No More “What Ifs” - How Nektar Takes the Guesswork Out of Scaling Bioprocesses
Culture's client Nektar Therapeutics is exploring new PEG-based modifications to therapeutic molecules. We spoke with scientist Jonathan Mott about lessons he's learned in scaling up processes for therapeutic proteins. Read on for his insights into Nektar's unique process development approach, balancing deadlines with flexibility, and feeling confident in scaling a bioprocess.
Culture: Can you tell me about Nektar’s research pipeline and the types of therapies you create?
Jonathan: Nektar is one of the world leaders in PEG manufacturing, science and technology. For a lot of people, when they hear “pegylated molecules,” the first thing they think of is half-life extension. That’s certainly one thing that PEG does really well. At Nektar, we’re exploring new and exciting PEG-based modifications to therapeutic molecules that really change their reactivity and behavior.
For example, we can: use different sizes and lengths of PEG, use different geometries (like branched or linear), use stable or releasable linkers when we perform the conjugation, or target different sites that might block or leave exposed different binding interfaces. When we bring all of those modifications together, they can cause drastically different responses in the body. A great example of this is our BEMPEG and our NKTR-358 molecules. Both of them use the same IL-2 protein. The only difference is the pegylation strategy, but it completely changes how the body responds. These two different molecules end up opening different signaling pathways, which makes them work for different therapeutic indications even though they have the same IL-2 backbone. It’s all about the PEG – that really drives those differences.
Culture: Does this unique drug development strategy necessitate a unique bioprocess development approach?
Jonathan: Our approach is a little different, mainly as it relates to the complexity of our drug substance manufacturing. For many biologics, you would first do upstream processing and then downstream purification to release the molecule, and you have your product. For Nektar, that’s just the first step; in parallel, we’re manufacturing PEG and a PEG reagent, and that involves a lot of tightly controlled small molecule reaction steps. So, there’s manufacturing, purification and release of the PEG and the PEG reagents - and we’re still not done yet. Now we bring these two released intermediates together and we actually perform the pegylation, and then we do another purification. After all that, we finally have our drug substance.
To make this complex process possible, there are a lot of different stakeholders from all the different pieces that need to come together. This requires that with every processing decision we make, we need to check in with all of these areas; we need to adapt to what the other areas are doing and make sure we’re creating one cohesive and robust process.
Culture: Are all of your therapeutics produced through biomanufacturing?
Jonathan: We have some molecules that are small molecules – these are chemically synthesized and then conjugated – and then we have a lot of biotherapeutics that are protein-based, and that’s what my group is focused on right now. An exciting challenge for us is the diversity of proteins we work with. Nektar as a company is really focused on where the PEG takes us. While there are a lot of companies that develop portfolios around a certain protein or a class of proteins, at Nektar we find proteins for molecules that would benefit from pegylation, so the proteins that my group works on are all very different.
While we’ve dabbled in trying to make some kind of a platform process, what we’re finding is that these proteins we work on are so different that we’re basically starting from scratch every time we develop a new program. There are some benefits to that; the fact that we’re not handcuffed to a platform process means that we can explore and find the optimal expression strategy for each particular protein. It also means that there’s a lot of development work that has to happen for each new molecule.
Culture: What kinds of challenges do you face with this approach?
Jonathan: Since each program requires a lot of new development work, the biggest challenge is managing resources across all of those programs as we work towards the next milestone for each one. This means that flexibility is really important for us. For a lot of our programs that are in early development, we’re in an exploration phase where we design an experiment and analyze the data to inform the next experiment. This requires fast turnaround of the data and the ability to change our experimental conditions rapidly. When we can make that happen, it enables us to change direction based on what we see in the data, which leads to the development of better processes.
Culture: What role does Culture Biosciences play in this work?
The guaranteed capacity subscription that we have with Culture provides the flexibility we need to achieve our goals. We know we have bioreactor runs booked at Culture every month, but we don’t need to commit to the specific run conditions. Culture makes it easy for us to quickly change the parameters of an existing process, or to implement a new or different process. I’ve been impressed by Culture’s ability to be so flexible – it is a key part of our partnership.
