Start with the end in mind: how ‘cost of goods sold’ modelling can drive the right decisions
The 'cost of goods sold' (COGS) is an important measure to understand when considering biomanufacturing a product. Without this measure, the focus tends to move to ‘yield’ only, which although important, is not the only determining factor to the final cost of goods and can quickly reach a level that can’t surpassed. By understanding the full picture, expression systems can be modified in other ways such as varying temperature or pH tolerance, which can have a greater impact on the final COGS. This case study highlights how a COGS model can be applied effectively.
Since May 2022, Phenotypeca has been working with Junlebao Dairy Group to generate a COGS model for human milk oligosaccharides (HMOs) and lactoferrin expression in yeast. A COGS model is the quantification of costs associated with scale up and production scale. A model is built based on prior process knowledge, internal and external sources and operational expertise regarding commercially operating fermentation processing. This model simulates the process to highlight the key drivers of cost per amount of final product. Final yield will always be an important factor; however, there is more to the story than just yield.
Junlebao Dairy Group is the largest dairy processor in Hebei Province, China, a national leading enterprise of agricultural industrialisation, as well as a sub-centre of national dairy R&D technology. Junlebao has established a business paradigm covering the whole dairy industry chain to ensure the coordinated development of upstream and downstream sectors, and to provide consumers with nutritious, healthy, and safe dairy products.
Phenotypeca claims it has the world’s largest unique collection of yeast strains bred and engineered for stable recombinant protein production. Coupled with a detailed and sophisticated COGS model, Phenotypeca can gain an early understanding of the order of magnitude and the interrelationships of the cost elements associated with the commercial production of recombinant products. In addition, the COGS analysis highlights the challenges associated with the development and scale up of commercially viable organisms and processes. This allows Phenotypeca to inform clients/partners early as to the commerciality of a process or product.
Objectives
To understand the order of magnitude and cost relationships of a product and associated ‘process COGS’ and ‘operational COGS’ prior to manufacture.
Process COGS are costs associated with and influenced by the process. They are primarily fixed costs with some variable cost elements. Operational COGS are costs associated with operating a commercial process. They are primarily variable costs.
Approach
The COGS model was initially setup by Phenotypeca’s in-house COGS expert to consider a range of processes and operational inputs including: fermentation titres and growth rates; operational elements e.g., equipment turnaround times; and raw material, operational, investment and utility costs.
The results from the analysis are demonstrated in Figure 1 (A and B).
Following the COGS calculation, a sensitivity analysis was conducted to understand the operational specific productivity. Together with the firm’s modelling partner BrydenWoods, Phenotypeca performed Monte Carlo analysis simulations. This helped it to understand which variables impacted the COGS and fermenter productivity, giving an indication as to where most cost savings could be made. The simulations were carried out on key variables/inputs over pre-defined ranges, as evidenced in Figure 2.
Phenotypeca successfully identified areas where Junlebao could make significant cost savings, not only in its process but also in its operational procedure. Junlebao was pleased with the results and subsequently signed up to a PhenoStart feasibility study. This Phenotypeca study will produce Research & Development strains of next generation yeast for the future manufacture of Junlebao HMOs.
Results
The benefits of COGS modelling for strain and process development include, but are not limited to, aiding the development team in setting targets and parameters with the understanding of potential cost impacts and benefits of a range of variables:
• Secretion levels of desired product (titre)
• Growth rate (feed time)
• Media constituents (media recipes)
• Plasmid stability (fill & draw campaign length/continuous fermentation options)
Conclusion
COGS modelling continually focuses the development team on the ultimate target of a commercially viable and competitive industrial scale strain and process. In summary, COGS ensures development proceeds with the end cost and operational goals in mind.
Mike Fudge at Phenotypeca comments: “My mentor used to say, “start with the end in mind”, and this is so true when it comes to bioproduction. Often, I see businesses approaching production with hope over intent. Companies believing that if they get the fermenter yield as high as possible, they’ll be competitive. As a result, this leads to complicated processing or scale up issues. The whole process end to end needs to be understood, so we can put the effort in the right areas to achieve the desired goal with intent. This is exactly what we did with Junlebao Dairy Group. By generating a COGS model, we have been able to successfully identify specific areas where Junlebao Dairy Group could make their product at the right COGS for their market.”
