Aligning Upstream And Downstream Development (Part 2)

In Part 2 of this conversation on “Better Biopharma,” panelists from the Bioprocess Online Live event “Optimizing Process Development Through Upstream And Downstream Integration” continue their discussion about how upstream and downstream development teams can work together to accelerate timelines and reduce risk and bottlenecks in process development.

Keep up with everything biomanufacturing at Bioprocess Online: https://www.bioprocessonline.com/

Follow Tyler Menichiello on LinkedIn: https://www.linkedin.com/in/tmenichiello/

Subscribe to our newsletters: https://www.bioprocessonline.com/user/edit/subscriptions

Tell us how we can improve. Take our 60-second survey: https://docs.google.com/forms/d/e/1FAIpQLSfZGJluCmEhm1MBQvcm7_gtBEXuTx5v42LJ1mx9bCUKo8fwfg/viewform?usp=header

Transcript

Copy Transcript Copied

SPEAKER_04 0:16

Hello and welcome back to Better Biopharma, the official podcast of Bioprocess Online. I'm your host, Tyler Minachello, and this is part two of a conversation with our panelists from the Bioprocess Online live event, optimizing process development through upstream and downstream integration. If you missed part one of the conversation, I'd recommend going and listening to the last episode before continuing along. If you'd like to watch the full live event recording, you can head on over to bioprocessonline.com. With all that said, I hope you enjoy the episode. You guys keep talking about process locks. This idea keeps coming up about process locks. And of course, we spoke about it during the event, but I wanted to kind of go through around the horn here with you all represent different modalities, more or less. I mean, we have Doug in the ADC world, we have Brandon in the in vivo carte, Eric in the vaccine space, and and Mark, I think you've seen it all, right? So if we think about the the hurdles to clear, so to speak, to get to a process lock so upstream can can halt and give downstream something to work with, right? In your respective spaces, what what are your how do you define those I guess parameters or or the what are those goals? Like what does that make sense?

SPEAKER_02 1:37

I think there's two kinds of locks. There's a temporary lock that helps the team do its job in a way that they can understand what's happening. Um and I think it's important that people know that these locks, these early locks aren't permanent, they're just to help take some of the noise out of the program. And then there's a different kind of lock that has regulatory implications. Um and just to be clear, I've really only been talking about the first one, the one that helps the the one that helps the different parts of the team not confuse each other.

SPEAKER_04 2:10

Yeah.

unknown 2:11

Yeah.

SPEAKER_04 2:11

So reaching getting to that point, like like you said, Mark, so we can just give teams a chance to synchronize in development. What do you think are the are the big hurdles to clear to get to that point? Um, and how does that change my modality in in your spaces, I guess, respectively, if we want to go around here and share.

SPEAKER_02 2:33

God, it's it's different in every place but that I've been, but you know, to the extent that you can get people into the same conversation and and find a compromise that sort of is okay for everybody. Um and then actually do what you said you were gonna do. Which is is is is often one of the fail uh it's often one of the fail mechanisms that everybody agrees and then they don't do it.

SPEAKER_00 2:57

Yeah. I think that is a good idea. I could elaborate a little bit on uh oh, somebody you're talking about. Go ahead, Doctor. Oh, okay. Um uh you know what to define what a lock is, and I know uh that was a good example of like there's temporary locks. Um, and I think the consensus that most people think is by the time you're making your talks test article, and typically lead lock stability, um, it should be a lock process, which which is ideal, but not always completely necessary. Uh you know, in a go fast sort of time, uh, you're you're always gonna probably make a few extra changes because scale up might just require it, you know, and so, or you might learn something from your medium-scale toxin Lelot batch that you maybe weren't expecting and you have to address it. And so, you know, and the same thing with like, do you need to have final clone, for example, for antibody by the time you get to talks? Not necessarily, right? You could do a mix of clones, potentially, even still be at a transfectant pool, potentially, right? As long as as you're if you're trying to get into a talk study, which is which is the gating study as fast as possible, then I think there's a regulatory justification you can make for even going that early, and as long as you select a clone eventually that has the same profile and product quality as what your pool or your mix of clones had. And so it's kind of you have to get everyone on board, including internal regulatory, and have these kinds of discussions when you when you're dealing with a timeline, you know, that is pretty tight, which is usually the case. You everybody needs to be involved in and it's kind of like support each other and determine what risks are worth it and what would just, you know, you're not ready to take yet.

