Latest / BioSpace / The supply chain fault lines of modern pharma
Transcript
- speaker-0: Hi, my name is Jennifer Smith Parker, Director of Insights at Biospace, and you're listening to Denatured. In this episode, sponsored by SK FormTechO, I speak with Arvind Singh, VP and Global Head of Supply Chain. We discuss how CDMOs are reshaping supply chain strategy for a new generation of complex modalities, from smarter raw material sourcing and precision cold chain logistics to navigating geopolitical and regulatory uncertainty. And deploying AI to build sustainable, scalable capabilities from clinical through commercial. Vin, thank you so much for joining this episode of The Nature. It's a pleasure to have you. If you can go ahead and introduce yourself, that would be great. speaker-1: Yeah, sure. Hi everyone. I'm Arvind Sang. I have over 20 years of experience in the life sciences industry, primarily focused on manufacturing, technical operations and global supply chain. And I'm currently with SK Fontech, where I lead the global supply chain across a multi side GMP network across the globe. And my responsibility is expansion across states and operations planning to procurement to capacity planning, all the way till supply chain ⁓ risk management and global strategies. speaker-0: Brilliant. Well thank you so much for the introduction. And let's go ahead and launch into the questions. So from a supply chain perspective, how does a C DMO strategy shift when transitioning from shelf stable small molecules to highly cold chain dependent peptides in ultrasensitive cell therapies? speaker-1: So one clear one quick distinction. For not every peptide would be a highly cold chain dependent with peptides, the biggest supply chain challenge is often the complexity of the manufacturing process itself and the availability of the specialized ⁓ materials and capacity. So with that clarification, the way I look at it is as we move across these modalities, resilience shifts from managing inventory to managing time, technical complexity, and ultimately patient identity. For a traditional small molecule, we can then create resilience through safety stock, qualified alternate suppliers, geographic diversification, and flexible production campaigns. Now, if a shipment is delayed, we may still have inventory to ⁓ you know available to protect the customer. But for peptide, the challenge changes. We may need specialized amino acids or amino acid derivatives, for example, resins, solvents, ⁓ purification columns, etc. Now there may be many synthesis steps in a peptide production and a small yield loss at each step can get compounded and become very significant at the end, right? So we need to involve in this case procurement or supply chain very early enough to be able to qualify preferred suppliers, procure the materials, right materials, understand upstream dependencies and reserve the right technical capacity. not simply purchasing more inventory to protect. And in case of cell therapies, which is, you know, and and more so pathologist, the patient is effectively the supply chain or it's part of the supply chain, right? Both starts from the patient and ends had patient. So it's, you know, think of this as the patient cells are collected, transported to the manufacturing site, processed, tested, shipped back, and infused into the same patient. So in this case, you cannot hold conventional finished goods inventory. It's essentially resilience in this case comes from precise scheduling, chain of identity controls, validated code chain logistics, real-time visibility, backup transport routes, and very exception management handling procedures, essentially. Now, as a C DMO, this is why we try to identify material capacity. quality and logistics risks during the proposal and product intake stage. We then carry these assumptions into supplier qualification, forward looking capacity planning, and our sales and operations planning. So again, the objective is pretty much the same across these modalities. It's essentially meeting our customer requirements on time and in full speaker-0: There are many steps and a lot of processes involved, but it's important, isn't it, when you're going to be following this to ensure accuracy? So peptides sit in a fascinating middle ground between small molecules and large biologics. So what are the hidden supply chain vulnerabilities in peptide manufacturing today, especially regarding raw materials like amino acid derivatives or specialized solvents that biotech sponsors often overlook? speaker-1: Yeah. Very interesting question. So peptides is a good example of why a supply chain can look simple at the top but become very complex underneath. What I mean by that is like let's say a a sponsor is seeing one peptide API, but the product may require numerous protected amino acid derivatives, speciality resins, coupling reagents, solvents, just huge conjunctions of solvents and peptide manufacturing, purification materials and and in in many cases. separately manufacturing these peptide fragments that has its own supply chain. So if even one critical building block is late or