3M Company (MMM) Earnings Call Transcript & Summary

Hello, everyone. Welcome to BioProcess International Spring Digital Week brought to you by the producers of the face-to-face BioProcess International events. My name is Barry Walsh, and I'll be your host for today's session titled Novel Approach to Bioprocessing - Replacing Conventional Anion Exchange Chromatography with Single-use Solutions. First, I'll cover some quick housekeeping items. If you experience difficulties with audio or advancing slides, refresh your screen with F5. If you are experiencing other issues, click the question mark button to receive assistance. [Operator Instructions] In 24 hours, you'll receive a link to watch the recording of this session. You can also download a few featured white papers in the Resource List box on the right side of your screen. Now let's begin by introducing our speaker from 3M. Joris Van de Velde is a Senior Applications Engineer at 3M Separation and Purification Sciences Division. Thank you for joining us today, Joris. Now I'll hand it over to you to begin the presentation.

Thank you, and thank you for the introduction. So today, I'd like to talk to you about a new product that we will be launching soon, namely 3M Polisher ST. With this single-use AEX chromatography device, you'll be able to replace traditional resin-based columns in your biopharmaceutical manufacturing processes. So I'll start this webinar with a brief introduction about the division I work for and how we help our customers solve their filtration and purification challenges, and I'll give some background on typical downstream processes, mainly focusing on monoclonal antibody production and highlight some of the challenges that are often seen. Next, we'll show some proposed strategies to address these challenges and introduce the Polisher ST product and its main characteristics. The ST in the product name refers to being salt-tolerant, as you'll learn. We will also dig into the more detailed performance of this product, showing the typical impurity reductions and viral clearance that can be expected. There will be also a short case study showing what implementation of this product could mean for a real-life process in terms of process simplification and efficiency. And to finish this webinar, we have a short summary for you. So in the Separation and Purification Sciences Division of 3M, we have 100 years of experience in filtration, and we apply that expertise across diverse industries. One of these is the biopharmaceutical purification market where we provide filtration and separation technologies for the manufacturing of recombinant therapeutic proteins as well as blood plasma fractionation and API. Our division is known in the industry for our innovation, collaboration and rigor. One of the key strengths is 3M's cutting-edge materials science which allows us to create next-generation technologies that really push boundaries of what is possible, and I believe today's presentation will offer an excellent example of that. We like to work very closely with our customers and apply our expertise in the lab to help solve our customers' toughest challenges. Now by doing that, we intend to build long-lasting, trusted relationships. Our global supply chain and our proven reputation provides the quality and reliability that are needed in this market space. Finally, we offer smart design for our products to make them ergonomic and user-friendly. The picture you can see in the bottom-right corner may be a bit small, but some of you may recognize our rotating capsule holder, which is used for our Zeta Plus and Emphaze products. This holder is loaded horizontally at waist height for the best ergonomics and then rotated to a vertical position for the filtration operation. That's one example of how we use smart design for our products. Now let's look at the manufacturing processes for monoclonal antibodies. Now in an ideal world, this would be relatively simple. The main steps in the process are to grow up cells and bioreactors of increasing volume and to induce product expression. And when we don't -- no longer need the cells, we remove them and remove the cell debris. That would ideally be performed with a single-step filtration under low-shear conditions. Then we have the capture step, the protein-A column. If that column worked perfectly, it would only bind the mAb, and all the impurities like the host cell proteins and the DNA would be removed. Then the only thing left to do would be to get the product in the right buffer for formulation and filling. Now as you all know, things are a bit more complicated in reality. The world isn't perfect, and the typical mAb process has more than 10 steps, including multiple clarification stages and one or more polishing columns, for example. For some reasons for this include performance limitations of available technologies. For example, the clarification train usually needs multiple steps. And if it's not properly designed and cells break open due to shear, the impurity challenges can even be magnified. The capture step also never gives you 100% clearance of impurities. They need additional polishing steps to remove those. A nonideal behavior may result in challenges due to product-related impurities or difficult-to-remove host cell proteins. The process design choices may also be limited by decisions made in the past and platforming requirements. The regulatory compliance finally also requires you to include certain steps, for example, for virus clearance. And then another challenge is related to the platform ability of these processes. Although many companies use similar process layouts for their different products, there's also been -- felt some limitations. For example, each molecule typically still requires significant process development effort, and one of the reasons for this is because the available technologies are not sufficiently