The biggest value of Culture, though, is having access to more bioreactor capacity: if we can do more runs, we can learn more. Running more bioreactors allows us to do more screening studies, do more robustness DoE studies, and it really supplements our internal capabilities.
Culture: You mentioned DoE – did you know Nektar has the record at Culture for running the largest DoE with the largest number of reactors in parallel? Can you talk about what you were testing there? What were the goals of the study and what did you learn?
Jonathan: That was a fun DoE for me to design and analyze; it’s rare you get such a huge dataset like that to dig into. In that program, we were early in development and we got lucky in that the first process we tried worked surprisingly well. We were confident that if it ran at the center point, it was going to be great, but we didn’t have any development data – didn’t have any robustness data – on what happens outside of those center points. And it was a little worrying…what if we were on the edge of something? What if something went wrong?
We were on a tight timeline to get this process transferred to manufacturing, so we reached out to Culture to set up a large DoE that allowed us to look at many parameters at once, including some of the mid-risk ones that we otherwise would have had to eliminate to save time. Ultimately, the study gave us a great dataset to take to our CMO. We were able to say that we’ve looked at all these parameters, we’re confident that this is a robust process, it works at the center point, and it allows for a little wiggle room throughout the manufacturing process.
Culture: Could you anticipate this type of study being more of a routine practice?
Jonathan: I think there’s always pressure to just go with what works: as soon as you have something, you move forward with it. We’ve had processes where we were forced to take that approach based on timelines, but when we went back and studied it later, we found that if we had shifted parameters slightly, we could have increased titer and productivity. Or conversely, we’ve seen a big drop-off in titer if a parameter shifted slightly outside of the range we thought was safe. You have to balance the speed of development with the robustness. As a department, we’re trying to do as much parameter screening and robustness testing as early as we can. We recognize that sometimes there’s acceptable risk to move a process forward before that, but if we can, we want to run the tests.
Culture: You mentioned that it’s not often you get access to such a large dataset. Once you started working with Culture and were no longer limited by bioreactor capacity, how did that change the way you think about a DoE?
Jonathan: Having access to unlimited bioreactors has been great for our group to unlock our creative thinking. We can think about all these "what ifs." What if this parameter was important? What if we tried a fermentation with this strategy instead? Rather than playing guessing games, we can test it and have the data.
Beyond being able to test creative solutions, we also have confidence in our process when we go to transfer; when we show these large datasets to our CMOs, you can see them breathe a little easier because they know that this is a well-developed and robust process. It’s something we’re confident that we can run well because it’s resting on this solid foundation of data.
Culture: When you say your CMOs breathe easier, that’s because if there’s some deviation during manufacturing, they are prepared for how to respond because you’ve already tested this condition?
Jonathan: Right. Deviations are going to happen – it’s not a perfect world. When something doesn’t go according to plan, quality comes to us asking, “what was the impact?” Because we’ve already tested it, we have data to go back to, so we can say “we did these development studies, this is what we found, and this is our data-driven assessment” rather than having to rely on theoretical arguments.
Culture: Speaking of working with CMOs, how much of Nektar’s work is done internally vs. externally with other partners?
Jonathan: It changes over the course of a program. For the early phase in development and initial design, most of it is internal. As the program matures, we might do some larger screening studies externally. When we do our refinement and robustness phase, that’s where we’re pushing for higher yields and higher quality and we’re also testing out all those parameters. The timeline on this phase is variable – we put as much time as we can into it based on other milestones and how the program is doing. When the robustness phase comes, we look at our timeline, we look at all of the options at our disposal – both our internal capabilities and our external partners – and we choose which parameters will be studied at which site.
When we do the tech transfer, it’s all with the CMO at that point, but then we almost always come back and do either a second round of development or identify some focused studies we want to perform. We then look at the process questions we have and decide which site is most capable of answering those questions through the work they perform. Sometimes it’s Nektar, sometimes it’s a CMO and sometimes it’s another external partner, like Culture. It’s helpful to have options because there are different resource constraints, expertise, and equipment at each one, so usually one stands out as the best choice depending on the type of work.
