Third Bridge专家称Charles River剥离CDMO后费用拖尾可能持续压制短期利润率改善

[[00:01:15]]
Regarding Charles River Laboratories' CDMO [contract development and manufacturing organisation] divestiture that was completed in May 2026 – why do unallocated corporate overhead expenses historically lag the completion of facility divestitures? What structural friction points inside the remaining operations prevent the rapid elimination of trailing costs?

[[00:01:39]]
When a business, for example, such large as Charles River CDMO and Cell Solutions unit is diversified, the manufacturing assets and employees and revenues, they usually transfer relatively quickly within a few quarters, as you have seen from their earnings calls. However, there are some corporate functions, which are more support functions. They always remain embedded within the organisation. For example, these are like finance and accounting, HR systems, legal and compliance teams, quality management systems and then there is a procurement, which is a very important function because there's a lot of contracts involved and enterprise software licences, which are linked to a lot of GMP applications, cybersecurity programmes and then, of course, shared services centres, which are shared through different business units and of course, executive leadership costs. Now these costs are often allocated across all the business entities within a business unit. Once the business is sold, for example, in this case, CDMO, many expenses cannot be immediately removed, cut and dry because the remaining companies still need those functionalities and capabilities.

In this case, it's not that simple. You ask me like the friction points, I would say, at minimum, these friction points remain, such as SEC reporting, which is the compliance, audit functions, investor relationships, internal controls and compliance infrastructure because the one division is sold doesn't mean that the other units don't need it. Then the headcount reduction timing, which could be another area, for example, the companies typically eliminate support positions months after divestiture and completion because retention agreements could be there, severance obligations are there, transition support requirements are there. Under the transition service agreements, which is we call that as TSA, there is sellers usually provide ID, in this case, is very, very important because of the data involvement, payroll, procurement, ERP support. Now all these entanglements take time, and they could vary from six to 24 months even after the deal is closed. Employees are easier, manufacturing facilities are easier. These are like typical hidden costs, which takes time.

[[00:04:28]]
What does the slow reduction you mentioned mean for the near-term margin targets of Charles River's highest growth business lines?

[[00:04:36]]
I think the most important aspect is going to be because CDMO has been not very profitable since they acquired the CDMOs in '21, and they acquired them at a peak and so is the Cell Solutions and because all the demand went down, so I think by diversification, this margin is likely to improve over time, provided the biotech funding remains strong and there is a recovery period. In this case, decoupling, for example, these CDMOs would improve the margins, but over time.

[[00:05:22]]
How can Charles River's management accelerate the removal of the trailing expenses? Or do you think those expenses are permanently baked into the remaining lab footprints?

[[00:05:33]]
I think the good news for Charles River is that these three CDMOs and even Cell Solutions were acquisitions, and they were like themselves as independent entities. As you can see, all these three CDMOs were their own manufacturing sites. I think that the easier way is to have their systems allocated and diversified as the deals happen and the most common shared systems would be probably quality management system and data retentions, which I think can be managed. I think that's the good news from Charles River standpoint. That's my view on this, but can it be expedited? Of course, it depends on the receiving unit also, like who is receiving it and are they capable and what is the agreement between seller and a buyer.

[[00:06:31]]
How seamlessly can a CRO [contract research organisation] decouple share digital networks, IT infrastructure and facilities without creating operational drag for remaining customers?

[[00:06:56]]
I think usually, it's not that as seamless as sometime it is presented. However, for a CRO because this is a very data-driven space and for clients which are the pharma companies in this case, they are very big on the data integrity, and it is mission-critical for them. The examples here, I would say, is like GLP, GMP quality systems, LIMS platforms, electronic notebooks, if they are implemented, clinical sample tracking, regulatory archives, validated manufacturing systems. Many of these systems are never designed to be separated, but like I said, that these were acquisitions and they had their own regulatory and all LIMS system was independent, so this could be an easier one. Then source of operational drag you asked, most common challenges include, for example, validation. If you have to do the duplication of the work and data migration risk because data was shared between different sites, cybersecurity we talked about earlier and then the systems, which are going to need new user permissions because now you are separating into new IT structure and do they need to be requalified. Then this kind of separation is a lot more tedious, and this could take like 12-24 months to fully separate. This is the area which is the most grey area and is regulatory burdensome. This takes time usually in these separations.