robust to deliver the expected performance under the wide range of conditions that you may find in these processes. Now at 3M, we cannot promise you the process of the perfect world, which I showed you earlier, but we can help you to get as close to that as technically possible. And we do that by trying to achieve very high purity early on in the process and simplify by reducing the number of required steps. Because with every step, you'll lose some product due to protein binding or just the associated liquid hold falling. And if you can eliminate some steps, obviously you decrease cost, but you also improve the product yield. Our process intensification strategies revolve around replacing multiuse unit operations with high regulatory burden by robust single-use solutions that drive efficiency and reduce the total cost of ownership. So in the past years, we've mainly focused our efforts on optimizing the clarification unit operation, using an open-grade depth filter with a high capacity like our Zeta Plus, which allows for direct targets without the need for a centrifuge. And when designed and operated properly, such a depth filter, can capture cells in a very gentle way with very low shear. The second depth filtration stage for removing small cell debris can then be replaced by the 3M Emphaze AEX Hybrid Purifier, and this chromatographic clarification unit removes 4 to 5 logs of impurities like DNA and chromatin based on the charge. Now chromatin is an important enemy of the protein-A cell. And by removing this component, the chromatin, you'll actually get an improved performance of your capture column, which will also result in 10- to 20-fold lower host cell protein levels than what you would normally see in your elution pool. The reduced levels in impurities can actually be seen visually after virus inactivation and neutralization. So typically, you see some turbidity occurring in this step as you can see in the left side. Now when Emphaze AEX Hybrid Purifier is used in the clarification, this turbidity will be much lower, as you can see in the file on the right. Now as a direct result of that, there is no longer a need for a depth filtration step after virus inactivation and neutralization to deal with this turbidity. And also, since the impurity levels are already an order of magnitude lower than usual, you can also think about optimizing the rest of your downstream process and making it more compact. So the next logical step for us is to replace the anion exchange polishing column by a much smaller single-use purification device. And this is exactly where the Polisher ST will come in. Now as you can see, there is no need for a depth filter or a membrane to protect this device, no filter between virus inactivation and the Polisher ST. So we combine multiple steps into one, and this ties in nicely with our overall goals to simplify processes and drive efficiency. So this is what we're aiming to do: replacing a reusable anion exchange column by a much smaller, single-use capsule for flow-through polishing application. The Polisher ST will have 100x capacity of a traditional Q resin with a recommended target loading of 10 kilograms per -- of mAb per square meter. The increased capacity means that a capsule can be used which is much smaller than a traditional column, as would be shown in the picture that, unfortunately, you can't see right now. It will be scalable from lab up to full commercial production scale. It is salt-tolerant, and it can operate at low pH conditions. The ST in the product name refers to the salt-tolerant. It will deliver robust viral clearance as well across very wide range of conditions, and it will remove turbidity, which is important for protection of certain other process steps like the virus nano filter. Obviously, this technology also offers the typical advantages that are associated with single-use membrane absorber products, which include, for example, short residence time, flexibility, elimination of compacting, cleaning, validation cost and reduced buffer usage. Now it may seem ambitious goals, we realize that our current AEX ligands will not be sufficiently powerful, and so I will now explain a little bit more about the chemistry behind. And so this is a slide that you should be seeing currently. I'm sorry about not being to see the previous slide. Sorry. This should be the slide. So like I said, current AEX ligands will not be sufficiently powerful because most of the current AEX technology is actually based on quaternary ammonium chemistry. And the main drawback of Q chemistry is that it is sensitive to conductivity, and it will lose much of its capacity at high salt conditions. The primary chemistry offers an alternative and is more salt-tolerant, but it doesn't perform well in polyvalent buffers, like, for example, phosphates. Now to cover the conditions where current ligand struggle, we'll introduce a new AEX ligand with guanidinium functional group. This ligand is 3M proprietary and bio-inspired. In fact, the positively charged guanidinium group, which you can see on the left of the screen, can also be found on the amino acid arginine. And in nature, this group plays an important role in stabilizing proteins and engaging in strong binding interactions due to its ability to interact with the carboxylic acid residues of other amino acids. You can see that in the middle of the screen. So in addition to the electrostatic charge interaction, it is also able to form hydrogen bonds, as shown here, with the blue dotted lines. By doing so, the coplanar ring of the 6 heavy acids is formed which provides a very stable bond. The positive charge is the lobe-like, which provides the salt-tolerance. Now if you think about the impurities that we want to remove in the polishing steps, you may realize that the negative charges on the host cell proteins and the viruses also come from carboxylic acid groups on