does not meet the required quality, the entire manufacturing can stop. So that's the first risk. This the second hidden risk is cumulative loss and kind of alluded to earlier question, but it's more like think of a peptide has 30 steps of synthesis and you know just take an example. And even if each step You have an yield of, let's say, 98%. The theoretical yield at the end of the ⁓ you know, 30 steps, but your final purification comes around to be around 55%. So small differences in your raw material quality can therefore have a very significant effect on output, impurity for impurity formation cost and and schedule of the project. The other underestimated ⁓ area is solvents. So peptide manufacturing can consume large volume of high purity solvents, sponsors. may focus on say price of the amino acid, but overlooked solvent availability, it's storage, hazardous material transportation, the waste treatment capacity of a manufacturing facility, and and changing environment requirements. Like taking example of straight-of-remuse closure, you know, has a downward pressure on the price of solvents and also the availability of solvents. So that's definitely an area to to watch out for. Now there's a The other piece with solvents is is the reverse supply chain that supply that gets often overlooked. Now large solvent volumes must be safely stored, recovered where possible, and ultimately removed and treated as a lot of waste. So if an operation site has waste storage that it reaches its waste storage limits, or a waste contractor misses a scheduled collection, or an incineration facility lacks capacity, production can stop. even when all the raw materials and you know and the resources that are available. So solvent recovery, you know, greener process development and qualified waste partners become very, very critical, both both from a sustainability perspective as well as from a business continuity measures perspective. So so the way to think about for peptides specifically is this that resiliency is not only about getting materials into the plant, but it's also about safely and sustainably getting waste out of the plant, just because of the huge, huge amount it has. And of course, then there are capacity constraints beyond the synthesis equipment, so they could be purification aspects, chromatography, analytical testing, et cetera, where you cannot always add a capacity as quickly as the demand increase. So we try to map these risks during the proposal state itself. Which materials are critical? How long are the lead times? Is there a qualified second source? ⁓ you know, are the two suppliers procuring from the same upstream manufacturer? We then use our referral partner network, risk-based supply qualifications, sourcing engagements, et cetera, along with the forward looking capacity planning to address the highest risks items. And that's basically the importance here is that buying the material is not the same as securing supply. The true resilience actually comes from your qualified materials, reliable suppliers, ⁓ available manufacturing capacity and enough forward visibility to bring all these elements together to ensure uninterrupted ⁓ tide. Manufacturing that's all. speaker-0: So with the type of minute detail that you're talking about, now applying that to the extreme precision required for ultra cold storage, chain of custody, and real time tracking when a patient's life depends on a forty eight hour delivery window. speaker-1: So when we talk about let's say resistance and pharmaceutical supply chains, it's important to recognize that every modality has its own critical control points. The requirements are different, but the objective is same, right? Ensuring that right product with the right quality reaches the right patient without interruption at the right time, of course, right? So right product, right quality, right node, downstream node, customer or patients, and then right times. And and if you lay that, layer that out for for various modalities, like for small molecule, for instance, precision starts much earlier in the supply chain. It means knowing where critical raw material comes from, ensuring supplier qualification, maintaining traceability, controlling storage conditions, ⁓ and having the right manufacturing and release timelines. So disruptions several steps upstream can ultimately impact your operations and patient supplier. For more complex modalities such as peptides, or you look at advanced therapies, the level of orchestration increases. Peptides may depend on a specialty amino acids, we just talked about purification capacity or tightly controlled manufacturing conditions. Whereas on gene therapy sites, viral vectors, for instance, ⁓ may require an ultra-low temperature storage, specialized packaging, quality or or Qualified transportation lanes and much tighter coordination between manufacturing, quality release, and distribution. So our approach is to treat the supply chain as an integrated system rather than a series of independent steps. What that means is that we're building visibility from suppliers through manufacturing, your quality