[[00:08:52]]
What do you think causes a long-term biopharma customer to take its business to an alternative provider during a complex data or facility separation?

[[00:09:04]]
Large pharma, biopharma companies usually are very risk averse. Let me say this, pharma in general is very risk averse, and the longer you go into the drug development process from preclinical to commercial side and we are talking about the post-launch and commercial side, they're very risk averse. They're very sensitive to taking any risk, so the pharma companies usually don't transition until they see this as a regulatory risk or there is a delay in the programmes, which can put them into concerns, which include data continuity, for example, who owns the historical batch records in case of separation of two entities, where the analytical data resides and in case of FDA inspector shows up, how they will retrieve the prior batch records. These are real challenges, and this could be seen as an ambiguity and that could force the pharma company to look for an alternative supplier. The second area is a quality uncertainty. For example, the quality is a very big area for pharma company because they do not want to sit in front of an FDA auditor and get a 483 on a quality. That is their biggest fear.

In this case, for example, if in these cases of buying and selling businesses, the quality personnel leave and there is a turnover there is SOP changes and that creates a validation gaps and there is an investigation. There's a lot of CAPAs are opened, and that takes time to resolve. They don't like to take these kind of risks, and this is where they will trigger, for example, dual sourcing. After the COVID, there has been a practice of qualifying a primary supplier and a secondary supplier, but it's not that simple for manufacturing space. The other thing is change management, which is a big burden because for them, they have to look for approved vendors, which takes time in getting that in GMP space, established audit histories, validated supply chains, and these all activities take a lot of time. If any of these areas, which I just covered, the pharma will see as a risk, they are going to really look for an alternate partner because they really don't want to take a risk. Cell and gene therapy area is very sensitive and that can ruin their whole clinical programme. This is where they don't want to take the risk.

[[00:12:00]]
Considering why biopharma customers experience switching friction during multisite data separations, how does this disruption impact contract renewal risk?

[[00:12:10]]
The data is the key here because, a, they're always worried about that the data crossovers to a different client from one to another, data integrity and data cybersecurity. The contracts are usually negotiated very heavily on these areas. If there is a buying and selling happening, that is the biggest area of concern. During the selling, I would say, the process, they have to bring sometimes the client's view into it to figure this out like how these contracts will translate to the receiving unit and how this will be perceived. I think this is an area where the contracts are usually having clauses and whenever there is a legal entity involved in these contracts, they always write some clauses. Usually, this area is covered under those clauses. Still, those are the key areas which are specific for the contracts.

[[00:13:22]]
Why do corporate overhead expenses historically lag the completion of facility divestitures? What structural hurdles inside a contract manufacturer's footprint typically prevent the rapid elimination of trailing real estate costs?

[[00:13:37]]
This is an interesting part of this whole thing because for the contract manufacturer, real estate is more than an overhead because these facilities are not like a simple facilities, these long-lived facilities may have a longer leases because you can't just have such an expensive facility built out on a shorter lease. Usually, these facilities may have like 10-20 years leases and then the equipment needed financing because sometimes CAPEX is so heavy and then usually some equipment could be financed. Then there is a maintenance contracts to maintain the facility and the equipment and the utilities in those because they're all under regulated space, and those are under FDA inspections. Those are not simple facilities. Now in context of cell and gene therapy, which is highly customised facilities because viral vector suites needs to be there, clean room suites needs to be there, GMP manufacturing areas are super specific. Plasmid production would be very specific because they need separation and different workflows.

These are not simple facilities. Because of high customisation, they are very well-designed, and they would have heavy operational costs. They will have also longer leases. Then on top of it, there is a compliance requirements for these facilities. For example, whether you use these facilities or not, you still need to have environmental monitoring to comply with the requirements. You have to have facility maintenance, security, utility support and regulatory record retention for all these records we are talking about. Whether you're utilising this or not, but if this is under FDA-cleared facility for certain specific operations, they have to maintain it and then warehouses where your finished goods are going to be stored or your raw materials are going to come, then there is, of course, laboratory notebooks, we talked about utility plants. All these would continue to be maintained until the receiving unit takes it on. There is always a longer lead time to clear these real estate and the facility and the utility management during these kind of transactions. It could take multiple years.