the amino acids, aspartic acid and glutamic acid. So as you can imagine, the guanidinium ligand is actually very good at finding the impurities based on the interactions shown here. At 3M, we're able to build long-branched polymers with multiple of these guanidinium groups spread over the length of the polymer. You can see that on the left, and so this provides a high charged density and a high binding capacity. The polymers are immobilized on a microporous membrane, as I'll show on the next slide. So the capsules of the Polisher ST will combine multiple layers and multiple ligands and are designed for flow-through mode operations. So on top, we've got 4 layers of an open polypropylene nonwoven functionalized with a positively charged quaternary ammonium ligand. For those who are familiar with our Emphaze products, you may see that these layers are based on the same technology. And just like Emphaze, these layers can deal with turbidity and are very effective at removing DNA. The Q chemistry is sensitive to high-salt conditions. But even at elevated conductivity, these 4 layers will still remove DNA and turbidity and protect the downstream membrane. So below, we've got 3 layers of the membrane functionalized with the new guanidinium ligand. And these layers will then capture host cell proteins amass. The guanidinium ligand is salt-tolerant due to the delocalized charge, so it keeps doing its job also at high conductivity. So with these multiple layers and multiple ligands, we basically have a divide-and-conquer strategy, if you will, where the top layers remove the turbidity in the DNA, protecting the membrane, which then take care of the host cell proteins and the viruses. And since this product is based on nonwoven and membrane technology, the flow is convective. And so the interactions are not limited by infusion like the resin-based system. This means that low residence times are required, and high flow rates can be applied. The capsules will be available in a complete range, covering from lab-scale to large-scale manufacturing, and the biocaps or BC probe that you see on the screen corresponds to the frontal surface area of the membrane. So these surface areas in the capsules are basically the same as those as from our other products like the Zeta Plus depth filters. The BC25, for example, has a surface area of 25 square centimeters, while the largest production scale capsule has 1.6 square meters. Now at the target loading of 10 kilograms per meter squared, this number also corresponds to the target amount of mAbs that can be processed. So the small BC1 capsule is designed for processing roughly 1 gram of mAb, while the largest capsule can process about 16 kilograms of mAb. The recommended flow rates in milliliter per minute are also the same number. So for example, 1 milliliter per minute recommended flow rate for BC1 and 25 milliliter per minute for BC25. So scaling up and down is rather convenient as all of the required information for scaling is basically included in the name of the capsule. Scaling will be linear across the different capsules and will be robust, and this graph shows also consistent BSA binding capacity from the BC1 up to the BC1020 capsule as an example. This test was done using 3 different logs of each capsule size. And based on this, we see that the scaling should not only be easy but also reliable. Now we get to the really interesting part, where I'll show and explain the performance of this product, and we'll start with host cell protein removal. Each Zeta Plus show hopeful protein removal in a wide range of conditions between pH 5 and 7.5 in the Y-axis and conductivity levels between 3 and 20 milliSiemens per centimeter on the X-axis. And the plot on the left shows data for the Polisher ST, and this data is compared with a typical Q chemistry from a geography resin on the right. Both were loaded with mAb solution to their typical recommended loading, which is 10 kilograms per square meter for the Polisher and 200 grams per liter for the column. And this data was generated with a mAb pool after virus inactivation in tris and acetate buffers and an HCP impurity level of around 500 ppm. For the Q column shown on the right, you can see that the HCP removal is best at low conductivity and high pH. Now when we look at increasing conductivity, you can see that the capacity goes down significantly, as expected, above, let's say, 10 to 12 milliSiemens per centimeter, the HCP removal capacity really isn't very good, which is shown in blue there. On the other hand, when we look at the results for Polisher ST, you can see that we don't get any cliffs where the performance drops off. First, at lower pH levels, there will be fewer negative charges on the proteins, so a lower percentage of load volumes. But in all the conditions tested, the HCP removal was actually higher than for the Q column. So for example, at low conductivity and low pH, the HCP removal is around 20% better compared to the Q column. And the differences become especially large when we look at the high salt conditions. So you do see a bit of a decrease in retention as pH goes down and conductivity goes up. But even at a pH of 5 and a conductivity of 20 milliSiemens per centimeter, so in the bottom-right corner basically, we still get more than 50% clearance. And under these conditions, the Q column has 0% clearance. That difference is clearly due to the unique properties of the guanidinium ligand. So overall, we get over 50% HCP removal in the entire range between pH 5 and 7.5 and conductivities from 3 and 20 milliSiemens per centimeter. Now at the same time, we see that we get over 95% recovery of the mAb across these conditions. This was tested in monovalent and polyvalent buffers, as you can see in the bar chart to the right, and we see that the recovery will probably be lowest at the higher pH level, especially if you're getting close to the