October 15, 2018 (Vol. 38, No. 18)
K. John John Morrow Jr. Ph.D. President Newport Biotech
Searching for Innovative, Cost-Effective Product Options
Biologic drugs, which are complex proteins manufactured by cultured cell lines, are coming to dominate the pharma market, now representing 25% of the total. By 2016, biologic drug sales had already reached $232 billion/year. There is no sign that the trend is slowing.
Such dramatic sales figures are a reflection of the fact that in many ways biologic drugs are more effective therapeutically than small-molecule pharmaceuticals. Because of their elaborate 3D conformations, biologics can interact with difficult targets that small molecules may be unable to reach. And their greater efficacy may mean that they can command a higher price.
But as Spider-Man reminds us, “With great power comes great responsibility,” and there is concern within the industry that costs are growing at an unsustainable rate. Thus, bioprocessing scientists are searching for innovative, cost-effective production options, and a number of experts addressed this issue at Cambridge HealthTech Institute’s Bioprocessing Summit, an event held recently in Boston.
Economic Drivers of Bioprocessing Technologies
Without a systematic approach that accounts for the entire range of variables and their synergies, an investigator may fail to order the relevant factors and determine critical interactions between them. At the event, this issue was addressed by Andrew Sinclair, CEO of Biopharm Services, who emphasized the economic drivers of bioprocessing.
“If we go back 30 or 40 years, pharma industry manufacturing costs were around 3–4% of the overall cost of goods (COG); today this proportion has increased to 10–20%,” Sinclair said. “So management of production costs is critical today.
“Moreover, there is a trend toward multiproduct manufacturing and more flexible facilities. Some facilities perform well, and some perform badly. Why is this? We believe that analysis of economic drivers can provide important insights. Given the strong commitment in the industry for flexible multiproduct facilities, COG models determine the value of novel new technologies. Modeling these allows our BioSolve Process users to adjust the parameters of the model to determine sensitivities to the value added.”
As Sinclair elaborated, “When comparing technologies, it’s important that you compare each with an optimum process configuration. For example, consider fed-batch processes in multi-bioreactor set ups. In the model, when you change the number of bioreactors that are pooled together to make a harvest, the outcomes will be different, demonstrating that there are a number of ways to make the product.”
And choosing among these options can have a profound impact on COG, by as much as 20–30%. These include concentrated N-1 seed, perfusion, and fed batch. But the task is not as straightforward as it might appear. “We find that upstream and downstream processing have to be considered as an integrated whole,” Sinclair continued. “What may improve performance at the upstream end may have negative consequences overall if you don’t integrate with the downstream end.”
To demonstrate the success of his modeling technology, Sinclair cited the case of a client who, working at lab scale, was able to correct shortcomings and avoid an expensive failure at production scale.
Continuous Production for Better Quality at Lower Cost
According to Thomas Müller-Späthe, Ph.D., a senior scientist at the Institute for Chemical and Bioengineering at ETH Zürich (the Swiss Federal Institute of Technology), “Continuous chromatography is a major facilitator of cost and time savings in biotherapeutic production.”
In a collaborative agreement with Bristol-Myers Squibb, Dr. Müller-Späthe and his colleagues investigated the use of twin column chromatography in conjunction with traditional upstream cell cultivation. “Our overall goal is to improve continuous production in order to generate a better quality product at a lower cost,” he said.
In this version of a processing train, two or more column chromatography processes can be joined in series for protein purification of antibodies or other pharmaceutical proteins generated by either fed batch or perfusion culture.
“We frequently encounter clients who don’t wish to abandon the fed-batch approach for cell cultivation of monoclonal antibodies, principally for economic reasons,” said Dr. Müller-Späthe. Perfusion culture requires a different fermentation setup and uses larger volumes of media, which can add significantly to the overall cost burden. “But in these instances, it is possible to combine a noncontinuous (fed-batch) process on the upstream end and couple it to a continuous process on the downstream side. This is a workable strategy because the continuous downstream flow will perform independently of the load on the upstream side.”