SPEAKER_02 4:46

You know, this is a little bit easier in the United States than it is in other, you know, many other jurisdictions. Um, I think you know, there's not very many countries who put the small letter C in front of GMP, and and and that's kind of a friendly aspect of the regulatory system we have here in the United States is the agency does assume that you're gonna keep improving at least early on.

SPEAKER_04 5:13

Yeah, I agree. Thank you guys. Brandon, were you gonna say something before?

SPEAKER_01 5:18

Yeah, there is a big Doug hit on a lot of it. There's a big communication aspect around the process lock. What is the purpose? What is the intent? Aligning on what that means for different functions, right? Uh, one another term that can be very complicated for people to uh understand or mean, or maybe there's many different interpretations of is what do we mean by having a platform, right? That means different things to different people. So if you are clear on what that means in your context for your situation, process log, platform, etc., that's super helpful, you know, managing expectations basically. Um so uh I think another uh key difference you asked about modalities. Um, maybe you could group them into two or three buckets. Um, if you have a well-known modality with a platform process, process lock is gonna look a lot different than if you're working with a new modality and you're working to establish your platform for the first time. There's gonna be a much uh different dynamic there. Maybe you need more of those interim locks when you're trying to establish a new platform for a new modality. Um, your analytical on a new modality, your analytical is probably your team is probably learning just as much as the process team, right? So there's a bit more likelihood of a moving target. So setting the expectations in those situations that we're probably going to go through multiple locks for multiple different purposes to get to where we want to go, versus where we're doing the 20th product in this platform, process lock should we could probably lock it at a priori before we even did any experiments to an earlier point. So it's all it's all super contextual.

SPEAKER_03 7:02

Yeah. I think one point to kind of go off just to summarize everything here is really there's a risk continuum and the understanding of risk and when to apply a lock in the process changes from stakeholder to stakeholder, whether you're in operations, your your MSAT function, quality, regulatory. So when you get to these conversations of a potential process change, removing or adding a process lock, define your rules of engagement as to how you interpret risk within your organization. Make it a guideline and don't make it a standardization to allow for ability to interpret and reinterpret. And make sure you have all those stakeholders there at that kind of key decision point to align and agree upon. Because the amount of times they've seen process lock or risk assessment that has had to go through multiple avenues of communication and go up and down the corporate ladder, it's probably one of the biggest uh aspects of bureaucracy that we deal with within the industry of really understanding and and ensuring that that risk and that lock is uh is clear to everyone. So define your rules of engagement, get all the stakeholders the right time, right place, and and and stick to a consistent plan. Yeah.

SPEAKER_02 8:20

Rules help.

SPEAKER_03 8:22

They really do.

SPEAKER_02 8:23

You know, um, we're talking about locks. One of the things um that you know, I I sometimes with some clients, you know, no problem, other clients, severe problem, is we're talking about, you know, a temporary lock for upstream to help downstream, temporary lock for downstream to help drug product. Um, but I've been in places where uh analytic method development doesn't lock. And you know, and that gives everybody a hard time. And and they also a lot of times they're kind of coming on board a little later than they should have, a little under-resourced and over taxed. And um they do have to improve their process as they go, their analytic methods. But also, there is a need to be able to have the same assay run on your different samples so you can tell what's happening. And and I think in some places I've been, that's done extremely well. Even if even locking into a very suboptimal method is still a lock and it helps. Um, but I've been in other places where it there isn't a lock and it adds a lot of confusion.

SPEAKER_04 9:32

That's something I didn't even think about. Is the where analytics comes into all this.

SPEAKER_02 9:36

I mean, we spoke about it during the event, but I didn't even think for a second about their own development and how they have to lock as well to keep everything on the same and at the same time, they have to give everybody visibility into what's going on in their labs, right? They they're they're they're not in an easy situation, they have to perform consistently while improving. Uh a little oxygen product.

SPEAKER_00 10:01

And Mark, you mentioned, I'm glad you brought up the analytical part about when when these analytical steps are considered fit for purpose or ready, ready to be locked. I've seen working with a variety of CDMOs that their internal systems at various CDMOs might might just be so rigid that they need fully qualified assays to release the talks test article, for example, which shouldn't really be the case, right? This is still development. It for GMP, definitely, but for TOX, and so I've seen delayed test article because the testing just wasn't ready. They had to finish because that was just their internal procedure, right? And then and there was no way around their quality group getting past that. And so those are things that are important to know ahead of time, you know, when you're trying to negotiate these master agreements and things, but you can't always think of everything, you know, it's difficult.