isoelectric point of the target protein. But in general, we thought that the product is salt-tolerant. It binds host cell proteins, even at low pH levels and the recovery is great. So let's look at the performance in some more specific conditions. Our target is being able to replace the AEX flow-through column in any mAb process, no matter the condition. Now what you see here are a few possible polishing train layouts. So after virus inactivation and filtration, some of the older legacy processes may start with a cation exchange column, which is operated in bind and loop modes. And then when eluting this column under high salt condition, the dilution will typically be performed before going to the AEX column. In this case, the virus filter could also be placed before the AEX steps, as you can see here. The more modern systems will often have the AEX column first, of course after virus inactivation, utilization and filtration steps. A minimal pH and conductivity adjustments may then be done to optimize conditions for the second polishing column. And in this process, all the steps can operated in flow-through mode. The newest next-generation processes are probably also completing a flow-through with conditions optimized to avoid any changes in pH and conductivity along the way. And if turbidity and DNA levels are well controlled, maybe by using technologies like Emphaze AEX Hybrid Purifier in the clarification, there may not be a need for a depth filter here, as you can see across that. Now based on these different process layouts, we came up with 6 model feeds, representing the conditions that you may encounter in these processes. So the table shows the conditions of the model solutions that we use to challenge the Polisher ST. I will not explain every line in detail, but I will give you some of the general ideas behind it. In the legacy process, the impurity level for the AEX column are probably the lowest because they have been reduced by the cation exchange step which comes before. In the modern systems, the AEX is the first column after the virus inactivation, which means you could expect higher impurity levels. And with Polisher ST, we want to replace the AEX column but also the depth filter and the membrane in front. So these 3 steps would actually be combined into one. And that means that there may also be higher DNA levels and turbidity to deal with. This turbidity was also included in one of these models and then for the next-generation process, the process has been more optimized, which means that you may find lower DNA and turbidity levels. You can also expect slightly higher pH and conductivity levels due to subsequent polishing steps. Now perhaps more important than which conditions can be connected to which system exactly is that with these 6 model feeds, we cover a very wide range of conditions. We have monovalent and polyvalent buffers, acetate, tris, phosphate and citrate, and we cover a wide range of pH and conductivity levels. We also have different levels of impurities, including presence of DNA and turbidity. So with each of these model solutions, we loaded the Polisher ST to the recommended loading of 10 kilograms per meter square. You can see that irrespective of the conditions, we get a consistent clearance. There is some stabilization time at the start. But after 2 to 4 kilograms per meter squared loading, you can see that we have reached a steady-state situation with very stable HCP levels coming out. There are no real signs of breakthrough, even near the end. Now to compare the overall HCP reductions in the pools, I have a bar chart on the next slide, which may be clearer. So you can see that for the pool reductions, we get pretty good host cell protein clearance for all of our model solutions. For some of the easier conditions, such as the tris and the pH and the conductivity 7, we get HCP clearance of over 80%. Some of the other conditions are maybe quite challenging. So for example, we've got acetate with a low pH of 5.5 or phosphate has very high connectivity of 20 milliSiemens per centimeter. And even for these conditions, we get more than 50% clearance. You can see that citrate is a difficult buffer for ion exchange steps, and it remains a challenge for HCP removal. The clearance in citrate at a pH of 5.5 was clearly the lowest. But even this result is actually better than what you could maybe expect with other ligands, which may have even lower or no clearance at all under these conditions. So overall, we got over 50% clearance of HCP for all conditions, except citrate. We had 99% mAb recovery in all the conditions. And although we don't show the data here, we were also able to clear DNA to detection limits for all the model streams. Now our next topic is the viral clearance performance of the Polisher ST. The AEX step is often one of the important and validated steps of the viral clearance strategy in mAb processes, and being able to show robust viral clearance under a wide range of conditions is therefore critical. So the scheme on this slide shows the standard channel of model viruses that are typically used. These most viruses are the ones that are of interest typically when using cultures. The 4 virus, as you can see here, include envelope and non-envelope models and cover very small sizes, like, for example, the MVM virus as well as larger ones. They will also cover a range of isoelectric points, which is of interest when trying to remove the basal charge interactions. The actual values of the isoelectric points are best seen as indicative for viruses, and different sources often report different values. Now these values may vary a bit depending on whether they are theoretically or experimentally determined and which kind of measurement method was flat. I suggest to have, first, have a look at the xMuLV, the Reo 3 and the PRV viruses. So we tested virus clearance with virus spikes and buffers at