Fusion Antibodies plc (FAB.L) Earnings Call Transcript & Summary

Good morning, and welcome to the Fusion Antibodies plc Investor Presentation. [Operator Instructions] The company may not be in a position to answer every question received during the meeting itself, however the company can review all the question submitted today and publish responses where it's appropriate to do so. Before we begin, I would like to submit the following poll. And I would now like to hand you over to CEO, Adrian Kinkaid. Good morning to you.

Thank you, Alex. Good morning, everybody. Thank you for joining us on this IMC. I'm, as Alex was saying, CEO of Fusion Antibodies. Very pleased to be here. I'm joined today by Stephen Smyth, our CFO. So without any delay, let's give you the latest update on Fusion Antibodies results for the year ended 31st March 2025. So as per usual, we have a disclaimer, which I'm sure all of you are practiced with this. We'll have read that in just a few seconds. So we'll move on, and I will hand over to Stephen to take you through.

Thanks, Adrian. Again, by way of introduction, I'm Stephen Smyth. I'm the Chief Financial Officer here at Fusion Antibodies. We'll start with our income statement abstract for the most recently audited year with full comparatives. So revenue increased year-over-year by 73% from GBP 1.136 million to GBP 1.965 million in the year ended 31st of March 2025. The first half of fiscal 2025 was strong at GBP 1.2 million, at which point we were slightly ahead of our internal forecasts. However, H2 was much more challenging with GBP 765,000 of revenue booked in the final six months of the year ended March 31, 2025. The rationale behind the lower-than-expected H2 was a combination of some projects being pushed out into fiscal '26, some being rescoped downward in size and in essence, in fee. And in one instance, we had a material project that was canceled. If we can extrapolate between revenue and gross profit, we can say that cost of sales for fiscal 2025 were GBP 1.535 million, so 78% of revenue versus GBP 1.181 million the previous year, which was 104% of revenue. So while that's an improvement, the company did undertake some very low-margin work in the previous fiscal year 2025 on the expectation that it would lead to follow-up higher margin work, this did not materialize. So I think what that tells me is we have headroom to get better in our cost of sales and pushing that gross profit and ultimately margin higher even based on a similar revenue number, which is obviously not the intention. That resulted in a gross profit, which improved from -- actually a gross loss of GBP 45,000 in the previous fiscal year to a gross profit of GBP 430,000 this year, margin jumping from negative 4% to positive 22%. But as I said, I think there's a lot of headroom in that margin to move that margin north. Admin expenses for fiscal 2025 were GBP 2.1 million versus GBP 2.02 million the previous year, an increase of 4%. The overall wage bill and the R&D spend was reduced year-over-year. However, an increased business development spend and increased rent cost for the company's premises and a higher provision against debtors caused admin expenses to creep up slightly year-over-year, resulting in an FY '25 EBITDA of a loss of GBP 1.674 million, an improvement of GBP 394,000 or 19% on the FY '24 EBITDA loss of GBP 2.068 million. We have moved to the summary of financial position or the balance sheet. Noncurrent assets, there were no additions in the year. So the movement from GBP 158,000 to GBP 63,000 is purely a reflection of the depreciation charge in the 12 months on the existing asset base. Inventory was reduced from GBP 460,000 to GBP 269,000. I think I've spoken about this in previous presentations, but this was the result of a concerted effort to reduce the level of working capital that we have locked up in inventory. So I think we've been somewhat successful at that. Receivables are slightly higher at GBP 632,000 of which GBP 333,000 are trade debtors. That's net of an GBP 88,000 bad debt provision. We have GBP 53,000 of prepayments, GBP 178,000 of accrued income and GBP 68,000 of a VAT receivable. Cash at year-end stood at GBP 359,000, down from GBP 1.2 million at the same point in time in March '24. This is somewhat skewed by both the level of fundraising activity in the prior year fiscal 2024 and the timing of it right at year-end. So that's -- we are -- second fundraise of 2024 right on year-end. The second tranche of that fundraise actually fell into the start of 2025. Current payables also slightly higher at GBP 623,000, of which GBP 372,000 are trade payables, GBP 158,000 are accruals and other payables. We have GBP 38,000 of deferred income. We have GBP 35,000 of social security and PAYE payables, and we have GBP 20,000 current portion of our higher purchase payables. Noncurrent payables of GBP 31,000 is the dilapidation provision that we have to restore the current premises back to its original prefusion state. So at the year-end, total equity stood at GBP 669,000, down from GBP 1.779 million at the previous year-end, again, due largely to 2024 having much higher fundraise activity and the timing right at year-end. Cash flow for the year, cash used in operations, GBP 1.364 million. So that's a reduction of GBP 405,000 or 23% year-over-year. Minimal amount of investing activity, GBP 1,000 of interest income last year, GBP 5,000 of interest expense this year. As you can see from the financing activities line, we had obviously significantly raised a lot more in 2024. And in fact, the GBP 529,000 that's in this year was the second tranche of the final fund raise at last year's year-end. It just fell into 2025. So cash movement in the period was a cash burn of GBP 840,000. And then if we do the math, opened the year with GBP 1.2 million, burn GBP 840 million, gives us our cash balance at year-end of GBP 359,000. Adrian. That concludes the finance slides.