[[00:16:18]]
If these legacy facilities are underutilised and seeing volume downturns, how can that impact be mitigated to protect operating leverage?

[[00:16:29]]
I think that is very hard to really maintain because this is where you are operating in a negative margin. You got to maintain because to get these licences through FDA and a regulatory pathway costs itself a lot of burden to get everything validated and under quality management system. You technically end up eating that cost. It is going to be a drag on the P&L.

[[00:17:00]]
Charles River management projects that removing the CDMO margin drag will clear a clean runway to lift structural manufacturing solution margins above 30%. How realistic do you think that is?

[[00:17:19]]
Let me say this, it's possible, not unachievable, but not guaranteed. It's a very ambitious target and aggressive target to achieve. The reason being is 30. If you want to achieve 30%, your asset utilisation has to be greater than 80-85%. It needs to be really specific your offering than just doing the manufacturing. The good news for Charles River is this in manufacturing, this is why I think it's achievable because the manufacturing solution had three different businesses: CDMO, biologics services and microbial solutions. With CDMO gone, the biologics services, which is very pharma-driven, routine testing for GMP applications, which is quite stable. Then Microbial Solutions is the very unique business, which has actually the products, not only services, and it has products which are sellable to pharma companies. They have hardware, software and consumables, which is razor-razorblade model, so there is a very well-positioned installed base from Microbial Solutions, which is a recurring revenue. In spite of, I would say, downturn, if the pharma customer have adopted that, they're going to continue to buy it, which is a more, I would say, stable revenue. Combining biologics services as well as Microbial Solutions, I think that will improve the portfolio provided DSA and RMS divisions do their part of the P&L contribution. This is why I'm saying it's aggressive, but it's achievable.

[[00:19:21]]
What's the standard facility process for rebaselining laboratory asset utilisation metrics once a legacy cell lines division is entirely closed or removed from the building?

[[00:19:35]]
I think this takes time. Typically the process is to remove all the legacy equipment, which is not utilised by the other business units because this was a manufacturing facility and that equipment, it was under GMP that needed to be decommissioned or they need to be recommissioned to new GMP areas, requalify laboratory space, reassign the scientists and quality resources, then rebuild the utilisation targets for the new business mix, if that is what is a plan. It takes time. If you are going for a non-GMP application, it could be very easier. I would say it could take 1-3 months, but if it is going under still GMP and regulatory heavy activities, then it needs to be compliant to the new standards, which would be implemented for the new lab space. Equipment needs to be reinstalled and requalified and then you have to do really assessments to see if this space is going to be the right fit for the new applications or new business. If the customer demand stays and there is, I would say, a better utilisation by a new business, it could be easily converted. However, it will take time, depending on what the use case is, GMP vs non-GMP.

[[00:21:19]]
How long would you say is the lag time between freeing up physical bench space and recording active revenue-generating study runs?

[[00:21:28]]
I would think that it could easily take from 3-12 months, like if it is just a custom work for creating new methods and research activities, those could be quicker depending on the utilities and all that facility design is suitable. If it is a GMP, that could take up to 12 months, which is, I think, quite aggressive target. I have seen that it could take up to 18 months to revalidate everything and now it will create a new revenue work stream.

[[00:22:06]]
How does the operational carve-out of legacy cell lines division structurally transform platform cost recovery metrics across shared manufacturing spaces?

[[00:22:17]]
I think that because those cell lines are not only used by CDMO or there are other methods which are being utilised in biologics, those are cell line based or cell-based, those cell lines will be still utilised and those were existing already in Charles River than Cell Solutions, which was acquired as part of the CDMO strategy. However, those cell lines are under GMP, and they are pretty well-maintained and those will be utilised under biologics testing. There are so many assays which are in the portfolio for Charles River, which are being utilised through pharma biologics assays to do multiple study designs, and those will be still very well-suited. I'm not worried about those. I think those are still very much into the strength of Charles River.