release and testing, all the way to logistics. So we use risk-based planning to identify critical dependencies very early. It's a single source material, constraint manufacturing network, analytical bottleneck or transportation risk. And for any temperature sensitive product, validated packaging, temperature monitoring, qualified logistics partners, and contingency plans become very essential. Now, and this is where you know think about the people, process, and system aspects where the technology itself will not just create resilience, right? The foundation of discipline processes, the accurate data, the clear ownership, the resources and rapid decision making is when something changes will become very important to be able to actually make your supply chain resilient. speaker-0: Okay. That does make a lot of sense. So when we're talking about sponsors and they naturally focus on clinical trials and regulatory milestones, where do you most frequently see a disconnect between a sponsor's expectations and the cold, hard realities of supply chain lead times? speaker-1: So the ⁓ the most common disconnect is that so sponsors understandably plant around their milestones, like the clinical milestone or commercial milestone. And while the supply chain, you must plan backward from that milestone in in many cases, many months in advance. Right. So sponsor may say, We need this material in let's say next month, September. But before manufacturing can begin, you'll need to secure customer raw materials, qualified supplier. You know, complete incoming testing, reserve equipment, laboratory, capacity, resources, et cetera, right? So for peptide, this also includes specialized amino acid derivatives or fragments. For bilobactus, it might include plasmids, cell bank, transfection materials, ⁓ single-use components. So you could see how the the complexity increases. Another disconnect is between forecast uncertainty and capacity expectations. So chemical programs are naturally changed peptide. manufacturing capacity, skilled people and long lead materials cannot always be turned on and off immediately. So if a technical decision or a forecast is delayed, it may be no longer possible to preserve that original production slot. So capacity I always say is this it's time sensitive. An unused slot today cannot be stored and used six months later as is. So that's definitely a a ⁓ watch out. The other important aspect is to always clearly understand end to end lead times. Completing manufacturing doesn't mean that product is immediately available. You're you're testing your data, deviations, final quality leaders equally is important and hence the resources in that area should also be ⁓ critically looked at. The one thing I would say is is that I don't see this as a sponsor problem or a C DMO. problem, right? It's usually an assumption alignment opportunity or issue, however you look at it. And our role as a strategic C DMO is to make these assumptions visible early at at even at at the proposal stake, for instance, we should clearly document material lead times, capacity holds, required customer decision dates, capital or equipment needs, testing and release timelines, and the consequence of any any late changes. And you carry then those assumptions into a joint demand plan, probability weighted capacity planning. And your regular SNOP reviews. Bottom line, the principle is very simple. You should never separate the commercial promise from the operational reality. speaker-0: Yeah. That makes a lot of sense. And I I did like making the point of a particular slot being open and I can't wait. ⁓ reminds from a layman's perspective, ⁓ flying airplanes and they have a slot that they have to leave at that time and you can't have that waiting ⁓ for like a later day or a later time it has to happen then or it's it's speaker-1: The consequences, yes, correct. speaker-0: Yes, exactly. Exactly. So I'm glad you mentioned about the straight different movies earlier, because given recent geopolitical tension and supply shocks, the industry has talked a lot about dual sourcing key starting materials. So how do you advise sponsors on balancing the cost of multi sourcing against the existential risk of a single supplier failure? speaker-1: The first thing we tell our sponsors is that resilience does not mean dual source in everything, right? So that's that's not the case. Resiliency essentially means what failure mode exists, where it would have the greatest impact, and then invest accordingly. So this avoids unnecessary cost, complexity, and ⁓ regulatory burden. We begin with a risk-based assessment. We usually ask four questions: How critical is the material, how likely is the disruption, how long it would take to recover, and What would be the impact on the program or our customer commitment ultimately to patients? So if a raw material is available from only one qualified source, it requires a long and complex manufacturing process or would take many months to replace that supplier, then the additional cost of qualifying a secondary source is effectively a resilience premium. Now that cost is usually very