Caitlin Smith has a B.A. in biology from Reed College, a Ph.D. in neuroscience from Yale University, and completed postdoctoral work at the Vollum Institute.
Bioprocessing refers to the use of living cells to produce a product. It is used for manufacturing products like therapeutic biomolecules, for example, or antibodies that are commonly used in research. A typical bioprocess workflow consists of three main components: upstream processes involving media and cell growth, a bioreactor in which the cells grow and downstream processes involving harvest and purification of the product produced by the cells in the bioreactor. Thus far, workflows have succeeded mainly by using a division-of-labor plan, with each section of the process cared for separately.
However, recent changes in manufacturing practices may alter the structure of the bioprocess workflow. A current trend in bioprocessing is to build smaller, local manufacturing facilities for more targeted products. This is a sharp contrast to the days of the “blockbuster drug” phenomenon, when huge facilities were built to manufacture one blockbuster drug. According to Michael Phillips, research fellow in processing technologies at EMD Millipore, companies in today’s environment want to be able to produce smaller amounts of targeted drugs and the flexibility to make different molecules in the same facility, rather than devoting an entire facility to the production of one type of molecule.
But to make less product in a smaller facility, they have had to rethink their entire workflow. What factors are most important in optimizing the flow of product through these stages? And how can one build flexibility into the workflow to accommodate future growth?
Constantly evaluate the available technology
The production and marketing of biomolecules is a competitive industry, and each company needs to take full advantage of the newest technologies that will benefit them. But of course, they can’t use every new development that surfaces. “One of the most critical considerations continues to be when to implement new manufacturing technologies,” says Jonathan Royce, bioprocess marketing segment leader at GE Healthcare Life Sciences. “The pace of innovation in both upstream and downstream bioprocessing continues to increase, and this means that process developers need to be on top of an ever-changing portfolio of technology options.” Companies need to weigh multiple factors in deciding what the long-term value of a new technology may be, including costs of implementation and any changes in the product’s quality and time to market. Anticipating a company’s future bioprocessing requirements, and then finding today’s technology that will still satisfy those needs long term, is extremely challenging.
So how do companies do that? “Technology mapping makes a great first step,” says Royce. “Identify what technology exists in your portfolio and compare this to the anticipated future needs. Use scenario planning to look at different situations—where might bottlenecks occur in the future? Where are there technology gaps that are going to slow down the pipeline? This information can then be used to prioritize future technology evaluation.” The scope of such technology mapping can be a widespread “facility of the future” project, in which each step in the workflow is scrutinized, including alternatives. “This can be a beneficial process, but it is time- and manpower-consuming, and not all organizations will have the resources required to undertake such efforts,” says Royce. An alternative is a more targeted workflow evaluation that can still result in increased efficiency and flexibility. “Think of it as a ‘workflow renovation,’ rather than a ‘demolish and rebuild’ project,” Royce says.
Move from unit operations to process compression
Most bioprocess workflows function as a linear series of sequential events called unit operations. Often each unit operation is staffed by a team of scientists who work to optimize and maintain their unit functioning at its highest level, to support the workflow maximally. The production of one biomolecule might comprise 10 to 15 different unit operations, each optimized separately.
But with companies increasingly building smaller manufacturing facilities to produce smaller amounts of biomolecules, they are beginning to rethink their workflow designs. “There’s a big push to compress the workflow,” says Phillips. “Instead of, say, 15 steps, can we do it in eight to 10 steps?” In addition, there is less emphasis on optimizing individual steps and more emphasis on optimizing the effects of individual steps on one another. “What is the impact of the previous step on this one, and the effect of this step on the next one?” says Phillips. “Most people are doing unit operations now, but almost every major bioprocessor is doing this process compression and thinking more holistically than just unit operations.”
Using an integrated approach to evaluate the workflow is also called holistic process optimization, an extreme form of which is continuous processing, where all steps are linked to continuously produce product. An example is the production of antibodies from cells grown in perfusion bioreactors. As the cells grow, you continuously add fresh media and continuously remove “broth” that contains used media, cellular waste and the product produced by the cells—but not the cells themselves. “The benefit is that the material coming out of the bioreactor continuously doesn’t have any cells in it, so there’s no need for clarification,” says Phillips. “The clarification was already done in the bioreactor.”
Power tools for bioprocess optimization
New tools are emerging for optimizing bioprocessing, and they are increasing efficiency. One of the challenges in bioprocessing is integrating the bioreactor with the purification step. “What if we can make more product and minimize the amount of impurities?” says Phillips. “Especially the amount of impurities that are difficult to remove. If we can solve that in the bioreactor, then it makes the downstream easier, and it makes the whole process more efficient.” New tools that facilitate this type of process compression, in which each step performs multiple tasks in the workflow, and that are also easy to integrate would be extremely valuable.
An example of these power tools is multimodal chemistry, which fits two chemistries or modes of chemical interaction into one device. This enables researchers to do things in the workflow that they’ve never done before. So-called smart polymers, which are used in the clarification step that follows the bioreactor, illustrate how multimodal chemistries work. When the cell-culture suspension is removed from the bioreactor, one of the first things to do is remove the cells; with multimodal chemistries, you can do that—and more.
“What we’re seeing now are technologies … that do more than just remove your cells,” says Phillips. “There are precipitates you can add that precipitate the cells but that also precipitate a lot of impurities as well. The advantage … is [that] while it helps the clarification, it also really helps the downstream purification, because it removes these impurities that would be much more difficult to remove in the next steps.” Removing such impurities may also increase the lifetime of chromatography separation materials or the overall stability of the product. Thus, new tools like smart polymers can be used as a unit operation, but they actually facilitate process compression, because they do more than one job.
Also allowing greater flexibility in bioprocessing is the advent of disposable or single-use technologies. “Single-use technology continues to be an important part of the solution in terms of creating process flexibility,” says Royce. “Even best-in-class operators who have minimized changeover times cannot match the flexibility with steel systems that can be achieved with single-use platforms. The basic number of additional variables provided by single-use systems gives greater scope for tweaking and customizing manufacturing operations by varying process volumes, flow rates and production capacities.”
An example of this is the use of negative—or flow-through—chromatography in the purification step, in which the chromatographic media binds to the impurities while the product flows through in the eluent (rather than vice versa). In the production of a biomolecule, negative chromatography can be an advantage, because usually there is far more product than impurities. Thus the amount of solution required to bind and remove the impurities is far less than the amount that would be required to remove all the product. “And so the devices are smaller and lend themselves more to disposable solutions, which leads to more flexible manufacturing,” says Phillips.
Room to grow
These recent innovations in bioprocessing tools are just the beginning. Royce notes, for example, that bioanalytics remain challenging. “We need better analytical techniques that are faster and more aligned with high-throughput process development tools which can produce hundreds of samples in very short periods of time,” he says. “I think almost every organization I encounter identifies that analytical throughput is an issue.”
As awareness and acceptance of continuous processing grows, manufacturers will probably begin to re-examine the functioning of their current unit operations. This “is likely to drive further integration of the bioprocess workflow, resulting in processes which are leaner and more productive,” says Royce. Today, the predominant organization of most workflows is in units, or at least in groups focused on upstream or downstream processes. “There are logical historical reasons for this,” says Royce. “But those who will be able to extract the most value out of their development work are the ones who are finding ways to integrate these two parts of the organization so that the synergies of the entire bioprocess workflow can be fully realized.”
The image at the top of the page is from EMD Millipore.