Okay. Thank you, Stephen. So what I'd like to do is move our attention looking forward. And to do that, I'd like to -- if you'll indulge me, just explain a little bit about the background to antibody discovery. So the front end of what we do, we discover, we engineer and then we supply antibodies and the various different engines or discovery platforms that we have for finding those antibodies. For the sake of this, I'm going to talk about human therapeutics, which is still the largest market for us. It's a global market and very much needed. So the top 2 platforms here, the mouse hybridoma and the rabbit B-cell cloning are pretty much traditional methods of identifying antibodies. In the first instance, it's a mouse antibody or a rabbit antibody that you would be presented with. This then needs humanization, which is part of our engineering process in order to make it into an antibody that would be accepted by a patient as being a human antibody, one effectively looking like one of their own and therefore, wouldn't elicit an immune response to the drug. It does add a bit of extra time, but a lot of people really want full-length antibodies to be working with. It's very important that you focus on the end point in what we're doing, and that is full-length antibodies. In this -- in the top 2 platform cases, they're just not human. If you do want to start with human sequences, then that's where the bottom three platforms come in. So phage display is something that's been around for a little while, and it's produced some of the best-selling antibodies, but it certainly hasn't taken over the whole discovery arena. So we're still seeing antibodies at a fairly constant rate in terms of marketed drugs having their roots in full-length antibodies from mouse and rabbit. So phage display does have some advantages. It allows you to get straight to human sequences, but it's got a severe restriction in the size of the protein that you can actually interrogate and look for. It's not a full-length antibody. It's only a fragment of that. So you then have a bit of a LEGO game to build these fragments together to make a full-length antibody. It does, however, have an advantage in that being a library, you can look at things that would be very similar and conserved between species. So again, in the top 2, mouse and rabbit. That's fine if you're looking for, say, infectives viruses, for example, they're alien to the mouse, they're alien to the rabbit, they're alien to humans, and you can generate antibodies against them directly. If on the other hand, you're looking for a therapeutic, which is deliberately going to bind and modify, say, an out-of-control protein, but a human protein, then you're looking for something somewhat different. The mouse and rabbit may well generate an antibody to that, but only if the sequences are different between what you're putting into the mouse or the rabbit. In other words, the human sequence is different to those species. If it's the same, it won't read an antibody to it because it will think it's -- the immune system will be thinking that you're asking it to make something that's autoimmune and going to do detriment. So there's this filter that you have with living species, immune systems that only allow you to get where there's significant differences. Some of the reasons why people look at more and more exotic platforms than mouse and rabbit is in order to elicit an immune response. So some people are using chicken antibodies and even fish sharp antibodies in order to find something which is binds to the human target protein. The further away you go from humans, the more that humanization is, it becomes more and more demanding. So of course, if you can start with human sequences, that makes it easier. If you can start with a library, you don't have that filter of the immune system. So that's beneficial. However, with phage display, including OptiPhage, you still need to put the sequences together and build them up. Another way of doing it is in the bottom triangle here, and that would be to build those sequences in silico and theoretical antibodies can be designed, which could bind to a particular target. But you have to bring those into the real world and actually test them to see whether the theory was correct. And with most AI/ML approaches, that means synthesizing each antibody one by one, which limits the number of sequences that can be brought into the real world. And it could be limited really to tens of antibodies. With our AI/ML-Ab, we can combine it with the mammalian display which is used in OptiMAL, and we can do libraries of many thousands of antibodies. So there's a lot more potential from AI/ML in our hands using mammalian display in conjunction with the in silico approaches. OptiMAL, however, is a fully diverse large-scale library, which can be used for screening pretty much any target and screening against full-length IgG antibodies yearend target, which are already human year-end target. So if you want to start with the end in mind and who doesn't, then OptiMAL is going to be a way to go. It's -- it can take any target. It's not directed on any particular target class or useful for particular things. It's universally applicable. So it's an excellent place to start. And really, it's got an awful lot of advantages not only to our clients, but also to us. Now I've touched on the antibody market and why we're doing this. But essentially, half of the drugs in development going through clinical trials, et cetera, are still antibodies. They come in different formats. They can be