Multivariate Calibration Advances Bioprocessing Design
Nicholas Trunfio, Ph.D., a research scientist at Sartorius elaborates on the use of multivariate calibration in cell culture monitoring, in which investigators follow the production and transformation of metabolites in order to optimize upstream processing performance.
“Multivariate calibration is a specific tool, and is basically a framework to manipulate data,” Dr. Trunfio explained. “It is an approach to precisely monitor variables, such as pH, dissolved CO2, glucose, and lactose production in cell cultures by creating a multidimensional model. These data are frequently generated by NMR or mass spectrometry. By integrating massive blocks of data through systems analysis, it allows modeling without having to actually build physical systems, saving time and money.”
According to Dr. Trunfio, because there are so many unknown relationships, investigators need a tool describing how these variables interact. Multivariate analysis allows the investigator to process this information, applying least-squares regression, cluster analysis, and other mathematical methods to generate data of metabolite levels quickly and cheaply in a dynamic cell culture situation. To do this, investigators use the best historical performance record as their touchstone, or basic set of observations. It is referred to as the “golden batch.”
Golden batch monitoring ensures that product quality will be retained. The FDA endorses the quality by design (QbD) concept, which is a set of predefined objectives that emphasize product and process understanding and process control. This concept fits well with the golden batch criterion.
“We can verify our model by looking at the relationships between our measurements,” Dr. Trunfio noted. “For instance, as the cell consumes glucose, it produces lactate, and our models exploit that structure, showing, for example, that as lactate measurements change from the historical values, we can ask what caused the change, and a signal will be produced.”
Dr. Trunfio said that there are competing methods, but he argued that the multivariate calibration models are superior: “There are other alternatives, but we seek a middle ground between knowing everything that is happening in our system and knowing nothing. We believe that the multivariate calibration model manages these tasks very well.”
He emphasized that successful application of multivariate calibration requires collaboration between biologists and mathematicians. “When you communicate sophisticated mathematical tools, there is always a frustration on both sides, because of the necessity of having to explain these concepts in biological terms. While models are very useful tools, they must be grounded in biological reality, otherwise when applied, they may vary far off base,” Dr. Trunfio concluded.
Biting into the Cost Rate of Production
Capital expenditure (CAPEX) is a driving obsession in the biotechnology industry. “If we can increase our rate of production of biologics, then we will require less holding capacity which directly impacts the CAPEX,” said Samir Varma, head of manufacturing and general manager at Enzene Biosciences. To achieve this goal, Varma’s favored approach is continuous processing technology.
Although fed-batch technology was for many years the platform of choice, in recent years, the continuous approach has been combined with innovative continuous chromatography technology, such as twin-column chromatography.
According to Varma, continuous bioprocessing allows up to a 10-fold reduction in bioreactor size over the fed-batch option, so a 2000-L bioreactor can be scaled down to 200 L. “The biggest advantage of continuous bioprocessing is that you don’t need a holding tank,” he stated. “We find that CAPEXes quickly add up as additional complications to the process arise.”
Reducing complexity is dramatically illustrated when converting a fed-batch operation to continuous perfusion. “Because we can downscale our holding capacity, we can cut the dedicated plant area for that operation from 3000–4000 square feet to one-fifth of that area, or 600–800 square feet,” Varma affirmed. And with continuous processing, Protein A columns can be smaller. “All of these alterations, when combined, can bring down costs substantially.” Under these conditions, the facility can generate 300 L/day, which can then be continuously processed to the final purified biologic product.
This means that Enzene can advance its program of making high-tech therapies available to the vast Indian middle-income market. With these cost-saving improvements, Varma anticipates expenditures for a biologic treatment protocol could be brought down to 40% of the innovator’s cost. But Varma believes that over time it would be possible to achieve even more dramatic savings, down to 10% of the innovator’s cost, which would make such treatments widely obtainable.
Currently the market for biologics is dominated by the U.S. at 50% and the EU with another 25%, leaving 25% for the rest of the world. Pursuing a low-cost strategy could alter the equation, making high-tech therapies available throughout the world. To advance this transformation, Varma noted, Enzene is building a cGMP plant that would have a fully integrated continuous bioprocess. “We have already completed pilot-scale work: a 50-L series of studies that laid the groundwork for the larger facility, now under construction.”