different conditions, again with pH values ranging from 5 to 7.5 and conductivities from 3 to 20 milliSiemens per centimeter, which is the monovalent and polyvalent buffers. And the results are probably the most boring from Zeta Plus one could obtain because for all 3 viruses, you can just see a plane of only one color, meaning that we get over 6 log clearance for all the conditions. Now the results for MVM are more interesting. Typically, this virus is considered one of the more difficult ones to clear. And at a pH 6 or higher, we get very good clearance, which is not affected by the conductivity. Again, this proves the salt tolerance of the product. And when we decreased the pH from 6 to 5, you did see a bit of reduction in clearance, and that can, of course, be explained because we're operating around the isoelectric point of this virus. Now what is interesting is that even at or slightly below the isoelectric point of the virus between, say, pH 5 and 6, we're still clearing around 3 to 4 logs. And that's quite remarkable since other chemistries, like quarternary ammonium, generally required to be at least 1 pH unit above the isoelectric point to get good clearance. We here show at least 4 logs removal for MVM at a pH of 5.5 and higher. Now the viral clearance was also tested in actual mAb solutions. And for these steps, we used the same 6 model feeds I've shown before. Again, these 6 models cover a wide range of relevant conditions and DSP processes for mAb, and they also contained host cell protein and DNA impurity levels that are representative of realized processes. Again, we load Polisher ST to 10 kilograms per meter squared, and you can see here, the results are on the right, in the bar chart, we showed robust viral clearance of 4 logs or more for all of the conditions, including low pH, high conductivity or presence of turbidity. Now interestingly, the robust viral clearance was also observed in citrate buffer. And you may remember that the host cell protein removal was much lower in citrates, so you can still reliably use this product for viral clearance under those conditions. Then finally, I wanted to share a case study with you. This case study was performed and published in collaboration with Bristol Myers Squibb, and what we have here is a protein-A elution pool with a host cell protein concentration of 2,400 ppm and DNA and a turbidity of over 60 MTU. And top of the scheme shows the original process flow, which includes a depth filter and a cation exchange column in bind and elute mode. At that point, the impurity concentrations have been reduced significantly, and a dilution is performed to reduce the conductivity. After dilution, a commercially available single-use AEX device was loaded to 1 kilogram per liter propulsion. You can see that the impurities are completely cleared, but the mAb concentration took a big hit due to the dilution. For the alternative approach, the Polisher ST was used and moved more upstream in the DSP train. So it was placed right out through the virus inactivation, basically taking the place of the depth filter. The cation exchange step is then placed behind it. Now the Polisher ST is obviously also the AEX step, so there's no additional polishing step after the cation exchange here. And with this train, the impurities are reduced to detection limits, just like slide before, but the resulting mAb concentration is much higher because the dilution step was avoided. So we've been able to simplify this process by reducing the number of polishing steps from 3 to 2 while reaching the same purity and obtaining a much higher mAb concentration. So this case study clearly shows the benefits this product can bring to your process. It enables process simplification. And due to its applicability in wide range of conditions, you can avoid dilutions and adjustments. In summary, in this webinar, we've introduced 3M Polisher ST. A single-use solution enables you to replace conventional anion exchange chromatography column by a much smaller and more convenient purification pipe. We've shown very high loading capacities of 10 kilograms per meter squared. The capsules scale reliably from the lab, all the way up to production scale with recommended loadings for capsule ranging from 1 gram of mAb to 16 kilograms for the largest one. The product performs well in a very wide range of conditions, as we've shown. It is salt-tolerant, and it can be used with different buffers and pH levels. We've also shown robust viral clearance across the same wide range of conditions. And finally, it is also able to perform in the presence of turbidity, which eliminates the need for a depth filter or a membrane to protect it. This means that multiple steps can be collapsed into one which further simplifies your mAb process and increases the product yields. So for those who are interested to learn more about this product or our other technologies, I would suggest visiting our website or contacting us at the e-mail address, which you can see on this slide. As mentioned before, the product is not available just yet, but it will be later this year. Now if you want to stay informed about the launch of this product and other existing development, you can go to our website and click the Ask the Expert button, where you can leave your contact information. You can also add any questions you might have, and we'll answer them as soon as possible. Now before we go to the questions for this webinar, I would like to thank the people who made this possible. So first of all, of course, I'd like to thank the organizers for their support, and then I want to thank many of the colleagues at 3M who developed this wonderful product, the people who generated these great data sets and those who helped to prepare this webinar. I also want to thank them for giving me the opportunity to present this to you today. And of course, thank you for your attention.