human or humanized. They can be engineered into bispecifics so they bind to two particular different targets and bring them together because that antibody then that bispecific antibody then acts as kind of a glue between two targets. They can be adapted into what's known as ADCs or antibody drug conjugates or they can be made radioactive by using radioimmunoconjugates or RICs. But essentially, they're all antibodies. So the market that we are addressing with these discovery platforms is huge. Now as I mentioned, the early approach way to develop an antibody would be to immunize different species, humanize them. And if you've got conserved sequences, I'm afraid between the human and the species that use, I'm afraid you're not likely to find so many hits. There is a risk to the affinity and humanization. You might lose some of the -- how well that antibody, that initial starting antibody binds. It might need to be compromised in some way. And there's also liabilities in terms of manufacturability as you move through that process. At the end of the day, you want a human antibody that can be manufactured by expressing it in a mammalian cell, standard methods for manufacture in the industry. So all those different bits sort of add to the time line. Now our approach, a more modern approach, if you like, is to discover a human antibody. And there are other ways of doing that. You can make transgenic mice with a human immunome built into them. They become very expensive, and you've still got to immunize in the traditional way. For sure, it's not always ethical to use humans to generate antibodies and use them as a discovery platform. So we can put that certainly to one side. So human antibody libraries like OptiMAL are really required to make improvements to the way that therapeutic antibody discovery is done. On minor caveat, we don't anticipate that actual drugs will come straight out of primary screen. We'll still need to engineer them. They might need affinity maturation, for example, making them bind at even lower concentrations and other engineering is possible, such as conjugation for ADCs, et cetera. The benefits of OptiMAL, however, for our clients is very significant. So we've got that speed, which I mentioned, and OptiMAL is really very fast in terms of achieving a fully human IgG. It may take longer to find some binders, but it is a fully human -- a full-length IgG molecule. So it actually works out faster. We're dealing with human antibodies all the way through, so we don't require humanization. We don't have any risks that, that process would carry with it. And they are full-length antibodies. They're not the phage display fragments. So again, we eliminate the risk of putting those fragments together and any errors that can come through that. Now in our OptiMAL platform, our mammalian display platform, it appears that we've developed something which is extremely efficient. So it's enabling the use of much larger libraries, and we're targeting 10 to the 9 antibody sequences in the screen, whereas from what we've heard from competitor platforms, mammalian display platforms, they're orders of magnitude less capable in terms of the total number of sequences that might be used. So they're down at sort of the 1 million target. And that's just not good enough if you're going in for what we would call a naive screen. So OptiMAL is enabling us to do much larger screens and go straight in with the OptiMAL platform against pretty much any target. We don't use animals in OptiMAL processes. It's entirely animal free, and that's an important thing. It's important also for our clients. They want to avoid using animals wherever possible. And indeed, there are such things as EU directives saying that it's a legal requirement to use non-animal-based processes where they are available and can work to the same capability. Developability of an antibody once it's found, can you manufacture it? And antibodies by means of OptiMAL are discovered by means of OptiMAL are highly developable. We know that because they've already been selected being expressed on mammalian cells. So we know that mammalian cells can do it, and that's what's going to be used to manufacture. They're human antibodies as well. So they're much less likely to cause immunogenicity problems, which might otherwise creep through in, say, a humanization process. Sequence liabilities, which we know about, we've got literature examples where people have found liabilities, things that prevent antibodies from working particularly well, being manufactured particularly well. They've already been removed from the OptiMAL library design. And that OptiMAL library design and method is now patented at least in the U.S. So we've removed those issues that can be found. So there's distinct advantages for our clients. One of the most predominant of which is the time line and just these are our estimates. So don't hold me to it. But these are kind of guidelines as it were, which sort of exemplifies OptiMAL being overall a faster process. So around 15 weeks is what we're seeing for an OptiMAL screen. Phage display, the third one down, such as OptiPhage is just a little bit longer than that, but the process is somewhat more complex, has different stages in it. And then you can combine phage display with mammalian display, which is probably about the closest competitor that we really have to OptiMAL, and it takes that a little bit longer. Transgenic mice are human antibodies basically being found by using mice still, and that takes longer still. And then the traditional methods of