Thermo Fisher Scientific Inc. (TMO) Earnings Call Transcript & Summary

Hello, and thank you for joining us today for this webinar, Late-Phase Biologics Development: Embracing Risk For Competitive Advantage, which is brought to you by Scrip and sponsored by Thermo Fisher Scientific. My name is Andrew Warmington. I'm the Manufacturing Editor at Citeline, and I'll be your moderator today. Before we jump into the presentation, let's go with some housekeeping items. First, if at any time you're having technical difficulties, click on the Test your Connection button, and this will take you to a connection checker and FAQs page. If you're experiencing any technical issues, there's also a Chat Now function to contact the support team for assistance. If you want to enlarge the presentation, just click the maximize icon or drag the lower right-hand corner of the presentation window to enlarge it. There will be Q&A at the end of the presentation, so if you have questions, we want to hear from you. To ask our presenters a question, simply type it into the question window on the bottom of your screen and then hit Submit. We'll be answering as many questions as possible during the Q&A session. And if there are any extras, we'll try to answer you after the webinar via e-mail. We'll also be conducting a short exit poll at the completion of today's webinar, so please stay around for 10 seconds after the end to complete it. Because we know you'll want to listen to this webinar again, today's session is being recorded and will be available to view on demand in about 48 hours' time. And now I'd like to introduce our speakers for today's session. First, Christy Eatmon is the Global Subject Matter Expert for Sterile Drug Products at Patheon, which is part of Thermo Fisher Scientific. Christy supports the global sales and business development teams in providing technical support, designing strategies and supporting new business opportunities for the sterile manufacturing business. She has more than 14 years of experience in the pharmaceutical industry with an emphasis on process engineering, product development, [indiscernible] manufacturing and filling. She has working knowledge of all phases from drug discovery to sterile product commercial manufacturing with expertise in the formulation of small and large molecules. Previously, Christy supported the site at Greenville, North Carolina as a senior principal scientist in the Pharmaceutical Development Services area. And Daniel Baskind is the Global CMC Lead for Biologics at Thermo Fisher. In this role, he oversees process validation and CMC operations for commercial programs across Patheon's global biologic drug substance manufacturing network. Prior to his current role, Daniel held multiple leadership roles at Thermo Fisher Scientific, including upstream and downstream process development, technical strategy and business development. The experience spans preclinical development through to commercial manufacturing across a wide variety of molecule classes. Before joining Thermo Fisher, Daniel worked in novel vaccine development at Novavax with time spent in both process development and manufacturing operations. He received both his bachelors and his master's degree at Cornell University. And with that, I'd like to hand things over. Christy and Daniel, the floor is yours.

Thanks so much, and it's such a pleasure to have the opportunity to speak. Good morning, good afternoon, and good evening, wherever you may happen to be. Thank you so much for the kind introduction. My name is Dan Baskind. I am the Global CMC Lead for the biologics drug substance unit within Thermo Fisher Scientific. Looking forward to having the opportunity to talk to you a bit about late-phase biologics development and how we view risk as an opportunity to create a competitive advantage. I'll take you through some of those details as it relates to drug substance, and then we'll hand that off to my colleague, Christy Eatmon, who will talk about it from the perspective of steriles. So in terms of today's agenda, we're first going to talk about what exactly do we mean by late-stage development? And what are some of the risks that occur in that space before we shift to focusing about some of the top risks and mitigations from a drug substance perspective, covering in kind for a direct product before talking about linking it into how we prepare for PPQ and commercialization. So jumping right in, let's talk about what we mean by late-stage development. And really, these are the development activities undertaken to design the manufacturing process that's planned for licensure. So we take a really high-level view of the sort of the key clinical milestones, right? That might be lead selection before moving into your first human trials or Phase I trials. And then ultimately, moving through Phase I, Phase II, Phase III to the point that you're able to file for licensure and launch. So we're really talking about the -- in late-stage space, it really incorporates everything from the end of Phase II meeting where that Phase III plan is identified through licensure. And a really key piece of this is ahead of your Phase III manufacturing, taking commercial process development activities so that you're ensuring there are minor changes that have to occur between the final process and tended at the licensing point and that's used to produce your pivotal study clinical material during Phase III. So some of the design features that we think are important when we talk about late-stage process development is one that is able to consistently achieve the required product quality and the process performance. This is a really key component, right? That robustness and the ability to be similar from batch to batch is certainly an expectation from regulators, but it's also important, just a reliable supply chain. On a similar note is ensuring that, that process is ready to be scaled to support whatever your commercial demand scenarios may be. And that may be different from year 1 versus year 5. And so thinking about, is this a scale-up challenge where you're looking at moving into increasingly large batch sizes and bioreactor sizes? Or is it a scale-out challenge, right, where you're thinking about running more batches per year to meet that commercial demand? And last but certainly not least is ensuring that, that process is ready for successful validation. And we'll get a little bit more into what we mean about that. But the intent is that the type of expectations for showing that a process is fully validated and ready for our commercialization are really different than an early phase. And that ends up being a significant portion of the investment of time and energy that goes into late phase development. So let's cover what some of the types of risk that you might run into when you're doing late phase development. The reality is that as programs enter late phase, they increase both in scope and complexity. And risk is a fundamental part of that. By embracing the presence of risk and focusing on managing that risk through both analysis and mitigation, there are opportunities to find -- excuse me, opportunities to find inefficiencies and increase speed that matches your specific program's goals. So the 3 areas of risk that we think about are technical risk, program risk and business risk. So diving into each of these a little bit at this level and we'll cover it more in the subsequent slides. When we talk about technical risk, thinking about how risk can be introduced through the availability of appropriate technical expertise, ensuring that the specific needs of your program are met by the experts that are available to support it, as well as the facility and equipment capabilities, where the manufacturing is being done but also from a process development perspective. Third is also the technology transfer process. The process of getting something that was created either in the research space, the discovery space and moving it into a process development lab or moving it from process development into manufacturing or perhaps from manufacturing