[[00:23:18]]
How does the ongoing wave of biopharma manufacturing onshoring to the US drive long-term sticky placements for automated rapid testing platforms over standard contract testing?

[[00:23:30]]
That is a new trend, I will say, because a lot of onshoring… Actually manufacturing returns to US, and I think these facilities increasingly will have a faster batch release, reduce QC bottlenecks because you're in within the same country, you can troubleshoot better, you can standardise the protocols, standardise the methods, real-time microbial testing can be done because now your turnaround time for microbial testing, which is required, that would improve. You can do digitisation of GMP workflows because you are within the same country and those can be under the same, I would say, ID space. This creates much stickier demand, in my opinion, and you can automate these quality control platforms over, I would say, traditional outsourcing testing. This, I think, would give some leverage to biologics testing, for example, in this case, in manufacturing, which is a very US-centric, I think, business, heavily U.S.-centric, although EU-centric spaces are there, and most of the testing for EU goes to the EU sites, but it will definitely improve, I would say, the batch releases faster and pharma also will go faster back to the pharma companies. That's what I think.

[[00:25:03]]
What operational friction or validation delays prevent a customer procurement team from adopting rapid quality control instruments earlier?

[[00:25:12]]
This is a very important topic, always comes and the major barriers include method validation requirements. Those are coming from USP standards or FDA regulations and a lot of other standards, which pharma company needs to comply with, for example, ICH, USP, EU standards. Whether you do the methods in development in US or EU, but you still have to comply and you have to do method validation and you have to create a package. Then FDA expect comparability packages, then quality approval cycles, of course, very, very heavily driven on a quality management system. You have to have SOPs and reports, staff trained and risk aversion during the regulatory inspections. I already talked about that because pharma does not like to get 483s and they want to take minimal risks. Most pharma companies would rather delay implementation than risk disrupting commercial manufacturing. This this is why it is a very slow adoption of new methods. It takes time. It could take up to 18-24 months, depending on how many resources are allocated.

[[00:26:34]]
The 18-24-month timeline is fairly lengthy. Can you discuss the upstream and downstream impacts from this delay?

[[00:26:48]]
That depends on how you're going to convert. You asked me earlier about when it will become revenue generation, so when you bring a new service or a new method and you have to do method validation and so on, it takes time. You've got to plan those ahead of time, and if you didn't plan, then you will be surprised that it takes that long and then it will take revenue generation. The plain fact in this space is that you have to have everything validated and if you're going to use that for GMP, and we are talking about manufacturing space, it's heavily regulated, and it is going to take most of the pharma-only moves when the regulations change. To do that, even if you create the method validation packages, until you don't submit those data to FDA under a given new drug product, the other pharma companies are not going to adapt because they wait for somebody else to get the acceptance on that method before they adopt it. They will continue to do the traditional testing to take minimal risks. This is why it takes a long time to really implement a new method or a new technology.

[[00:28:09]]
Is there anything else you'd like to highlight in terms of what prevents a sponsor from taking its batch validation work to offshore providers if domestic lab capacity becomes constrained?

[[00:28:19]]
I think there are only two or three things, which really trigger anybody to take to a different supplier. Number one, quality issues. If a provider continue to have quality issues in a campus and there is a delay in the timelines of giving the results, that is another area. The third is regulatory compliance. Nobody wants to take a regulatory risks. Those are the three major areas. Of course, price. everybody gets hung up on price, but I think price is more secondary, but these three things are very, very important, quality, product delays and regulatory compliance.

[[00:29:05]]
Regarding primate supply and toxicology study start predictability, how does moving to an 80% internally controlled NHP [non-human primate] supply chain alter study staging, velocity and timeline reliability within a safety assessment pipeline?

[[00:29:22]]
I think this is a very good move, usually because if you have all this in-shore 80% internally, I would think you will have a study start predictability, which is much better. You can have scheduling certainty. You can forecast the studies better. Your customer confidence is higher in that sense. Resource planning, you can plan resources much better when you're dependent on external vendors. I think that will definitely improve the things by having in-shore internal 80% NHP supply.

