Two examples of selective outsourcing at Serono will now be described. One is related to a specific technology that supports
speed- to-market and quality objectives, and the other is part of the company's drive to achieve platinum quality.
EXAMPLE I: PURIFICATION PROCESS IMPROVEMENT
The problem to be solved was that the current capture step (first recovery) of the process for clinical Phase III trials
did not meet Serono's target levels of quality and efficiency. Therefore, the objective was to improve the purification process
of a new molecule to meet clinical Phase III standards within an established time plan. This is illustrated in Table 4.
Table 4. Serono International - Purification Process Improvement Plan
An extensive development program was launched, and over 50 commercial chromatographic resins were tested, but without significant
improvement. It then became clear that alternative technologies should be used to develop a simpler process with platinum
quality results (HCP <5 ppm and aggregates <1%). One of the approaches to reaching this goal involved developing a custom-made
affinity ligand that would provide the desired selectivity. The development was outsourced to a highly specialized company
with proven records of expertise in this field. The development study was based on protein structure, and a ligand was developed
with high specificity of binding to the target protein, as shown in Figure 1.
Figure 1: 3-D Protein Structure Model and Potential Binding Sites
The most desirable features of the prospective ligand were: high selectivity, chemical robustness, low cost, industry proven,
and high purity. Within 10 weeks, 1728 ligands were screened and an optimal ligand was selected. This process is shown in
Figure 2. The development activity performed by the outsource could not have been done in-house at Serono due to a lack of
the specialized expertise that this contract research organization (CRO) brought to the project. Definitely this collaboration
shortened the time to market and improved the quality and efficiency of Serono's development process.
Figure 2. Ligand Development Study
EXAMPLE II: FINE DETECTION OF PROTEIN X AGGREGATION
The objective of this project was to develop methods to detect, quantify, and classify (characterize) protein X aggregates
to prevent or at least minimize their level in the final product. Understanding of the type of aggregation (covalent versus
non-covalent bonds, hydrophobic or ionic) would permit kinetic and thermodynamic studies of the aggregation process, of the
mechanism of aggregation, and of the factors affecting their appearance or their disintegration. This information would impact
on the ability of manufacturing to minimize the level of aggregation by improving the storage, handling, and shipment conditions
of the intermediate and final bulk. The issue was that the analytical methods were not sufficiently sensitive to reach the
objective; thus, new, sensitive methods had to be developed to quantify and classify protein aggregation.
This task was outsourced to a specialized CRO that developed and introduced state-of-the-art analytical tools, such as AUC
(analytical ultra centrifugation) -sedimentation velocity, and new high-performance liquid chromatography (HPLC) methods to
enable the detection and qualification of all classes of aggregates.
With these characterization technologies, it was possible to obtain the following results:
- In "normal conditions," non-covalent aggregates constituted the majority of protein X oligomers
- Reversion to monomers was possible upon addition of new excipients
- Dimers could be reverted to monomers upon proper manipulation and processing (as per standard operating procedure)
- Storage temperature was critical (e.g., freeze/thaw cycles)
- Aggregation could be significantly reduced by mixing with a cryoprotectant (prior to freeze/thaw)
In this case, the specialized CRO helped Serono to better understand the drug molecule, to improve final product quality to
reach platinum quality standards, and to provide strong support to manufacturing.