facilities to manufacturing facility. There's lots of opportunities for things to be lost in translation or mismatched and inadvertently creating risk. There's also the specific means and approaches required for process characterization and validation that can create an area of risk and the suitability of the process to meet those needs as well as ultimately the commercial process control strategy, right? When you -- there are certain parameters that have been identified as critical for control, ensuring that your PCS puts you in a position to deliver consistency is a really important part of late-phase development and needs to be considered from a technical risk perspective. Now when we talk about program risk, really, there's 2 components that jump out. The first is your time line requirements. In terms of understanding the specific milestones that are needed for your program and is the scope that's required to get you there and match, as well as the regulatory strategy that your program is taking in supporting those requirements. Now shifting to think about business risk. Really, the 2 pieces here are, what's your funding strategy? And does your drug development program align with the capital available to invest into either the development and manufacturing activities? And are you getting enough of details around the technical requirements in order to meet the milestones that need to be delivered on for your investors? There's also the reality of asset development, where if your intent is to move this asset either by spinning off or moving into -- sold-off to another company or to achieve partnership, what are those external parties looking for in terms of the state of technical development for that asset? Those are the type of risks that also need to be considered. So the key takeaway here is, as you're looking at these areas of risk, it's really, really important that there's full transparency from both sides, meaning specifically the sponsor and the manufacturer as well as with regulators, right? There needs to be a really considered and thoughtful approach in how that communication is done and ensuring that risks are proactively flagged and that a strategy is developed proactively as opposed to reactively. So with that as background, we'll jump to some of the top risks and some thoughts on how you can mitigate them effectively. So the first thing it relates to is process materials that may not be appropriate for late-stage manufacturing. This is really just, if you're in a situation where your commercial process is relying on materials that are either hard to procure perhaps due to long lead time, maybe are highly variable because particularly, a key consideration is things like animal-derived materials, which is certainly more complex to handle than those that are chemically synthesized and defined. So thinking about how to mitigate some of those challenges, the first point that jumps out is using platform materials. But what we mean by this is materials for which there's existing knowledge and successful practices that can be leveraged in order to use those materials for a new program. This is something that your manufacturing group may have existing information, for example, on extractables and leachables approaches and how that's defined. Rather than having to invent the wheel new every single time, having the opportunity to look at existing practices puts you in a position not just to find efficiencies but to buy down risk as well. Second, you need to think about your supply chain strategy. In terms of ensuring that you have consistent supply, that you're not in a position where you're not able to finish a batch or complete a batch due to the absence of materials, thinking about working with a partner that has existing agreements that can support that supply chain can be really helpful. But there also may be instances where you have a critical material for your specific process and then a new agreement is needed for those critical materials. So just thinking proactively about the type of materials and what's specifically required to keep them in a position of security and inventory is a key piece of this. Another strategy that some of our customers have used is dual sourcing, right, where there are certain components that maybe are not as specialized to your process and just to mitigate against supply chain interruption, there's an opportunity to have 2 parts that are considered synonymous during the process and ultimately validated as such as well. Now in terms of addressing material variability as was brought up before, there's really 2 components that I think are important to consider, the first being mitigating that through supplier quality, especially with materials that are used in biologics processes. There is a significant quality infrastructure that exists on the supplier side to ensure that what leaves their door and is pulled into manufacturing for clinical material or commercial material does have appropriate quality standards. And in some instances, especially if you're looking at building out supply for a commercial program, you can discuss with your supplier your quality needs to ensure that, that variability is appropriately tight. In instances where that may be challenging to control, taking on the understanding of that variability through process characterization activities can be really credible. And there's a long history of this approach, where folks have done things like studied the ligand density on a resin where it's -- that has been soon turned to be a critical parameter. And using that as an opportunity to evaluate intake of materials to ensure that it's fit for purpose for your process can be really valuable as well for mitigating that risk. So now move on to our second risk, which is gaps in process and product understanding. Now this is a risk that comes up with some frequency for a variety of reasons: first, that it's actually very standard to see relatively lean approaches taken to early-stage process development really for 2 reasons: one, to accelerate speed to clinic. That can be a really, really important perspective, one, in terms of moving your program along as quickly as possible but ends up being a very common funding milestone as well. But the other part of this is that it may make sense to defer investment in products or process understanding until later stages when the path to commercialization is clearer. And so for a variety of reasons, it's often to see that we just don't have that much to go on with things that have only developed through those early stages. But there's also challenges that can emerge when it comes to intercompany knowledge management, right, especially when you're acquiring an asset that started outside of your company. We often find that we've had customers that have maybe less information available than they had originally hoped. And last but certainly not least in terms of areas that justify this risk, there's frequently misunderstanding around accelerated designations, specifically in terms of appreciating what is required for whether it be Fast Track or orphan drug in terms of reducing regulatory requirements. So in terms of mitigating some of these pieces, I think it really starts with quality by design. And we think about this in 2 stages, right, either [ perspective ] QBD or retrospective QBD. Now when we talk about perspective QBD, this is things that you can build in knowledge through platform approaches, right? It's very typical, especially in early development that when a company is using a molecule type that fits to an existing platform, they can use, for the most part, the same unit operations, the same raw materials. And that itself allows you to have a high degree of understanding and cross-supply knowledge from 1 molecule to the next. The other element of prospective QBD is doing early-stage design of experiments. Now DOE, maybe 5 or 10 years ago, would be something that was seen mostly as a tool to support process characterization. But in the modern era, the reality is that it's perhaps the most efficient way, in