[[00:30:07]]
What does secure animal supply mean for win rates on major MSAs [master service agreements] when competing against unintegrated offshore alternatives such as WuXi AppTec?

[[00:30:20]]
I think this is a major competitive advantage over like, for example, WuXi AppTec or other competitors because you have guaranteed allocation of animals to the study, which gives you better predictability of study start. You can have a better timeline commitments to your client. You can have less, I will say, supply volatility. This is very important when large pharma tenders are filled. Reliability very much matters to them as equally as price because they don't like disruptions because now it will take them longer to complete the studies. This definitely will improve the MSA win rates.

[[00:31:06]]
What specific bottlenecks or quarantine disruptions at domestic ports of entry are completely bypassed by owning international breeding infrastructure?

[[00:31:23]]
I think just to make sure that there is going to be a regulation which will apply, which is CDC guideline USDA and the transportation regulations still apply. However, owning the breeding infrastructure reduces exposure to import restrictions, air freight delays, port congestions and third-party shortages. You could, I would say, predict your quarantine disruptions. You can plan better, like I said earlier. Planning is key. This definitely improves study start and finish predictability and you can become more reliable.

[[00:32:11]]
How does phasing out open market animal sourcing in favour of an 80% captive supply chain model compresses structural unit costs and drives visibility margin accretion over a multi-quarter horizon?

[[00:32:27]]
I think over multiple quarters, the breeding can show lower procurement volatility, like earlier I said, reduce broker markups because now you're not paying to a third party in between, you can improve your utilisation, you can reduce study cancellation because that is an important part when I'm talking the reliability. You can actually reduce that study cancellation because you didn't get animals or you didn't get a number of animals you need to study. The result is gradual structure margin expansion rather than an immediate step-up. I think over time, it will give better margins and you become a reliable partner. In this space, I think the trust is a very big thing, reliability and your regulatory compliance is a big thing. Those are really, I would think, the power of Charles River in this space because they're really good at regulatory compliance and having all this in their wheelhouse will definitely put them as a very strong partner.

[[00:33:45]]
How insulated is a standard Q4 operating margin exit rate if open market primate pricing experiences unexpected volatility towards year-end 2026?

[[00:33:56]]
I think those are very unpredictable indicators. However, like I said, you own the breeding, you own all these process workflows and that will at least make you reliable. You can predict better. You can communicate to your clients better, and you can keep them into the loop rather than unpredicted disruptions, which is very hard to manage. If you don't know and then you cannot communicate to your client who is waiting on results, so I think this whole cycle actually improves. Definitely, once you lock a customer because these are multi-year studies, it's not like onetime study, so pharma always look for a long-term strategic partners. That definitely improves Charles River's position.

[[00:34:53]]
How can a CRO structure multi-year MSAs to insulate operating margins from extreme price volatility in the open animal sourcing market?

[[00:35:04]]
I think that is where you have to build that in your contract when you write the study contract. Open market, you are dependent on a third party and those prices could fluctuate. You have no control over those. If you own the breeding, you can control those and predict the pricing better in the contract when you're establishing that. You can also show the expenditure, which goes around it on the animal studies and you can write better budget planning as compared to open market. In open market, you have no control and any fluctuations in the supply or tariffs or any government and political aspects could actually impact your margins and impact your clients as well.

[[00:35:55]]
Why do post-acquisition compliance delays or biosecurity alignment hurdles at the newly acquired KF Cambodia Farm pose a risk to study start timing reliability within Charles River's domestic DSA [Discovery and Safety Assessment] pipeline?

[[00:36:10]]
I think that is an important aspect of this discussion because the key risks include they still have to have a biosecurity harmonisation, regulatory compliance upgrades, they might have to do it because those animals will be used in GMP and in unregulated drug development space. They still have to comply that. Animal health monitoring is a very important aspect of having healthy animals, you have predictable indicators for study design and outcomes, and that is super important. They have to bring that to the standards, which FDA or other regulatory agencies are expecting. Transportation qualification is another one. Any of these 4-5 areas, if there is a delay, this can have a ripple effect on Charles River DSA scheduling and the study start because this could disrupt. In my opinion, after this KF Cambodia acquisition, they have to put a lot of effort from a regulatory standpoint and bringing them to the standard, which is expected from regulatory agencies.