small compared to a delayed clinical trial, missed launch, or interruption to your commission supply. Now there's a caveat. Now, dual sourcing has to be real. Two distributors may purchase from the same underlying manufacturer, or or two suppliers may operate in the same geography and depend on the same raw materials upstream or raw materials manufacture upstream. So one has to look beyond the immediate supplier and map upstream dependencies, manufacturing locations, logistical routes, financial health of your suppliers, quality performance, and and of course available capacity to be able to fulfill the requirements. A second source must always be technically qualified and approved by quality, right? So that's that's another important aspect is that it has to be part of your regulatory filing strategy. If it has never supplied material at the required scale, for instance, or it's not qualified, then it will not provide much protection during a crisis, even if you have identified it early on. So for the most critical materials, It can be very helpful to give the alternate supplier some recurring volume, for instance, so that the process and relationship stays active, keeping the line warm. For lower risk materials, full total sourcing may not be justified. You may have other production other protections, including a strategic inventory, reserved supplier capacity, a technically assessed backup source or or an agreed recovery plan with your suppliers. These are some of the reasons why we have a preferred partner network across multiple reasons, geographically dispersed. We segment these suppliers and materials based on criticality, spend, revenue drives, quality, capacity, overall strategic value, in other words. And that allows us to investe qualification resources where they create the most protection for our supply chain. And the one thing I would highlight here is that the lowest unit price is not always the lowest total cost. The resilience sourcing decision balances Today's material cost against the financial, operational and and project or patient consequences of not having that material tomorrow, right? And that theme actually comes across even in your public policies. So for instance, look at FDA, identifies as single or limited suppliers, foreign dependencies, transportation disruption, any geopolitical events as an underlying supply and chain risks and wants the companies to consider the ⁓ alternate sourcing and and mitigating those risks. The EU on you look at twenty twenty six Critical Medicines Act agreement similarly emphasizes on diversified and reliable sources rather than purchasing it decision based on just lowest ⁓ cost or or price. speaker-0: Yeah. I think those are good words for wisdom, not necessarily just looking at lowest cost, ⁓ you're s saying here. There's a lot more to look into. So when a client transitions a project from early phase clinical supply to commercial scale, where do supply chain bottlenecks usually hide? And how does an experienced C DMO de risk that handoff early? speaker-1: In these cases, usually hides at the interfaces. So the areas between process development, procurement, manufacturing, quality, regulatory strategy. Some of some examples, right? We can take certain examples. Let's take raw materials again. At a clinical stage, a project may use very small quantities from a specialized supplier. But at commercial scale, that supplier may not have the required capacity or quality systems or lead time management resources perspective. or ability to produce the larger batches. So minimum order quantity, shelf life, storage requirements, alternate source qualification can become very significant as you move further in in your project. The second is facility and process fit. A process that works well at a small scale may behave very differently at larger scale. Right. So look at processes, mixing, filtration, heat transfer, et cetera. Of course they impact or change based on the equipment. But on the supply chain side, you look at solvent handling, waste capacity, ⁓ yields, all can change. So it's very critical to be able to actually manage that aspect. The third is on the laboratory and quality release, capacity resources side of things, the equipment and resources there. So, you know, most of the time we're looking at, let's say, ⁓ sponsors maybe focusing on manufacturing, but commercial supply can also require validated analytical methods, reference standards, ⁓ stability programs, trained analysts. So a full manufacturing suit doesn't help if the batch cannot be tested and released, right? So that's another aspect where we'll have to be cautious of. And then other risk is knowledge transfer. Critical process knowledge may sit with a few development scientists at an early stage, but may not have been, let's say, fully captured in the documents being transferred to the commercial team or or another site. So that so the tech transfers become tricky in that case. The way we manage these is is we de-risk this transition well before the pivotal or validation batches and how we do that. So we conduct a commercial readiness