And thank you, Joris, for an excellent presentation. So we have received a few questions already, but we'll give the rest of you a moment to enter your questions in the Q&A box to the left of the slides. So while you're submitting your questions, I'll run through some brief announcements. First, I'd like to thank 3M for sponsoring this digital week. Next, I'd like to quickly draw your attention to our face-to-face BioProcess International events, which we'll be visiting Boston in September and Milan in October this year. Additionally, BioProcess International Europe will be delivered as a 100% virtual conference and exhibition on July 13 through the 17. Also, be sure to check out the resources list to the right of your screen, where you can download a few featured white papers. Now back to Joris to begin the Q&A.

And our first question is, can the solution be scaled up beyond 16 kilogram mAb loadings?

Yes. Thank you. Indeed, the BC1 -- sorry. Yes, you can definitely scale beyond the 16 kilograms -- sorry, can I -- sorry. I'll just answer your question. Yes, you can definitely scale beyond the 16 kilograms per mAb loadings because you can actually stack the larger capsules. So the smallest one, the BC1, up to the BC1020 are stand-alone units. And then the BC2300 and BC16000, you would use a holder. And with these holders, you would be able to stack multiple capsules to increase the capacity. The holders are actually the same ones as would be used for our other products like the Zeta Plus that filters Emphaze AEX Hybrid Purifier capsules. There are 2 designs of the holders. The smallest one would take one of the bigger capsules, so the BC16000, and there's another version of that, which takes 3 capsules. And the largest one, the production-scale holder would be able to hold 7 capsules. If you want to scale even beyond that, of course, you can use multiple holders in parallel.