[[00:37:28]]
Why do large biopharma accounts face significant operational switching friction and regulatory hurdles when attempting to reallocate validation budgets from traditional in vivo studies to new approach methodologies?

[[00:37:43]]
This is a very new subject, to be honest, because this was a regulation change in 2025, April 10, when FDA Modernization Act 2.0, FDA brought it up that the monoclonal antibody drug development, they would be willing to accept NAM and AI to reduce animal reduction in the studies because their point of view was that the animals are not closely predicting all the toxicology and everything as they would do that for other small molecules and so on. NAM is a very, I would say, broad subject. The biggest challenge, whether it's one technology or other or this method or that method means organ-on-chip, organoids, next-gen sequencing, you name it, new technologies and new methods. The biggest barriers are these. Large pharma always want to have a regulatory validatable methods. They have to compare this with the existing validated methods. They don't want to take regulatory uncertainty, like I said earlier.

Then if there is a change in a drug molecule like they are just doing the modification of the same structural molecule, they want to have historical toxicology data, which they want because if that was done on the animal models, then this became a big barrier for them to adopt to a new method and figure it out how to correlate these data. Of course, there is always internal QA resistance because all this really cost you money and then it will increase the validation burden in the beginning. Whether pharma will switch that in an ongoing drug, which is very difficult, they may start using these, once enough data are generated and enough regulatory standards are created for a new drug, so not so much into the existing drug.

[[00:40:07]]
What technical validation blocks or regulatory hesitation from large pharma teams, thinking of QA [quality assurance] units, are currently slowing down the commercial adoption velocity of molecular assays and NGS [next-generation sequencing]?

[[00:40:23]]
I think the major part is a bioinformatics AI-based libraries for next-gen sequencing. There's insufficient regulatory precedent. Regulatory pathway and the standards are not fully standardised. Limited cross-company validation datasets are created. Of course, like I said, lack of harmonisation in the global standards, like USP or FDA even doesn't have all these standards, which are referred for regulatory guidance. There is a traction from EU in context of reducing animal testing. Then, of course, using AI, you can have it, but how do you explain that for a GMP, so you can't really go into the deep learning mode. You have to have a supervised learning going with these databases, and they should be explainable to a regulatory body. Again, I cannot emphasise the fact pharma does not like to change until it is regulatory compliant.

[[00:41:37]]
Why do technical validation hurdles and regulatory safety standards prevent biotech sponsors from fully replacing traditional in vivo models with lower-cost automated screening cells, especially if the cost profile is significantly cheaper here?

[[00:41:55]]
I totally would think that, that logically that should happen. Again, how to generate the data, which can explain the previous data, which was generated through animal models and how to show that these data are acceptable for a regulatory body and methods are validatable. That is the biggest barrier. At a research level, of course, this is easier to adopt. When it goes to GMP application, that's where the biggest hurdle comes. For non-GMP applications and upstream research or pre-drug discovery, these are very, very much adoptable and easier, I would say, to generate preliminary data. However, as you move to the GMP space, that's where the hurdle starts to become harder.

[[00:42:49]]
How can a preclinical provider realistically redeploy legacy rodent vivarium space to build high-margin automated screening cells to support large-scale computational target partnerships?

[[00:43:03]]
This is a very interesting question. I would say the vivariums are for animal colonies and their design of the vivariums and the associated, I would say, utilities and everything is very, very different than in automated screening labs where you're thinking about AI-enabled discovery platforms. The infrastructure is very different. Facility makeup is very different. You have to recreate these spaces into a very different use case. In this case, you could have challenges like HVAC design because now these data centres are going to need a lot of energy. You're going to have to do utility upgrades. Equipment needs to be installed, uninstalled, decommissioned, cleaned these areas, decommission them means decontaminate these areas and then repurpose them and then you have to do the validation again. For advanced analytical centres, you have to have really the powerhouse and the utilities and the cloud services, all that available. The infrastructure changes are very different and needs cost, and it will need a lot of CAPEX, a lot of changes into the equipment and the facilities. Then you have to really commission them for utilisation. Can it be doable? Yes, but it will take a lot of investment, and it will take time.