assessment, identify scale-dependent risks, map critical and long-lean materials, evaluate our suppliers, and determine whether capital equipment or facility modifications will be needed. So and it's also beneficial and it has helped us to establish a clear stage gates for capital, material readiness, analytical readiness, and technology transfers as the evaluator of project. or or different phases of manufacturing. And finally, what you want to do is is you do want to look at this all of these aspects holistically as part of your, you know, CSN operations planning or your integrated business planning. Because the goal is again to preserve your process knowledge and make commercial decisions while there is still time to create options. If we wait till the demand is confirmed, the only option might be you have a added cost, delays or unnecessary technical risk to your project. speaker-0: I think that's important as you're saying about planning ahead and and really not waiting until the last minute. So speaking about speed, advanced therapies are moving globally much faster than in previous decades. How do you keep supply chains compliant when dealing with divergent regulatory frameworks for gene therapies across the US, EU and the Asia Pacific regions? speaker-1: Yeah, so the the key point here is that compliance cannot be inspected ⁓ into supply chain. So you can't you can't add it at the very end. It has to be designed into and supply chain. So the way I I categorize it is is these plan, source, make, deliver. And yeah, all these four buckets, you have to design your ⁓ compliance in into these. So what we do is is we use a global foundation with regulatory overlays approach. local regulatory overlays approach. And what I mean by that is the global foundation is standardized. It's basically based on common GMP and ICH principles, qualified suppliers, controlled materials, data integrity, traceability, validated processes, documentary change control, clear quality agreements, etc. So that's all very standard. Now these principles should remain consistent regardless of where the product is manufactured or supplied. We then add the requirements of The specific modality and target market. That works must begin during the, let's say, initial stages of project validation or intake. For example, the focus may include the regulatory starting material definition, impurity control, supplier qualification, and material traceability as it moves across the supply chain. The regional differences you pointed out are also very important. The US, EU, Asia Pacific generally share many quality principles. But they may differ in their some product classification at the local level, the documentation requirements, importation. There are some might be some ad hoc local testing requirements. So you have to comply to those. And our first approach is to confirm where the sponsor in the intends to conduct the clinical trials or launch the project. And we then develop a market and modality regulatory metrics, kind of where covering suppliers, trading materials. manufacturing side, specifications, testing and storage requirements, et cetera, to be able to actually not just produce it, but also be able to actually deliver to our customer for for the markets. Change control becomes a another important aspect. ⁓ a supplier process, analytical method, et cetera, manufacturing side of our suppliers, it cannot be changed without assessing the technical quality, customer or regulatory impact. So wherever possible, we design alternate suppliers and manufacturing options into the regulatory strategy. very early because a technically capable backup provides the limited that protection. If the approved dossier doesn't allow you to that supply to be used, then it's a limited production if it's not qualified. And our presence from a SK farm takeover, SKPD perspective, our presence across the globe gives us the access to regional experience and the ability to understand requirements closer to the relevant markets. So that's ⁓ And you know, I think that's what we're that's where we are leveraging our people and process and our knowledge of the system to be able to actually meet our customer commitments. speaker-0: So speaking of hot topics, ⁓ we hear a lot about AI and drug discovery, but less about AI and logistics. So where is your team actually deploying digital tools or predictive analytics today to prevent supply chain disruption before they happen? So speaker-1: Few things to maybe maybe clarify first, maybe separate it out just from a distinction perspective. So digital systems can tell us what is happening, the the way we have it. And I'll I'll draw it from ⁓ from SKPT is that digital systems can tell us what's happening. Analytics helps us understand why it's happening and then AI can help predict what may happen next. So today our most practical focus is creating reliable, connected visibly. across demands, materials, supply, capacity, manufacturing, quality and logistics all the way, right? So end-to-end. And through our SNOP process, we're bringing that together, all of that together, including the customer