some instances, to develop process understanding and optimization even in the earliest stages. That knowledge is directly leverageable as you move into the later stages as well and may actually create some efficiencies and minimize the amount of time taken in those later stages. But that's a really solid bedrock for understanding the connection between your process parameters and the critical quality attributes necessary for your drug's efficacy. However, in instances where that information may not be readily available and you need to fill in the picture as you move to later stages, retrospective QBD is a totally appropriate approach. And really, the bread and butter here is using high-throughput process development. What we mean here is the opportunity to map the space very quickly and in parallel. Some of the key equipment that we like to use for things like this are the Ambr 250 for bioreactor upstream work as well as the RoboColumns and Tecan's systems for automated liquid handling the downstream space. So it allows you to cover an enormous amount of space in a short period of time and can build on your knowledge base for your program very quickly. The third point I'll mention is making sure that you focus on early identification of critical quality attributes. A very common mistake is really finally committing pen to paper about what the specific CQAs are for your molecule at the point that you're putting to light your filing together. The real thing that you want to make sure you avoid is finding new CQAs that require control after your Phase III process has been defined, mostly because, one, it creates challenges in terms of commercialization approach in terms of finding retrospective information that would have been easier if it was in place. But even beyond that, it creates perhaps a necessary risk as you move into your Phase III clinical trials. If there's a critical quality attribute that needs to be highly controlled and is not well understood by the time that you're ready to start your Phase III trials, it puts those trials at risk, and it's very challenging given the scope and expense of those trials to take that on the chin easily and move forward. So in terms of the third risk, let's talk about complex or nonrobust manufacturing processes, right? So when you're using a unique process that's maybe less typical than what's usually seen in the manufacturing space, this could be because you have a molecule type that's different than what's standard or perhaps because the process was developed in a research lab without manufacturability being the key point of focus. This introduces risk in a number of ways, right? Any time that a process is more complex or has more manual operations, the risk for deviations and ultimately for batch failure and manufacturing is higher, without a doubt. And so the simplification could be a really important priority as you're moving into later phases. This also can create challenges with process transfer. When your process is more complex, that means it's more difficult to describe, right? And there might be tribal knowledge that made it effective with the unit that it started with that are challenging as you transfer into another space. It also can create complexity with scale, right? Just because something is able to run reliably at bench scale, where it has the opportunity for greater tender loving care compared to what's achievable in a manufacturing environment, that certainly can be an area of risk as well and where you may have mismatches between what's achieved at scale versus what's on the bench. And lastly is just is having a -- really considering that compliance with modern regulatory expectations is a really, really key part of moving things through commercialization. And when you're working with like legacy programs where perhaps there were less robust technologies that were totally appropriate for use 20 years ago because they represented the modern standard, today, they may struggle to move through that same regulatory scrutiny. Some great examples from our clearance are things like Triton X-100 Detergent, which has moved to certain regulatory risks related to the environmental hazards. And this is something that you'll have a challenge introducing into processes in Europe, or virus filters, right, where the type of control that goes into manufacturing those filters today vastly exceeds those from several decades ago. So when you think about mitigation approaches, one is really making sure that you identify high-impact process improvements, right? And this doesn't mean that all components of your process are equally important to focus on for optimization, but certainly, replacing legacy materials. In some instances, it's not a significant development exercise, changing viral filters, yes, it may require additional viral clearance studies. But other than that is a relatively simple and straightforward development change that should not really pose any risk or introduce any need for clinical bridging. There's also a need to focus on manufacturability through simplification, right? If this was something that, for example, used grading elutions or used a number of buffers that upon second analysis maybe could be reduced in terms of the number required, that simplification can pay major dividends in terms of reliable manufacturing in the commercial space. There also needs to be an early evaluation of atypical or complex process steps, right? So in some instances, there are process steps that are just key to how your process performs. And it's not a matter of replacing it with something else, but making sure that you have your eyes open to what the validation risks may be for that complex process step and what the mitigations needed might be with time to address them far ahead of the validation campaign is a really, really important approach to take as well. So the fourth risk we want to highlight is if you have a process that's not capable of supporting your commercial supply needs. So this is really about low output per batch. This could be related to perhaps due to low volumetric productivity, which could be things like a low-titer cell line and your bioreactor or perhaps because your downstream yield, the amount recovered is relatively low compared to the amount of starting material. This also relates to scale out, right? So aside from just sort of the batch size that you're using, how many batches per year is another key consideration. So how we mitigate this is, one, thinking about making sure you have a high-performing cell line in place and this is something that's not just about titer. The amount of material produced per cell is obviously a very important consideration. But developing your cell line to deliver the required product quality for your materials, it's a really valuable resource, right? Rather than turning that into something that's being managed through process development in the bioreactor or in the downstream cascade, having a cell line that puts you in a position for success from the get-go is a really, really valuable approach to take. Also be considered in the purification unit operations that you're using. One of the ways that you can improve yield is by having fewer downstream steps, right? And especially with the availability of modern resins that allow you to either have completely custom affinity resins that bind to either a specific -- your specific product or to a specific impurity that needs to be removed, or by taking advantage of complex chemistries like mixed mode chromatography, you're able to have unique separations that sometimes can replace multiple steps and certainly allow you to have a more precise control of selecting your product from the impurities and having higher yields within that unit operation. The last approach is making sure that you align your scale with your commercial demand. This is something that needs to be considered not only in terms of your batch size, right, is this going to be a 500-liter reactor, a 2,000-liter, 5,000-liter, single-use bioreactor or even into the stainless steel space, whether