Next question is can you elaborate on the performance in monovalent versus polyvalent buffers?

Absolutely. Let me check. So can I share some more slides? Yes, I can. Sorry. Can I go back to the slides here? Do you see my screen now? Okay. So yes, let me elaborate on that. So what you see now is a different data set, which shows the performance in monovalent and polyvalent buffers in a little more detail than what I showed before. So here, we're showing data at pH 6 in acetate and citrate and pH 7 in tris and phosphate. So we've got data at low and high conductivity. And the Polisher ST was loaded to 10 kilograms per meter squared as before. As you can see, the HCP removal in acetate and tris is very good. It's best at low conductivity, as expected. And then with, for example, phosphate buffer, there is some reduction in performance and low conductivity because the Q chemistry of the first 4 layers would be slightly less effective. But the performance in phosphate doesn't actually go down at higher conductivity. So at high conductivity, there's basically no difference in performance between tris buffer or phosphate buffer. Citrates, as mentioned before, is more difficult. This, you can see again. But again, the conductivity doesn't seem to have a significant impact. So even at 20 milliSiemens per centimeter, we do get HCP clearance in citrates.

The next question is can the product be sterilized or sanitized?

The answer is yes. So all the capsules can be autoclaved prior to use or can be base sanitized with up to 1 molar sodium hydroxide for up to 1 hour.

And can this product be used outside of the shown range?

The answer is yes. So we've shown a range of pH 5 to 7.5, and that is the most likely operating window. The same is true for the conductivity. So as for pH, when you're lower than 5, you'll be getting close to or even below the isoelectric point of many of the host cell proteins. And of course, if the host cell proteins are not negatively charged, then they don't -- they won't bind to the membrane. So that's why lower pH levels are typically not desirable. As for pH levels greater than 7.5, you need to take care about getting closer to the isoelectric point of the target protein. As you'll get closer to the mAb, there is an increasing risk of binding the target mAb, which obviously would result in yield loss.

The next question is when Emphaze AEX Hybrid Purifier is used in clarification, how does that influence the downstream AEX step with Polisher ST?

Right. So indeed, we showed a process scheme which shows the different products being used. I'll see if we can pull this one up. This one should be coming up now. And so the thing is the Polisher ST is designed to work under all the conditions. So whether you use Emphaze AEX Hybrid Purifier upstream or not, you can still utilize Polisher ST to replace the AEX column. Now of course, there will be an effect on the capacity. So the target loading of 10 kilograms per meter squared for the Polisher ST is just a recommended value, which will depend on how clean the product stream is. If there is a high turbidity, lots of DNA and high concentrations of host cell proteins, then capacity will be exhausted more quickly. When the Emphaze AEX Hybrid Purifier is used upstream and the amount of impurities after protein-A column are much lower, then the Polisher ST will obviously have a much easier job. So in those conditions, the capacity may actually exceed the 10 kilograms per square meter loading. And with such low levels of impurities, you may be able to load 12, 15 or more.

Will the use of the 3M Polisher which reduces number of steps also result in the reduction of process time compared to the traditional AEX use?

Potentially, that will depend on the exact cycle that you do on the AEX step currently. So obviously, we do divide its single use, so you would just equilibrate. You would load. You would maybe chase the products, and you're done. So you don't have those regeneration cycles that you would have with a traditional column. The flow rates are also higher. But typically, what you would do is you would have a much smaller device. So for the loading, you go faster, but you also have a smaller device. So let's say, for the loading steps, the time would be approximately the same, and then you would win time for -- because you don't need the regeneration and the cleaning of the column basically.

The next question is will the load turbidity negatively affect the Polisher performance?

Right. Good question. The answer is yes. So it is designed to deal with turbidity. These first 4 layers are based on the same technology as the Emphaze AEX product, and that one's used before the protein-A column where the impurity levels are much higher. So it can handle quite a lot of turbidity, and it can handle quite a lot of cell debris. But of course, once the capacity of those layers are exhausted, then any turbidity or the particles that cause durability would start to hit the membrane, and you would see that as an increasing pressure during the run. So as long as the pressure doesn't exceed the maximum limit, it should be fine. But in extreme turbidities, the capacity may be limited because of that, yes.