forecast, project milestones, material readiness, et cetera. We're also strengthening our planning systems to master data, Rafate Capacity Planning, Supply Intelligence through our preferred partner network. Now there are tools that exist and and these tools already help us identify conflicts earlier, for example, when it when a customer programs require the same equipment, laboratory resources, et cetera. Now, you know, one may think, ⁓ these are not very exciting for AI, right? But without the accurate data and disciplined planning, an AI model simply predicts the wrong answer faster. So so the data becomes very critical. A practical step is to use when we look at predictive analytics or AI, energetic AI. So where I think is is it come into these following four areas. First is demand forecasting capacity. A C DMO pipeline contains signed business, verbal awards, and proposals with different probabilities. Predictive and generative models can combine probability expected start dates, material requirements, and honored manufacturing slots, et cetera, to show where future constraints are likely to emerge. Second is supplier risk, and and we're exploring ⁓ some AI tools here in this area where We can combine our internal data, lead time changes, late deliveries, quality deviations, capacity requirements, responsiveness of the supplier, and then combine that with external signals such as weather, geopolitical developments, tariffs, ⁓ code congestion, financial results for suppliers. So that objective there is to flag a ⁓ developing problem before a supplier formally misses a delivery or impacts our operations. The third is on the operational execution side. where the I can identify patterns that indicate potential schedule delays, equipment failures, material expiry, loweries, ⁓ etcetera. So so that more from a predictive maintenance perspective and and all production planning adherence perspective. Fourth is the logistics aspect where temperature sensitive products, real time location and temperature data, you know, so so think about a temperature sensitive products I'll back up. Real time location and temperature data can be combined with weather. route and flight information to predict a delay or excursion and allow intervention while the shipment can still be protected. I would say though that AI as an I I don't view AI as an autumn ⁓ autonomous decision maker. It's in a regulated environment, the model must have reliable data, a defined purpose, appropriate validation, cybersecurity controls, and and very important, explainable outputs, right? So ⁓ AI system may ⁓ suggest to you an ⁓ alternate supplier, a ⁓ shipping route, ⁓ but you still need a qualified expert in a company and supply decision. Yes, correct. speaker-0: You need that real person. Yeah. Yeah. You hear that all the time when when it comes to AI, the kind of the new tools. You cannot just rely on the machine to tell you everything. You have to cross-check that. So let me ask you about advanced therapies again, of requiring intensive cold chain infrastructure, liquid nitrogen shippers, specialized packaging, rapid air transport. How can CDMOs reconcile the strict temperature requirements of complex modalities with growing corporate sustainability goals? speaker-1: Yeah, so so first, very important. Product safety and product integrity must always come first. Right. We should never reduce a validated temperature requirement or a control system to simply to claim a sustainability benefit, right? So that's important just because of the industry we are in and it's ultimately serves the patient. However, sustainability and supply chain resilience, the way I see it is not necessarily competing goals. In many cases, they reinforce each other. ⁓ give an example. So let's say a temperature excursion that destroys a high value drug or product, it creates an environmental impact of that original manufacturing, packaging, transportation, et cetera. And you still require the product to be manufactured is shipped again for the patients. So preventing product loss may be the most important sustainability action that we can take in pharmaceuticals. So and light census industry. And we approach this through several practical labels, right? So the first is packaging. We're validated for the product and route, we can use reusable shippers, right size packaging, selecting lower impact materials, and established a take back programs with logistics partners. So reverse logistics again in this case becomes very important because a reusable shipper is only sustainable if it's returned, inspected, reconditioned, and and reused for all practical purposes. The second is route and network design. Qualified direct routes, shipment consolidation, reliable carrier capacity, and better forward planning can reduce unnecessary handling and emergency air freight. So your regional manufacturing can also bring production closer to the customer and patients. For SKPT, let's say we our viral vector presence in the United States and France supports your regional supply model and can reduce dependency on a single site or transportation