Culture: When thinking about integrating with external partners, how does working with Culture compare to other CROs?
Jonathan: We like working with Culture whenever we need more bioreactor runs to answer our questions. One of the benefits I mentioned earlier is the flexibility – being able to reserve our bioreactors in Culture’s lab and then figure out what goes in them later. Culture has also built this great web interface that we like working on; it’s easy to set up runs and see the data from past runs.
Sometimes it’s hard working with other CROs who operate only via email; we have to dig through our inbox and find the report or excel spreadsheet they sent us a long time ago. With Culture, we don’t have to deal with that because all of our data is in the cloud and we can access it on your website.
Culture: In talking to other companies in biopharma, I always hear about the importance of timelines and the urgency of getting to market quickly. How do you balance the need for flexibility with the need to meet deadlines?
Jonathan: We always take a risk-based approach when determining how flexible we can be. We look at when the next milestone is and consider what we absolutely need to do beforehand versus what is “nice to have” versus what can be pushed off until later. We rank all of the research questions this way and then work through that list as we approach the milestone.
Working with Culture allows us to make more data-driven decisions. We can look at some of our medium or low-risk parameters and find out if they are critical parameters. Rather than guessing based on previous work, we can say with confidence that for a given parameter in this process, we tested it and it has a wider range that we can work in - it’s really robust. Or, maybe it tells us the parameter that we didn’t think was important actually is! If we know that earlier in development, we can find the optimal level and we can start testing the robustness of that parameter sooner.
Culture: Are there any lessons learned in developing or scaling up processes for therapeutic proteins that you can share with our readers?
Jonathan: When I talk to my colleagues in the industry, one of the most common things that comes up is compressed timelines and pressure to move a process forward even if it’s resting on limited data. If you think about the big picture, we’re developing these life-saving treatments for some terrible diseases, so anything we can do to move that forward faster is absolutely worth looking into. It’s always going to be about balancing speed and safety.
The way we manage that at Nektar is risk assessment. We try to quantify as best we can the risks that our process has, the risks the different parameters within that process have, and where we should focus our limited development time and energy. The best way to decrease risk and improve robustness is to do more runs and generate data. The small-scale fermenter technology Culture has is key for what we do because it allows us to get that data and perform those runs rapidly , at a price that is about the same as what it would cost us to do in house. With Culture, we can explore design spaces, prove robustness, and have data in hand when we move our manufacturing processes forward – and we can move them forward quickly and with confidence.
Culture: So, the lesson is, when in doubt, just put it in the bioreactor and run it?
Jonathan: Yes, do the test! No more “what-if” games. Let’s throw it in a reactor and see what happens.
Culture: During this time of COVID-19, other companies have told me their normal lab operations have been affected, and that their CROs and CMOs are backed up. How has having access to Culture throughout this time period impacted R&D at Nektar?
Jonathan: Our interactions with Culture really haven’t changed during COVID-19. Even before all the shutdowns, everything was mostly virtual with Culture because your web interface makes that easy. It feels like everything else in the industry has changed, but everything with Culture has gone very smoothly through all this.
Culture: Does having access to your bioprocess data in the cloud impact how you can collaborate with your team remotely?
Jonathan: I think it’s helped us with organization. We’ve tried on our end to take anything we get from the CRO and put it in our Nektar cloud storage. That way we don’t have to go inbox diving. But even so, there’s always the question of whether you have the latest version of the Excel spreadsheet. With Culture’s cloud service, we can always go directly back to the data rather than having to question what the latest update was on the Nektar side.
Culture: What’s next for Nektar, and how does Culture fit into that picture?
Jonathan: We have a robust pipeline of pegylated protein therapeutics, and I’m really excited about it and all of the new projects that are coming through my group. There are projects both in clinical and preclinical development, and we’re working hard on designing scalable manufacturing processes for all of them. I’m really looking forward to the future: all the different processes we’ll be able to develop, the experiments we’ll be able to run with Culture, and ultimately all the therapies we’ll bring to patients.