By:Diane Retallack, Ph.D., Director, Molecular Biology at Pfenex
At Pfenex, we are in the business of developing biosimilars. And, as a result, on a daily basis we produce dataÓlots of it. This data enables the team at Pfenex to rapidly demonstrate analytical similarity between a Pfenex biosimilar candidate and the reference drug product. The ability to verify the quality of a biosimilar candidate early on in development enables Pfenex to efficiently progress its entire pipeline of biosimilar candidates. Our protein production platform, Pfenex Expression Technology and our extensive analytical biochemistry capability allows for the parallel evaluation of protein produced from thousands of unique production strains. Throughout this process we employ the Core LIMS to enable tracking of results from strain construction and screening through fermentation and protein purification.
We implemented the Core LIMS to organize our in-house data. Our product development capability relies on very high-throughput screening. We are generating a thousand or more production strains for every experiment, which results in several thousand samples to analyze. We need to track the construction of each one of those strains and link the associated data. This information includes: the genotype and lineage of the strain, and information about the protein expressed; the quality of the protein expressed in that strain; which strains were selected to move forward into scale-up; and the data observed with the strains that were scaled up.
We closely examine a subset of strains before choosing a production strain. Once we select the production strain, we optimize the fermentation so we can increase our titer of soluble, active protein and then develop a scalable purification process which can be implemented for cGMP manufacturing. All of these development efforts are underpinned by an extensive analytical capability that continuously generates large amounts of data for ongoing assessment of biosimilarity of our product candidate to the reference drug product. The Core LIMS is invaluable in cataloging this data to ensure ease of retrieval and efficiency of assessment.
One project requires a tremendous amount of data collection and analysis – but when there are five simultaneous experiments, manually keeping track of the data becomes challenging. Before the LIMS, finding information on these projects could entail several hours of sifting through folders to locate the right data. Now, with all data in the LIMS, it is possible for a single user to immediately access the information they need for a particular product candidate, rather than having to search for it.
Even though our scientific team works very closely together, we foresaw how the LIMS would make it simpler to track samples and data. With the barcode as our guide, we can now track location, origin, fermentation type, conditions, genetic makeup of host strain and more. It is possible to quickly compare analytical results during process development, and know whether the differences between the samples may be due to different fermentation or processing conditions, differences in production host genotype, or perhaps due to a different expression strategy. The LIMS is used to track experiments, inventory our components and hazardous chemicals, and automatically calculate key information generated. Overall the LIMS dramatically increases the efficiency of our product development efforts which directly translates to rapid, low cost biosimilar development.
When implementing any new solution one must establish a bridge between the old approach and the new solution to enhance adoption. When the LIMS was set up, the automated naming function was different from the method our scientists had been using, so we developed an alias attribute. Initially we created a system which would allow our traditional experiment-based naming convention to be connected to the new strain number and barcode. Now our researchers don’t need the old names, they just use the strain barcode. We assumed this would happen over time, but were pleasantly surprised that the adoption progressed relatively quickly. We believe the naming alias was instrumental in speeding the user adoption process.
The LIMS is also useful when working with development partners. The naming alias has helped us keep track of samples as we bring them in from other third party sites. We assign the sample a new name in-house, but the barcoding and the LIMS allows us to easily correlate the two sample names.
This has been a staged LIMS implementation. We are still configuring the system and continually coming up with new uses for the LIMS to further enhance the efficiency of our scientific team. We have already observed significant time savings across our entire product development efforts. Researchers are finding it easier to determine what samples they have, and to keep track of all the different components used in their experiments.
Diane Retallack, Ph.D., Director of Molecular Biology atPfenexwill discussPfenex Expression Technologyduring herpresentation, Meeting the Need for Rapid Protein Expression and Development with a Cloud-Based Informatics System at PEGS 2015, May 4-8, in Boston.