corridor. Right. Same is true for our small molecule facilities across the globe. So that's where that geographic presence becomes important. The third is the digital visibility, real-time temperature and location monitoring combined with productive weather, flight and route information can identify a developing delay early enough to intervene. So better visibility, it creates data to optimize your packaging, have more efficient transportation. without compromising the product quality. Switching gears. Within the manufacturing facility, so that's that was transportation, but when you look at manufacturing facility, sustainability includes energy efficient ultra low temperature freezers, preventive maintenance, efficient use of liquid nitrogens and refrigerants, you called out renewable electricity where available, and reducing product and packaging waste. And and finally, we need the right measures. of sustainability. So we should look beyond, let's say, carbon per shipment and consider carbon per successful dose delivered, temperature excursion rates, emergency shipments, product loss, packaging returns, et cetera. So I think that's becomes a more meaningful metric, maintaining the validated conditions required to protect speaker-0: Yeah, that's a lot to think about. So the final question is a crystal ball question, as I love calling them. So ⁓ looking ahead for three to five years as new modalities like MRNA, radio pharmaceuticals, or antibody drug conjugates green traction, what supply chain capabilities should biopharma companies be building today to be ready for tomorrow? speaker-1: So the like you said, it's crystal. But the the answer lies in keeping it simple and and basically sticking to basics, right? So the first capability companies need is the ability to understand the complete supply chain for each modality, not just one step. So plan, source, make, deliver. You need to understand it end to end. And what does that mean for like these various modalities that you mentioned? So let's say MNR ⁓ R N A. It means visibility across your plasma DNAs, enzymes, nucleotides, capping reagents, et cetera, in temperature control distribution and follows. For radio pharmaceuticals, the model is completely different. Your companies need reliable isotope and ligand supply, licensed facilities, radiation handling capabilities, specialized transportation and manufacturing located close enough to the treatment centers to manage products. Whose use for life is actually measured in hours. So very critical on the distribution front, logistics and distribution. And ADC's interesting case, it combines the three supply chains a biologic antibody, a highly important small molecule payload and linker, and the conjugation ⁓ conjugation and the final product process. So these activities require different technical capabilities, contact containment strained and standard suppliers and quality systems. But they all must arrive at the time and meet at one place or site for the integrated product specification. The second ki capability is flexibility and modular capacity. So companies should avoid building facilities and processes that can serve only one product or forecast. Platform processes, scalable equipment, strong analytical methods, and transferable technologies allow capacity to adapt as our pipeline changes. Third is Supply and geographic optionality. Critical materials and capabilities should be mapped several tiers upstream with qualified supply networks, strategic inventory where appropriate, and regional manufacturing options. But optionality must be technically qualified and incorporated into regulatory strategy before it's needed. And I think I've mentioned this a few times now. Your fourth is. Integrative planning and digital traceability. So companies need to connect the probability-weighted pipeline with material lead times, RD hours, manufacturing capacity, logistics requirements. Your predictive analytics or AI can then help identify where several future programs may compete for the same scarce material equipment or technical expertise. And the final piece is your talent and your partnerships in public-private partnerships. No company needs to own every capability. But it must know which capabilities are critical to control internally, which can be provided by a say strategic C DMO or supplier, or how or and or how these suppliers will operate through common quality standards, data governance, and escalation processes. In summary, what I would like to say is that the companies that'll be ready in three to five years are not necessarily those that build the most capacity. They'll be the ones that build the right optionality, your technical platforms, your supplier ecosystems and partnerships before the demand or or the modality becomes urgent. speaker-0: Well, thank you so much, Robin, for this fascinating discussion about supply chain logistics. It's been very informative. If you'd like to listen to more episodes of The Nature, please turn to biospace.com. And if you'd like to send me a pitch, please send me an email at jennifer.smithparker at biospace.com. Thank you so much.