SPEAKER_02 10:57

And I think people don't agree on the words, right? Uh, phase, appropriate validation, qualification. Um, you can have a lot of different opinions about these common words and what they mean.

SPEAKER_00 11:10

Yeah, exactly.

SPEAKER_04 11:12

Yeah. That's a great points, guys. Thank you. Thank you, thank you. Well, we are getting close on time here. So I wanted to ask you guys the question that I ask all my guests on the show. And Eric has already answered this question before, so I hope he has a new answer prepared this time around. But that question is how do you think we can better biopharma? And Doug, we can start with you since you're on top of my screen. We'll work our way down.

SPEAKER_00 11:40

Oh, sure. Um and kind of to maybe to resummarize a little bit what we've already talked about is um, in my opinion, stop redeveloping platform processes, you know, get to a consensus internally with communication among all the groups of what's good enough for fit for purpose for the stage of development, you know, and and and that's doing various things like we've already talked about, using clone pools or even um just pools for getting to certain material deliverables, um, and and then have some predefined sort of um acceptance criteria and some metrics that are applicable to the overall program versus each individual step of the program, um, is is one way to I think to do it better. And I know most of my answers are around faster, because that's sort of where I am right now in in my current role. Um but I understand faster it means riskier, you know, at times. And so um, but I but I'm still a believer that the at least in the US, and I know uh FDA is very collaborative um and and partner, you know, they're they're like a partner. You can submit an IND and get it through with flying colors, but then you know what's the risk? A month or two later, you're trying to submit an IMPD and you get like the thousand questions because you didn't put as much stuff in there, you know. So it's risky in that way. Um, and so engaging health authorities early, like if you're proposing some kind of unique control strategy or or a continuous process, um, don't wait until you think everything's perfect to engage a health authority in some kind of a pre-IND meeting, for example. Try to do that earlier so that you can you can have them as a partner in less of just judging the work that you've put to you've worked so hard to put together and hoping for the best.

SPEAKER_04 13:34

That's great. Thank you, Doug. Brandon?

SPEAKER_01 13:38

I would say uh point I've made a couple of times, making sure your work and your goals are connected to the business drivers, right? Being clear on what those business drivers are and what risks you're willing to take, aligning on the risk levels that you're comfortable with as an organization to achieve those goals. Whether in Doug's example, it's speed, um, maybe it's getting into a new jurisdiction, maybe it's improving COGS, um, maybe it's other drivers, but making sure that your goals aren't too separated from that, um, and then um being clear on the risk amount of risk we're willing to take to achieve those goals.

SPEAKER_04 14:20

Great, Brandon. Thank you. Eric, part two. How can we better buy a pharma?

SPEAKER_03 14:25

Part two, and I think in the spirit of this conversation being uh creating more effective upstream, downstream teams, it's sharing. So it's about sharing your resources appropriately. It's about sharing those experiences and creating those cross-collaborative experiences between your upstream and downstream teams. A great point that was brought up earlier is sharing data, understanding the um what that data means and having that marriage between discrete and continuous data sets to really secure and define your design design space appropriately. And ultimately share, share your successes, share your metrics of of what success looks like because a team that's able to share better will perform better.

SPEAKER_04 15:10

Great, Eric. Thank you. And Mark?

SPEAKER_02 15:13

Well, I don't think I can add anything that uh that my colleagues here uh haven't already said with respect to how the team works. So I'll pivot a little bit and talk about governance. And you know, we work a lot with um, you know, companies that are you know series A, series B kind of funding. And I think from a governance standpoint, um if if the leadership of the company, you know, after Series A is really thinking hard, why do the series B people want to come on board? And you know, and then further downstream, series B, right? Why why is the market going to invest in us? Why is pharma going to invest in us? I think if you get a good answer to these questions, you tend to do your CMC a lot better. And that brings upstream and downstream and AD and drug product along for the ride. And it it helps them get onto the same page. Um, yeah, I know that's kind of a you know a bean counter way to answer the question, but um, I think I think money or lack thereof creates focus. Mostly the lack thereof.

SPEAKER_04 16:25

Yeah, right. Well, thank you, Mark. Thank you, Doug, Brandon, Eric, Mark, for being here on this episode of Better Biopharma, the official podcast of Bioprocess Online. And thank you out there for tuning in. We'll see you next time.