that be 10,000 or 20,000. But also being considered the advantages of single-use bioreactors versus stainless steel. If you need to move to a 10,000-plus liters bioreactor, stainless steel might be the best choice for you, but the amount of capital investment and time that needs to go into setting up a facility to work with stainless steel should be a consideration, right, especially balanced against using single-use bioreactors that are a bit more nimble and where there's increasing volumes that create -- that minimize that gap between single-use and stainless steel. And the other piece of this is that batch cadence, right? If you're in a situation that you need to run 100 batches per year, that's going to be challenging to support with a single facility. So making sure that your batch sizes support the variety of commercial demand scenarios that may come out is something to really identify early and make sure that you're not looking at major changes in terms of batch size or style at the 1-yard line as you're moving into commercialization. Next risk I'll highlight is the misalignment under accelerated regulatory designations. And so we view this as a program risk, right? And the reality is just that when you're working under those accelerated designations, you may decide to take an atypical CMC or validation approach, but those approaches can create difficult to anticipate gaps, right? So understanding which components of your CMC and validation program are must-haves versus nice-to-haves versus those that you may defer as part of your licensing approach, it can be extremely product-specific as well as facility-specific. The other part is really just considering how you're communicating with regulators around the type of Fast Track -- excuse me, the particular designation that you're using and the -- what you're able to use as far as communication pathways, that's a really important part of managing that risk as well. So in terms of approaches here, really, it's about ensuring that there's really robust risk analysis taken on to support that atypical CMC approach. So making sure that, as you're considering, okay, we're removing this part or we're adding this part in, or these -- this is not an interdependency that we're using relative to the standard, thinking about that from a change management perspective and where it's introducing new risks and how they're mitigated. So ultimately, you're presenting what continues to be a holistic and defensible CMC plan. And the second part of this is communication, communication, communication, right? Not just with the regulators but with your manufacturing partner. Take advantage of those designation privileges. Often with those designations, there is an opportunity to communicate more frequently and to receive more contemporaneous responses from regulators to the type of inquiries that you have. Take the guessing game out of it, right? Have those conversations proactively and then make sure that's something that's really well defined and aligned with your manufacturer as well so that the scope that they're taking on, especially as it relates to validation, ultimately supports that bespoke approach taken for your program. So moving on to the sixth risk is really in terms of when is too long to have your capacity planning put together, and this is more business risk, right? The reality is that it's -- and it makes sense, right? A lot of people are reticent to sort of initiate discussions on the specific commercial manufacturing needs until they're really at the point of licensure, right? I mean, one is because the capital committed for that type of manufacturing can be significant as well as there may be still developing clarity on the amount of manufacturing required around those time lines, but as well as recognizing that market conditions can change, right, that perhaps even post licensure as a drug is being -- moving into -- through the market and people are reacting to it, things can change at a dime and the ability to make sure that, that's really well communicated with your manufacturing team is a really important part of this as well as recognizing that if you're working with an external manufacturing partner or frankly even an internal manufacturing partner, their available capacity, we're working in a space where capacity is very highly utilized, especially in the drug substance space. So how do we mitigate these challenges, right? One is just not being afraid of having an early discussion on what your supply needs, right? Providing an opportunity both for your internal team as well with the partners that you've been working with to have a window in, okay, what if this scenario manifests? What if this scenario manifests? What are the type of agreements that we need to have in place to ensure that we don't have supply chain gaps? And also making sure that you have demand projections driven by effective market intelligence, right? Having -- it's hard to say which challenges works, right? Overestimating your demand or underestimating your demand. But making sure that you have a really tight understanding of sort of the window that you'll be operating in year-over-year really puts you in a position to make sure that your capacity planning is accurate. And in some scenarios, you may even consider a secondary supplier, right? If there's something where perhaps the number of batches required is not exactly supported by a single supplier or just to create robustness within your supply chain, having a secondary supplier available to support your drug substance program can be really valuable. And the last risk we're going to cover today is related to the business risk of misalignment between your CMC plan and the funding strategy. As mentioned a little bit earlier, additional funding rounds or certain capital milestones may have specific requirements, right? Or if you're looking to develop an asset that's -- it needs to be in a certain state of affairs before partnering, making sure you understand those expectations is a really, really key part of ensuring that you don't have a situation where required -- the required activities needed to support access to funding that are not achievable, given the misalignment with your CMC plan. So some of this may include thinking about if you're deferring certain CMC activities because of the funding situation for the specific program, think about those mitigations ahead of time. It's totally appropriate and relatively standard. You've seen that certain companies may say, you know what, for our particular risk appetite and strategy, we're going to move this element of our CMC program to the later stages. And that may mean that it needs to move along quickly and with greater resourcing at that later stages. But having those mitigations in place so that once the funding is available and you're ready to start, you already have that plan available to execute, could be really, really important in mitigating that strategy. There's also ensuring that the technical program delivers the asset value that's required. It may be that if you're working with a partner or an acquirer for a given asset, they may have specific expectations around, for example, protein characterization that may not be stringently required for the clinical program that you're running but may introduce value to them. And making sure that there's -- that you're clear on what that may need to be from a business perspective can be really important to have that fed into your technical program as part of the scope that's developed. So in summary, when we think about the strategies to mitigate and embrace risk, the 6 points that I really want to highlight are: leveraging platform materials and approaches. So you can use that existing knowledge and feed it into your new late-phase program; thinking about adequate supply chain planning and the multiple approaches that one can take for mitigating challenges there; thinking about quality of design, not just as a prospective approach but retrospective in some instances, when needed to fill in gaps in your process understanding; as