[[00:44:43]]
What would you say are the facility modification economics and equipment downtime metrics that are required when retrofitting older vivarium real estate?

[[00:44:56]]
I would think that first, you have to… Vivariums are like animal hotels, let be simplistic terms. You have to fill those and the more studies you do, your hotels are filled. Now when you decommission a vivarium, like I said… The KPIs are very different for vivarium and facility utilisation KPIs are very different as compared to data centres or for a new analytical lab. The KPIs cannot be just one-to-one replacement. That's number one. For converting these vivariums to a different laboratory space, you have to really assess what is return on investment and what the KPIs would look like before you go that route. Like I said, it is not one-to-one replacement. The KPIs would be how many facilities or hotels you have converted to these labs and what they're producing. You may not need all vivariums to be converted into the data centres, maybe one or two, and the KPIs would be driven by what the expected outcome from that laboratory is, what are the key parameters you're going to measure.

[[00:46:17]]
How does deploying large-scale historical data tools for virtual control group validation ultimately compress the operational timeline from a customer proposal to active revenue generation within a regulated safety assessment framework?

[[00:46:35]]
I think this virtual control groups will definitely do faster protocol design because the study design itself takes a while. That will definitely improve the type of the studies and then faster protocol design. It will definitely reduce proposal cycle times because there is a lot of back and forth between establishing the contracts, and that depends on the statement of work and the protocol design. That definitely will go hand-in-hand. You can have better study planning. Because now you can alert your teams also that this is coming and that allows you to do better resource planning, all the things could go wrong and you can figure it out how you're going to mitigate those risks. You can have more efficient customer discussions because, obviously, you can figure things out ahead of time, proactive planning. Customers do proactive planning and transparent discussions. I think that will definitely improve. Then, of course, regulatory part would also improve because now you can have proactive planning, you can have more discussions with the regulatory bodies if you are interfacing them or your client is facing, you can participate. I think those can help in a better adoption and a better smooth submission and clearance for a client as well.

[[00:48:15]]
How long can a core safety assessment footprint insulate its margin profile from fixed cost facility deleveraging if biotech demand remains soft or uneven?

[[00:48:28]]
I would say usually a good safety assessment network can absorb softer demand for 4-8 quarters. Either you do pricing discipline and you do workforce management and facility optimisation, which if you look at Charles River's past earnings for the last two years, and they have been doing that, and that is not a unique thing to Charles River, everybody does that. Beyond that, if this is a persistent underutilisation, then it will start putting pressure margins. I would say if you look backwards and then you can look at how many quarters it has been under the pressure and then how the soft demand continues and how it will project, but after eight quarters, I will think it will be harder, then there needs to be more drastic measures.

[[00:49:38]]
What workflow automation or digital productivity tools do you think can be deployed to mitigate underperforming volume growth?

[[00:49:49]]
I think the digital tools can do so much on this aspect. You can standardise the things, you can bring, for example, LIMS, you can bring electronic notebooks. No, I would say it actually have a lot of impact at the bottom line because those could become a barrier if you don't do it well because if you're still doing paperwork and you're wasting time and you need to have people to do the paperwork, digitalisation can improve. If you do it right, I just want to emphasise the point. If you do it right, because if you have different LIMS system, electronic notebooks, which are not talking to each other, then you are site-specific. That really does not help you to look at the data across and you cannot use the protocols from one site to another. You cannot replicate the workflows from one site to another. They are not going to give you the effectiveness and the, I would say, return on investment you're looking for by implementing digital tools.

