Supplement: Visions of a Continuous Bioprocess Future

September 1, 2018 (Vol. 38, No. 15)

Challenges Remain in the Switch from Fed-Batch Biomanufacturing Operations

Considerations of cost, quality, speed, and flexibility are driving the bioprocessing industry to look for ever better ways of producing their wares, whether they be monoclonal antibodies (mAbs), recombinant proteins or extracellular vesicles. Incremental tweaks in how the processes are carried out are enough when the call is for 10-fold increases in productivity.

Wholesale changes, however evolutionary they are, will likely be necessary to achieve the goals of future labs. Among those changes is a switch from fed-batch to continuous processing—upstream, downstream, and perhaps in-between as well.

Last year, the BioPhorum Operations Group (BPOG), a consortium primarily of biomanufacturers, released its Technology Roadmap looking at where the industry is going and where it needs to be in the next ten years. They found typical market trends impacting the traditional business drivers, “which makes sense,” says Michael Phillips, Ph.D., director, next generation bioprocessing R&D, MilliporeSigma.

Yet, they were looking for a 10-fold change. That means that cost, including the cost to build a new facility and the cost to manufacture a particular molecule, needs to decrease by a factor of 10. Similarly, they were looking for ten-fold changes in quality, flexibility, and speed.

“It can’t take five years to build a new facility,” emphasizes Dr. Phillips, who adds that to accomplish a transformation of this magnitude, new paradigms, rather than incremental “step” changes, will be necessary.

Dr. Phillips is part of the NextBioPharmDSP collaboration, funded as part of the European Union’s Horizon 2020 initiative, and tasked with leveraging single-use technologies in a fully integrated continuous downstream manufacturing platform. As part of his presentation to the Cambridge Healthtech Institute’s Bioprocessing Summit in Boston last month, he compared the factory of today with visions of a factory of the future.

Large Infrastructure

Today’s factory could be a large facility that could take five or six years to build, he explains. It uses primarily stainless steel and is focused on unit-operations. There is a large infrastructure, including clean rooms and steam generators, “because everything has to be steamed in place.” There are large QA and QC departments. Everything is paper-based “as digital bioprocessing is still in its infancy.” And there is a big inventory area.

In contrast, factories in the future will be much smaller, much less capital intensive. They’ll be based on single-use technologies within a closed process, “meaning that everything is connected such that the process itself is protected from the environment, and the environment and the people are protected from the process,” continues Dr. Phillips. It will be intensified, and will use inline sensors to automatically and constantly report whether the product is within spec or not.

“You won’t have a QC lab anymore because as soon as you’re done processing it’s ready to be released,” according to Dr. Phillips. Because it’s fully interconnected or continuous, there will be much less labor involved, it can have much higher productivity, and everything can be smaller. And because it’s single use “you don’t need a facility for each particular molecule, you can run many different molecules in one facility”—one that can be built in a year because it doesn’t require a huge infrastructure, and it can support local manufacturing “which is a big trend that people are seeing.”

The transition to tomorrow’s factory will likely be evolutionary. People start with a batch process. They have a bottleneck that they need to overcome, say, in capture. “So they go to multi-column continuous capture because it’s more productive and will relieve their bottleneck. It’s still a batch process, but within their batch process they’re using an intensified unit operation,” explains Dr. Phillips. They next start to connect two things that make sense, and ultimately, they become comfortable connecting unit operations and go fully continuous.

The same thinking applies to format. It used to be stainless. “Today it’s a hybrid—a mixture of stainless steel where it makes sense and plastic where it makes sense. The future may be more single use,” points out Dr. Phillips, with the ultimate end state being a single-use fully closed system.

Regarding analytics, today you sample and send it to the QC lab. “The intermediate would be online sampling—you’re sampling, and your QC lab is now on the manufacturing floor. The future will be inline sensors,” he predicts.

And you can say the same for software, automation, and digitization. “They’re going to all go through an evolutionary approach. And not until all of them get to the final end-state will you be able to realize this pipedream,” says Dr. Phillips.

Match the Batch

Continuous bioprocessing begins upstream, which can present its own challenges—especially when switching an existing process from (fed) batch to perfusion culture.

Momenta Pharmaceuticals has “spent a lot of time optimizing feeds” of its biosimilar cell cultures to match the innovator drug and “we kind of know how to do that in the batch mode,” explains Elsie DiBella, Ph.D., vice-president, biological process and chemistry, manufacturing, and control (CMC) development.

“But when you’re in the perfusion mode you’re balancing cell growth and cell productivity all the time, and that makes it a little more challenging to get that glycosylation pattern where you want it.”

It turns out the cells in perfusion mode “are healthier—you’re always re-feeding them—and the glycosylation profiles are a little more complex than we needed them to be,” she says. By reducing the perfusion they were able to shift the cells to more of a protein production than a cell growth mode.

“You can take a fed batch process that’s operating at, for example, 20 million cells and turn it into a perfusion process operating at 80–100 million cells,” adds Dr. DiBella, who also was a speaker at last month’s bioprocessing summit. That can be done with the same cell line and with not a whole lot of work, “with a product similar to fed batch.”

Unpack the Column

Capture and polishing needn’t be done in batches, either. Instead of using packed beds, ChromaTan’s continuous countercurrent tangential chromatography (CCTC) system allows resin to flow through its continuous hollow fiber system.

“It recirculates the resin really rapidly through every step, and within every single step you have a countercurrent exchange of the buffer and the resin. That way you have binding, washing, elution, stripping, and equalization happening at the same time,” explains Dmitriy Fedorenko, the company’s associate director, process science.

CCTC can be integrated with any sort of process analytical technology (PAT) because of its ability to modify chromatography conditions mid-run, with a measurable rapid response. According to Fedorenko that “very rapid response within the system to any sort of change in buffer input that would typically necessitate a full cycle time to be completed within a batch operation.”

The company has been working on integrating CCTC with sampling devices, more fully automating the system, “and having it give you feedback and feed-forward controls,” says Fedorenko, who also presented at last month’s bioprocessing summit.

There is a big focus in the industry on developing these PATs. But “even if you develop a sensor that can give you a really rapid response, it doesn’t have the same utility if your system can’t respond rapidly to it,” he says.

Exosomes to Scale

Exosomes are vesicular packets emitted by cells, decorated with proteins and containing a payload, that can be picked up by other cells. Codiak BioSciences is “focused on using and delivering them as a therapeutic modality,” says Andrew Grube, principal research associate, upstream process development. But “the traditional approach for exosome manufacturing is using ultracentrifugation, [where] scalability is totally impractical.”

The company adapted an advanced perfusion process, using a chemically defined, protein-free media to grow the cells to high densities. This “enables us to squeeze a lot of high productivity from small bioreactors [and] process intensification is a major aspect of that,” notes Grube, who presented his work on developing the upstream aspects of the process at the bioprocessing summit in Boston last month.

The continuous aspect of the process allows retention of the cell mass while continuously harvesting the bioreactor, “which enables us to have many times the productivity of our fed-batch model,” he says. Another difference from manufacturing exosomes in a batch process: exosomes get resorbed (“eaten”) by the cells that produce them.

In perfusion there is a much shorter residence time “because you’re constantly feeding and removing media, and what happens is you take out the exosomes before they are being consumed. This changes the equilibrium in the bioreactor and results in much higher productivity per cell.”