well as thinking about building out a robust process and planning ahead for the scale-out challenges that may emerge. You also need to think about deep diving into risk analysis and making sure that you're not trying to figure things out on the fly. You're making sure that, that's a proactive approach taken as well as proactivity in the communication, this being probably the most important part of any risk management, making sure that there's really close communication between the sponsor of the drug as well as the regulatory bodies involved and the manufacturers supplying it. So with those pieces in mind, I want to bring up a case study we ran into not too long ago at Thermo Fisher, which really related to poor performance ahead of process performance qualification. And this was a midsized biotech company that had an execution molecule. This was a process that they developed in-house and then transferred to us for Phase III manufacturing with the intent to move quickly into PPQ, given their particular expectations around the commercialization pathway for their molecule. The challenge that we ran into is that, that transfer cell culture process showed significant variability at scale. And where this really created the issue is that the transfer downstream process was simply not able to support the wide range of outcomes from the bioreactor, specifically in terms of the output titer. And so this is something that we had a challenge to sort of flag early because it really wasn't until it was performed at scale that we saw this issue. As I mentioned before, this is a set of activities, both from a PD perspective as well as a PC perspective were performed in-house by that customer. And the exacerbating situation was that this process was one with duplex bioreactors, meaning 2 bioreactors that were run in parallel into a single downstream train. So what did we do? Well, we realized that this was an issue that needed to be addressed to put us in a reasonable state of validation. And so the team took on the goal of performing really rapid process characterization and optimization studies at Thermo Fisher using that high-throughput process development equipment. Now this is that we have a long history of doing and understood very well about its success in moving and scaling up to the manufacturing scale it was being executed at. And so we decided to leverage an agreement with the client that long history of scale-up qualification. In some instances, it may be preferable to do scaled-up qualification before starting your PC activities. In this instance, it was decided it was a tolerable risk to reform that qualification in parallel. Beyond that, we used a very narrow design space for development that was evaluated to ensure that whatever solutions that might be identified through that experiment were easy to implement and didn't create risk to the filing. So we didn't consider things like changing the media composition, right, or looking at a different set of feeds, right? We kept the materials the same and instead focused on parameters that were relatively easy to control and that weren't expected to be seen as much more than tweaking optimization of the process. And lo and behold, the team was impressively successful with this, right, that because of that constrained approach and the focus on the type of risk that was going to be embraced in this particular strategy, those changes were successfully implemented as part of our PPQ program that completed its qualification successfully. Now while this program was successful and that is a reason to celebrate, I think it's an important takeaway from this is that there were a number of challenges here that we will say we got lucky to some extent, right? We have phenomenal expertise, but luck was definitely a component of our success here as well. And if you really want to avoid these issues, thinking about probing the robustness of your process earlier on, right, or even having the opportunity to do more evaluation as part of the tech transfer program, right? Where are the details around variability that weren't seen in small scale and that could have been better realized if it were included as part of the tech transfer program? Those are all details that I think pulled earlier could have minimized the risk here. So that is the end of what I wanted to cover from a drug [ substance ] perspective. I'll be handing it over my colleague, Christy Eatmon, to talk about this through a drug product perspective. And certainly, if you have any questions about this section of the presentation, feel free to pop those into the chat. We'll be happy to follow up afterwards. Christy?

Thanks, Dan, for sharing risks and mitigations on drug substance. We'll switch now to drug product. And although drug product is much less complex than drug substance, especially for biologics, there still are some risks and some ways to mitigate those as we move into late phase. So typically for drug product to biologics, the process is pretty standard. It's what we call thaw, pool, filter, fill. So we receive drug substance frozen, typically at minus 70 degrees, sometimes at 5 degrees if there's stability information at 5. Then we would thaw the drug substance typically statically. Sometimes at the commercial scale, we can have big kind of cryo vessels that we would use kind of in active jacketed thaw tank for that. But we do a thaw process so thawing the drug substance from minus 70, bringing it up to 5 degrees C so it's liquid. Then we would pool it so there's typically multiple bags or bottles of drug substance. We'll pool it together into some sort of mixing vessel Depending on the scale, we might use different vessels. And then we would filter it through the 0.22 sterile redundant sterile filtration and then fill it into vials, prefilled syringes or cartridges. So for today, we're going to assume that that's the process we use as a thaw, pool, filter, fill and that we're filling either into standard glass, vials, syringes or cartridges. And I think these risks would cover most of those and we'll talk a little bit about this. So there's a few risks here. Many of them are technical when it comes to drug product as we move from early phase into late phase. So the first risk that I've identified is changes in presentation. So what I mean by this is that many times for biologics, specifically those that might be targeted for at-home use, the presentation changes from early to late phase. So by that, I mean that we would move from a vial to a syringe. And typically, we fill into vials for early phase. There's a couple of reasons why. The first is that it's pretty quick, it's pretty cheap. In terms of the components, a vial is certainly much cheaper than a prefilled syringe. So that's pretty much the quickest and easiest way to get into the clinic. Additionally, sometimes in early phase, we don't understand the dose maybe, don't understand the frequency. We're looking for all those things as part of those clinical trials. So once we do understand the dose and the concentration we need to be at and the route of administration, we can settle on the presentation. So that is a risk because we need to maybe, at some point, look again at stability. So typically, if we're moving from say, a vial to a syringe and moving maybe from IV to subcutaneous administration, we'll have a concentration adjustment. So the concentration will most likely go up. Typically, in early phase, we work at concentrations, depending on the molecule, but in the 10 to 50 mg per ml range. And then sometimes, based on the dose and the volume that our clients want, we might have the concentrate to 100, 150, definitely in that range for subcutaneous delivery. So as we adjust concentration, we also have to adjust excipients, right, to make sure that we're stable, that we're not having any aggregation issues. So any changes that we make in terms of formulation or presentation, we have to, of course, get some stability of that new formulation or in that new presentation. So for example, if we move from a vial to a syringe, we'd have to do