The next question is have you proven that the virus removal is not size-based? If it were, this would prohibit you from also claiming the LRV from virus filtration.

Right. Good question. So we've used different virus models, including very small ones, which we don't expect -- which couldn't possibly be retained because of size. The membrane has a pore size of 0.8 micrometers, so the smallest virus would definitely wouldn't be retained based on size. The pores are way too open for that. And we have done experience as well with high salt, where if you hit the membrane with very high salt conditions, you can actually elute off the viruses. So the -- yes. The mechanism of taking out the viruses is definitely charge-based.

The next question is what materials are the new Polisher ST columns made out of?

So the layers themselves, the first 4 layers are based on a polypropylene nonwoven where we attach charged polymers with a Q ligand. Then below, we have the membrane with the guanidiIium ligands. This membrane is a polyimide membrane. And then for the capsule construction, they're basically the same materials that we use for our other products. So we've got some polypropylene internals, some plastic components, and then the capsule shell will actually depend a little bit on the scale. So the smallest ones are polypropylene. The scale of capsules are polysulfone, and the largest ones are a copolymer of polystyrene and polyphenylene oxide.

The next question is will the larger 3M Polisher ST membrane cartridges be amenable to cleaning and sanitation for reuse for the same protein?

No. The Polisher ST is really designed as a single-use device. We do not have data to show cleaning and reuse of the device. And because of the guanidinium ligand, remember that this is a salt-tolerant ligand, so it's not like you can elute it off at higher salt conditions. So that's one of the reasons why it really is a single-use device because probably, even when you clean it, you may not be able to get everything off. And of course, if you don't use it as a single-use product, then you'll lose some of the benefits of having single-use technology, which is the fact that you don't have to clean, you save on buffers and labor cost and so on.

The next question is, is it cost-effective in respect to DE resin?

Is it cost-effective -- sorry, cost -- in respect to what resin?

DE resin?

Okay. So we've done some cost modeling on this. And when we launched product, we'll have some documents around that. So what we see, just in general terms, for cost is that, obviously, comparing to a reusable chromatography resin, the consumable costs will go up. That will be largely compensated for by a reduced labor cost, a reduced cost in capital investments as well as having a higher recovery of demand because the device itself will be a lot smaller than the column. So you'll lose less mAb inside your column or inside the device which will help your product recovery, your yield, and hence, your cost.

And the next question is how will the use of the 3M Polisher compare to traditional methods in terms of cost effectiveness?

So I guess this ties in with the last question. Consumable costs will be slightly higher than a resin that is used for 100 cycles, 200 cycles, but it would be compensated for by labor, reuse, buffer reuse and so on.

And is there an integrity test for this device? And can it be correlated to viral retention?

Yes. Exactly. So we do a pre-use check based on the differential pressure to check if everything looks fine. And then post use, there will be a bubble point test, which can be correlated to viral retention, indeed.

And how do you demonstrate HETP column performance such as symmetry?

So I mean this is not a column, so it's not resin-based. It's a membrane. And so in terms -- when you install it, you would check for the differential pressure, and then you've got that post-use integrity test, which gives you basically the green light that the device performed as it should have.

And the last question we have time for is can we see the virus reduction performance being appropriate for cross-platform applications like plasma fractionation?

Potentially, yes. So the product was mainly developed for monoclonal antibody production processes, as I showed today and most of the virus data, what was generated for that purpose. Now of course, I mean, we saw quite a broad range of different viruses, very wide range of conditions. So indeed, going forward, we do see possibilities for viral clearance in other types of processes as well. Yes, absolutely.

So that's all the time we have for questions today. Thank you, Joris, for a great session. If anyone submitted a question that wasn't addressed, keep in mind that the speaker will reach out to you directly. This session was recorded. You'll receive a notification in 24 hours when the on-demand session is available for viewing. Before you log off, please take a moment to complete the feedback form, so we can continue to improve your digital week experience. And on behalf of Informa Connect and Life Sciences, I hope you have a great day.