If you implement correctly and standardise across multiple sites, they can be super useful. They can also cut down your paperwork and your hand, I would say, batch records, which causes a lot of, I would say, errors, especially in the manufacturing space. That could reduce the burden of human errors and the CAPAs and root cause investigations, which we were talking earlier and also improve your turnaround time because if you have less investigations, you can actually release the results faster. Digital tools are very useful in improving, if you implement right. If you don't implement right, then it can become also a barrier because you invest so much money and they are not fully realised among different sites. That could also become barrier because you can't see the data across different sites and you cannot implement the same protocols.

[[00:52:02]]
How do you stress test the operating leverage of a contract lab business over the next 12-18 months if net book-to-bill tracks near a 1x baseline?

[[00:52:15]]
I think this is a very simple in my mind, maybe I'm oversimplifying. 1x book-to-bill ratio generally indicates a stable backlog, but limited growth. I would focus if the leading indicators such as a backlog quality, cancellation rates, average study size, pricing trends, utilisation and customer mix, then book-to-bill ratio alone. Those are very important indicators to look at it. If we are going to just look at only book-to-bill ratio, then the base is like 1x, that means it's a stable revenue with modest margin improvement driven by productivity and cost control. You can do so much cost control on the bottom line. You have to improve the top margins, and if it is greater than one, that means you are growing. You have a stronger indicator for growth. You are not only improving your operating leverage, but you're also growing the business. Less than one is underutilisation of your facilities, your delayed starts of the studies and you have more problems and then you have to really continue to do restructuring, consolidation and cost cuts.

In simplistic terms, if I was looking at Charles River business, I would not only look at book-to-bill ratio, I would also look at other indicators I just mentioned. Bull case in this case, would be stable backlog, high utilisation and margin expansion. Base is like, yes, you are meeting your numbers. Your revenue is roughly flat. Margin improves through restructuring, not through growth. I think that's the key one, and you can continue to be in a low single-digit earnings growth. Bear case is like you have continued to have study delays, but you cannot continue to say softer growth for pharma, provided there is a bigger rationalisation happens in pharma, which happened, but I don't think so that will continue to happen. That is actually improving. There is a sign of recovery. There is a more funding for smaller biotechs. I think those are good signs. Now it is up to the business to operate and perform.

[[00:55:07]]
How do aggressive price discounts from smaller unintegrated independent CROs disrupt the ability to secure large pharma renewals without executing margin-dilutive concessions?

[[00:55:23]]
I think that you can play for a while, but you cannot continue to only look at pricing. because everybody is playing in the same macro environment. It is not different for pharma companies. They will understand those for a little while. They will also understand, for example, your pressure from your political, geopolitical issues, but you can't continue to play the pricing game for the long term. That will definitely start diluting because there will be pushback of price increases to pharma. To me, for example, the best indicator would be DSA study start volume, what that shows, NHP supply reliability, book-to-bill ratio, which we said needs to remain above 1.1 and successful removal of standard overhead costs we talked in the beginning. Those are like 4-5 matrices than just pricing.

[[00:56:34]]
How do sequential scheduling delays and elongated procurement cycles on high-margin toxicology runs ultimately impact the trailing 12-month pricing power of a clinical testing segment?

[[00:56:51]]
I think that should improve because if you look at the strategy under the new CEO, Birgit Girshick, she's very pragmatic, and she has been indicating on the improvement on efficiency on the workflows and better investment in all the processes, which includes toxicology, and they have been indicating through the strategic review that Charles River is going to continue to focus on their core strengths, which includes preclinical drug development, which includes your DSA, toxicology, research models and they divested the businesses which were underperforming. I hope that overall, not only toxicology, but all areas should improve in the coming quarters. They will get back to their operating margin, which was expected by the investors. If they don't, then I think there are some other issues which we talked about, which they need to resolve.

[[00:58:07]]
Thank you for that. We will now conclude the Interview. Thank you for hopping on the line and all of your insight. Yes, it was a really great Interview, and hopefully, we can work together again in the future.

[[00:58:21]]
Absolutely. Thank you for the opportunity. I really appreciate that.

[[00:58:23]]
Thank you so much. Have a good one.

[[00:58:26]]
You too. Bye.

[[00:58:27]]
Bye.

Transcription ends at 00:58:29 of the recorded material