silicone spiking studies to prove that we have stability in the presence of silicone. That can be a challenge for large molecules because they do sometimes aggregate in the presence of silicone. And all prefilled syringes are silicone to some degree. There's different ways to siliconize those syringes like spraying on silicone, baking on silicone, but it's still part of that contact material. So we need to make sure that we test the stability. Also, syringes will have like a tungsten rod that makes the board for the needle so we have to do tungsten spike as well. So those are the kind of risks that we may have as we're moving from early phase where it's pretty simple standard type 1 glass, FluroTec-coated stoppers. The FluroTec-coated stoppers should remain even with a syringe. But any time we change formulation, we need to sort of start over again when it comes to stability. And stability for a drug product can be a bit different than drug substance because we're filling a much smaller volume so we have a much larger kind of volume to service area of that container. And we're performing another process on it, right? So we may be going for a free thaw. Additionally, sometimes in late phase, we change the storage conditions. So maybe in early phase, we're just freezing for safety purposes without a lot of stability at 5C and now in late phase, we're getting closer to that commercial presentation and we're changing to 5 degrees. So we have to gain -- get extra stability at those conditions as well. So we talked about this a little bit already, another technical risk, not only changing presentation from a vial to a syringe but changing formulation as well. So also, we have scale-up in terms of this as well. And also, as I mentioned, moving from a subcutaneous or from an IV to subcutaneous. Some other changes could be maybe moving from a liquid to a lyophilized formulation if we just don't have enough good stability at 5 degrees then sometimes -- it's pretty rare for biologics, but sometimes we would potentially move into a formulation. It could also depend on the market where the product is going, where we might decide to do a free [ tried ] formulation for maybe for stability at room temperature, depending on where the product is going. So those are the top 2 risks that I sort of see in late phase as these changes and any time we make changes, we have to kind of, not validate, but we have to prove that those changes isn't having any impact on the quality and certainly the efficacy of the molecule. Another change that we often see is what I'm calling failure to meet stage gate requirements. So at Thermo Fisher and probably most other CDMOs, we have sort of check boxes as we move from Phase I to Phase II, Phase II to III. Moving from Phase I to Phase II, there's not many requirements, right? But as we move to Phase III, it's really preferred that the Phase III product be manufactured in the same way as the intended commercial product. And then we also have some other things that we need to do as part of Phase III. So things like method validation of analytical methods, we want to make sure those methods are validated to commercial standards. And again, we might have to be redoing some evaluations and some development if we get concentration changes or excipient changes, right, making sure that those methods are still appropriate and that they're still going to work and we can validate them to a commercial level. The other thing, the big thing that I'd point out here is filter validation. Filter validation takes a lot of material, which we'll talk about next, but our requirement is that we do part of the filter validation. so microbial retention all in the intended filters. So we need to understand what filter we're going to use for commercialization. For biologics, I haven't seen many that don't go with a PVDF filter membrane. There could be other reasons why we use other filters, but we have to understand filter sizing per share based on the batch size. So we have to understand sort of scale-up, but there are certain requirements to move to Phase III, and that's where we start, as Dan mentioned, we talk about late phase and what are the differences. So from a drug product perspective, some of those differences are making sure we're in our commercial presentation, our commercial formulation, having filter validation and having enough of these kind of non-GMP background tests that I'll talk about a little bit as well to make sure that we understand that as we move from early to late phase, we're really planning for PPQ, right? So we need to have enough data to make sure that we meet our requirements once we get to PPQ. So the last one I have here is underestimating volumes that are needed. We see this as well. And it's a business risk because if we calculate that we need X number of vials and I see customers do this all the time, they'll calculate, oh, I only need 300 vials for the clinic, right? But -- and maybe I need 500 to 600 vials for stability, but they're not factoring in things like testing. They're not factoring in other things that may not still be validated like sterility or things that take a good number of vials. And they're certainly, as I mentioned, maybe not factoring in the volumes needed for filter validation. So we need to think about all those different volumes that are needed for different non-GMP tests or kind of off-line tests that would allow us to move to late phase and eventually PPQ. I called out 2 here that are compatibility studies. So that's product contact part compatibility. So as most of our filling lines, we have some stainless steel lines, but a lot of the lines these days as well as compounding vessels are disposable. So as we work more and more with single-use systems, we need to evaluate the stability or the compatibility in the single-use systems. So some of them will be kind of submerged or in dwelling, right, like a compounding bag. The LDPE bag is kind of like a submerged piece, whereas silicone tubings, polystyrene connectors, polycarbonate connectors, those would be kind of flow-through items. But in any case, we have to test the compatibility of the material at the concentration that we need for commercialization at the formulation with all the right excipients there to do those studies. And those studies can take a couple of liters because we're going to take samples at certain intervals and test them, right? So calculating volumes needed for that is important. Also other process studies, and there's a bunch, but just to name a few would be like a loss on filtration study or filter flush studies. We also might want to do pumping studies with large molecules to understand at what conditions do we get aggregation, right, especially with higher concentrated large molecules. We want to make sure that our drug product processes aren't having an impact on the quality of the molecule. So drug substance has spent all this time developing this and they're complex, it's so robust and takes so long. So we want to make sure that on the drug product side, we're not doing anything that's going to impact the quality, right? So typically for large molecules, that means aggregates. So pumping studies to kind of process and process again, kind of create a worst-case scenario, mixing studies, maybe things like that. So we would do those studies kind of off-line in a non-GMP environment. We could use potentially non-GMP material or maybe pilot material, but it just has to go into the planning of the late phase. If there are formulation changes or process changes made upstream at the drug substance side and they run a pilot batch, then we couldn't use that pilot batch to do some of these studies. So again, it brings the importance of transparency, communication and planning to make sure that we meet all these stage gates, that we have enough volume and that we're preparing for all the requirements for late stage, check off all those boxes. Do we have compatibility in the components? Do we have compatibility with contact materials? Do we have all of our methods validated? So just check off