAS software.
For reduction in the experimental effort, the full response surface model was used with all two factors interactions that allows the evaluation of the main effects aliased with 2 factor interactions.
This model allowed the evaluation of the continuous factors (mAb concentration, pH, and polysorbate-80 concentration) over the selected ranges.
Estimated effects from fractional factorials 2k design were calculated from the JMP software.
14 formulations were manufactured and the following effects were evaluated: purity by SE-HPLC , related substances , particulate matter and pH.
Results: The response attributes were measured by a variety of analytical techniques to assess the chemical and physical stability of protein on storage and after dilution in IV bag
• Purity, total aggregates (% HMW) and fragments (% LMW) by SE-HPLC
• Particulate matter (≥ 2, ≥5, ≥10 µm and ≥ 25 µm particle size) by light obscuration
The percent main peak by SE-HPLC assay resulted to be affected most by pH with better stability at high pH (see fig.2).
The total aggregates (% HMW) were influenced by protein concentration and pH, with maximum stability observed at low protein concentration and at low pH (see fig.3).
Fragments (% LMW): Low pH showed lower % LMW compared to the high pH (see fig.4)
Conclusion: • DoE approach resulted in a real optimum formulation, not an “acceptable” one
• DoE approach maximizes formulation understanding with a reasonable number of experiments
• DoE approach helped understanding the impact of formulation variations on critical quality attributes when the process will be scaled up to GMP production area