all those boxes to prepare for late stage. Again, this isn't an exhaustive list. Just a kind of a list I came up with based on what I see in dealing with programs and also with customers that are moving from early to late phase. So how do we mitigate? We talked a little bit about how to mitigate through planning. Again, as Dan mentioned, transparency between CDMO and clients is very, very important. Having those technical area to technical area discussions, right, planning every aspect and making sure that we have all the data because sometimes, as Dan mentioned as well, like assets may be moved around from company to company, and it's always important that the data accompanies that transition, right? So we want to have transparent communication but also transparent sharing of learnings. We don't want -- we want to eliminate tribal knowledge, right? So again, along with that transparency is the integration between drug substance and drug product. And of course, not every drug substance or drug product is filled within the same company. We understand that. In Thermo Fisher, we can do both, at least [ for 1 million ] cell-derived biologics. So if a biologic or any API or drug substance comes from another vendor, we need to make sure that we understand the timings, we understand how it's going to come. So whether we have an overarching project manager that covers all that or it's PM to PM of 1 group discussion or PM from our clients to PM of our teams, we need to kind of make sure that the time lines coincide and that they make sense. So we need to be prepared when it comes to time line. And having 1 source is obviously the easiest way to do that, but if that's not possible, there's other ways to make sure that we understand the timings. We also -- I sort of mentioned it but use standard processes as possible. So this really, really reduces the risk. We can use standard processes whenever possible. So in our network, we have a range of things that are prequalified or that we've used before, and that can mitigate the risk on new projects because we can say, okay, we've managed this type of product. We've done X, Y, Z. We understand it. We understand the nuances, and we can achieve a successful late-phase batch, right? So new and novel processes are always going to be out there because that's how we advance technology, right? That's how we move forward with our science. And that's how a lot of new things are developed. But when it comes to fill finish, it's pretty standard, right? So hopefully, we can always use our standard processes. We're not -- we don't want to recreate the wheel, especially moving from early to late phase. If we can kind of keep the process similar, that helps to derisk the late phase and then even derisk the PPQ. And some of these processes do very well scaling up so it mitigates the risk of moving to late stage even further. I talked a little bit about this but leverage key learnings. So if you're working with a CDMO, the likelihood that we have managed a product similar is high. In the past, I get questions from customers all the time, how many mAbs have you worked with in the past year? It's too many to count, right? How many oligos have you worked with? Well, maybe not as many but still several. So we can utilize our learnings. Of course, we can't transfer knowledge from 1 product to another in terms of studies, but we can use our knowledge based on what we've seen in the past, what's worked for us. And then that reduces -- hopefully reduces the risk. Also when we get to the FMEA stage, if we have -- we don't have sort of platform processes as drug substance does, but we have standard processes. So when we get to the FMEA approach, we can say, "Oh, we've run this process before. Here's the risk that we identified. Here's how we mitigated and it worked well, and we're going to use that for your program." So always important to leverage other programs but also leverage learnings in early phase, right? Because as we're making early phase batches, we're getting to know the molecule a little bit, getting to understand any parameters that we need to use specific to that program. So leverage learnings from early phase and from other similar processes. And that kind of feeds into keeping really good records in early phase. So this is particularly important if assets change hands from 1 company to another. We typically try to create documents that capture everything so that if there's a long pause between early and late phase or if the asset gets moved from 1 company to another, we understand what we've done in the past. So not that we can't manage something in late phase where we don't have a lot of data from early phase, but it's always helpful if we have as much data as possible. And then we're ensuring that we're not creating redundant work, right? That if something has been done in the past, that we can check it off the list that it's done but also so that we can maybe look at what's been done in the past and that will help us, as we go on to late phase, establish CQAs, critical quality attributes, CPPs, critical process parameters, right? Those are all things we have to identify for validation. But if we get started thinking about earlier how to build that, then we'll have an easier path to get to validation. So very critical to keep record keeping, especially when it comes to things like analytical methods, things that might be more complex than the thaw, pool, filter, fill process. And of course, plan early, take calculated risks. So that's the whole concept of what we're talking about today, right, understand where the biggest risks are and definitely mitigate those. But there might be other risks that are minor and maybe we can work around those or provide rationale. That's specifically critical if we have accelerated time lines, right? So if you have all the time in the world and have all the material in the world, you can do every study under the sun and make sure that you have a really robust process and you understand all the fine minutia of the process. But most of our customers don't have that luxury. They're trying to really get things in the clinic, prove safety, prove efficacy, and get things on the market, right? So we don't always have the luxury of a lot of time or a lot of material so we have to use it wisely, especially for a large molecule. They're just very precious material, right, really expensive. So we need to make sure that we plan everything out really, really well how we're going to use every drop of that material, right? It's really important. So again, those are sort of some of the top risks and how we mitigate those. A lot of it's going to come down to just being practical, doing FMEA, saying, okay, hey, we have a high-risk item here. How do we mitigate that? And working together with your CMO or your CDMO to make sure that we're all on the same page through transparent communication. I'll also share a case study. It's a little bit different than Dan's, but I think it still speaks to the risks that we often see moving to late stage and then also kind of how to mitigate them and how we can work together, client and CDMO. So we -- this is a product that I manage. So it's going back a couple of years. But we had a client that was kind of a small biotech company. They had an orphan drug, so they're targeting rare diseases. It was a mAb that was made, I think, 50 mg per ml so very standard mAb, which was good, right? And they -- I think they may have made like 1 pilot batch and then maybe 1 batch that covered Phase I and Phase II. So going into Phase III, we just didn't have a lot of data. We had hand-filled in the laboratory, so not a very controlled environment, the first kind of pilot batch from drug substance to have leading stability data in the vial. And then we made this other batch. And on the batch, we just used our standard processes, right? So we weren't challenging any ranges. We weren't even identifying what the CPPs were at that time because it's just standard, thaw everything, okay, does it look thawed? Yes, it looks thawed. Is there any ice left? No. Pool it together and mix it for 15, 20 minutes. Okay, homogeneous. Will filter it. No filter flush studies, nothing like -- just all standard processes, get this thing in the clinic as quick as possible. So by the time we got to late phase, there just wasn't a lot of data. So we're thinking forward to PPQ. And with the knowledge that we're not going to have enough data to check off all these boxes, right, to make sure all those stage gates are met, we can -- we were far enough ahead that we could plan to make more material so they could scale up from like a 1,000-liter to 2,000-liter bioreactor, but there's no way to create data that we don't have. So we thought, well, if we run the Phase III batch just by standard processes, that's not going to do as much good because we won't be able to challenge any ranges or really identify CPPs that we need to move into validation. So basically, what we did was ran the Phase III batch a little bit as like a pre-PPQ batch. So although it was a clinical batch, we added some things like we don't have a lot of material to do all these studies in the lab, right? This material is really precious. So let's see what we can do on the GMP batch and get quality buy-in. So that's always important to get quality buy-in whenever you're doing something different, and validation buy-in on how we can best use the material and how we can prepare for PPQ. So on the Phase III batch, we had to take the material for filter validation. So that meant that we had to -- after pooling, take some off, send it out to our filter vendor and then actually hold the batch until we got that data to release the batch. So that's not standard. But again, with nonstandard processes or nonstandard situations, we kind of have to be creative. So we planned to run it as like a pre-PPQ, like I said, where we would challenge the hold times, take extra samples. We didn't have material to do a filter flush study in the lab. So we said, "Well, let's just take the first bit of material that comes off the filter, test the concentration, see if we have any loss on filtration. We'll take the first few vials that come off the line. We'll test them to make sure that we have homogeneous drug products. So running that Phase III batch like a pre-PPQ batch really gave us a lot of data, right? So we can't test the low end. Well, we could test one or the other, right, low or high when it comes to mixing speed, mixing times and things like that. So we figured that high would be worst case so we let that mix for a really long time. We tried to -- we stopped the batch actually kind of like what we do on immediate fill, stop the batch, hold it and then start filling again to test like time on the filters and things like that. So it was a risk because if anything would have gone wrong, we would have kind of had to -- we would have had deviations on the clinical batch and maybe risk the clinical time line, but we determined that the benefit of having that data available moving into PPQ was more beneficial than the risk of a potential loss of this Phase III batch, right. So again, it's up to a good conversation with client and CMO to figure out what the best solution is. I'm not going to say this was a perfect example of what we did, just 1 thing that we did. So in the end, we did have 3 successful PPQ batches. So our -- we always have to run 3 consecutive PPQ batches successfully to prove validation. And they even did something different with those PPQ than typical. Usually, PPQs are made for commercial launch. These were kind of a combination of backfilling clinic and also commercial launch. And then, of course, working with the regulatory agencies, not from our perspective as much, but from the customer to understand continuous process validation and what needs to happen annually or biannually as we make more batches and submit more information, supplemental information to the agency. So I think that's all in my case study. Let's move to kind of the next steps here. So what happens after Phase III? So after Phase III, typically for large molecules, we move into validation, right? So this slide, I'm not going to read everything on here. But this slide is just a picture of all those check boxes I've been mentioning for the whole time we've been talking here. So as we think about what happens after late stage, start to prepare for PPQ, so make sure we're starting to think about FMEAs, right? Any changes to the process, changes to those batch records, any changes to the bills of materials, we need to change those. Like I said, product contact compatibility, identifying what those CPPs are. Sometimes, the CPPs and drug product aren't very clear. At quick look, it's like what's the CPPs fill weight or just they're not always super clear, mixing times, mixing speeds, those type of things. But we have to make sure that we understand what they are and make sure that, of course, all the requirements for the facility that you're working in are met like change controls, documents that we have to write. So these are examples from ours. Every CMO might be a little bit different. But in general, there will always be stage gates, all these things that we have to think about as we move from late phase to PPQ. And I think at Phase III is the time to start thinking about PPQ. Even if there is a long break, it could be several years between Phase III and PPQ, but we still need to make sure that we have all the data, that we're gathering the data so that we're ready when the time comes. So key takeaways here, again, are there's always risk, right? So as we're manufacturing these really complex molecules and managing processes that take many days, happen at various facilities, right, everything doesn't happen in the same building, there's always going to be risk there, but we need to understand what the risk is to be able to mitigate it. And then, of course, try to anticipate what could happen based on our experience. So the good news is we're pretty experienced. Most of our customers are pretty experienced, and we all, in our everyday life, right, use our experiences to guide what we do. So we always look for ways to mitigate and are always open to that transparency, right? So when things go wrong, let's be transparent. Let's sit down, come up with a solution that works for both parties and really figure out ways that we can work better together so that we can make successful products, high-quality products, safe products that impact the lives of all of our patients, right? So we really want to make sure that we have good quality product that gets out there in the market, whether it's a clinical trial or a commercial product that we feel good about the products that we're making and feel good about how it's helping patients around the world with better quality of life and carrying diseases and things like that. So again, thanks so much for your time today. On behalf of Dan and myself, really appreciate you joining our webinar. There will be a time to write in questions. So please, if you have any questions, type them in the chat. And at this point, I'll say goodbye and pass maybe back to Andrew just to finish this up. Thanks.

Thank you, Dan and Christy. As I mentioned at the start, we have an exit survey which 2 questions about the webinar so please stay with us. First of all, we're asking what topic are you interested in learning more about. The options are technology transfer and scale-up, demand forecasting, switching drug formats in late phase or commercialization, quality by design for biologics, preparing for PPQ, biosimilar development, fast-track designation and patient centricity and drug design and delivery. And secondly, would you like to connect with Thermo Fisher to discuss your project in detail? Please just let them know, yes or no. And on that note, I'd like to thank our speakers, Christy Eatmon and Daniel Baskind for that great presentation and our sponsor, Thermo Fisher Scientific for making this event possible. On behalf of Thermo Fisher Scientific and Citeline, have a productive remainder of the day, and thank you for watching.