mouse hybridoma and rabbit B-cell cloning will take you around six months or more to get through. So we're looking really at three OptiMAL screens being done in the ear versus two rabbit or mouse screens being possible in the year if you were to do them end to end. So there's some distinct time savings in the way that we think that we can get this done. So I've announced previously and updated you previously that OptiMAL works, and we're now in a phase of working out how well OptiMAL works. So these are the kind of criteria which I want to be reporting to you on. As I've said, we're looking to screen around 10 to the 9 sequences, fully diverse screening. So that's a sufficiently large number of sequences that is appropriate for a good hit rate from a screen for pretty much any target. The target classes that people are most interesting are proteins and peptides. So we really want to make sure that we can deal with those. And then we want reasonably high affinity antibodies, not necessarily expecting to be drugs, but reasonable high affinity antibodies coming out of that. And the measure there is what's called the KD or dissociation constant. And ideally, we'd be looking for tens of nanomolar, say, 30 nanomolar, 50 nanomolar or something like that coming from the screen and being a good starting point, we might do affinity maturation on those. We want to be able to tune the process so that we can actually get things that are less capable binders as well. Sometimes that fits the target product profile, and we want our platform to be as widely possible -- widely applicable as it possibly can be. So the ideal if we could take that to thousand nanomolar and still be able to extract those antibodies from the screen, if that was what we were looking for. On the other end of the scale, we also want to make sure that it will pull out antibodies, which are particularly tight binders, but that's in terms of platform development is less challenging. We need selectivity between different targets. So we need to be able within the process to do what's known as negative selection. So we want an antibody to bind to this version of the protein, but not that version of the protein. That would be something that we could potentially build in if we can do that, and it makes the platform very much more applicable and salable. We want to -- without -- with all of this, we want to be mindful of time. So getting to the clinic in that 4- to 5-month time line is helpful. So we're not necessarily getting to clinic unless it's something that drops straight out, but it's the overall time line to the clinic that's important. And the faster we can make the discovery process, we can keep that discovery process down to four to five months, then that's really working. Now as you know, we've been working with the NCI doing a validation collaboration on using OptiMAL in their labs, and we want to be able to make sure that they can do it because that shows that our platform is transferable and we can move OptiMAL to other clients. So we're looking to derive antibodies through or for NCI to discover antibodies against three particular cancer targets. The benefits of OptiMAL to the company and therefore, to our shareholders is also very significant. So obviously, revenue. OptiMAL is designed as the premium platform for antibody discovery and applicable to a very wide part of the end of therapeutic antibody discovery market. So we want to be able to generate significant revenues from this at good margin prices. It's proprietary. It's therefore, going to be only us that's able to enable clients to use this approach. Obviously, we can't overprice it and price ourselves out of the market, that will be foolish, but it is a premium system with advantages for our clients. It would be good if we can reach through and being proprietary and a premium process with those advantages, we're in a better place to negotiate milestones and royalties as well as, of course, follow-on work. Milestones and royalties are very difficult to come by. People are -- our clients are very resistant to agreeing to them, and you really need a proprietary position to stand much of a chance of being able to get those. And of course, OptiMAL gives us that. It's a very wide appeal, as I was saying, in terms of the target classes that it can do, and it's, therefore, a very wide appeal to client types. So because it's unique and widely applicable, we believe that it's going to get attention from big pharma and major biotech, which have been rather difficult for Fusion to engage with at meaningful scale. We've engaged with a lot of them -- they know that we do quality work, but we're looking forward to having a USP that will get our levels of engagement with them an awful lot more. And of course, smaller biotech, it's very applicable to their needs as well. Transferring OptiMAL to other laboratories enables us to change the business model so we can license it to those clients. We then cease to be a rate-limiting step in terms of the number of OptiMAL screens that can possibly be done in a period of time. We're enabling our clients to do that in return for a license fee and a license fee in crude terms is the best margin that you can get. Scalable, as our own capacity ceases to be a limiting factor, we can run many more screens than it would otherwise be possible. The library itself is enormous, 10 to 45 in theoretical terms. We can't make all of those antibodies. It's highly diverse. It means we can sample for different clients. So if we're making 10 to the 9 for a screen for company A, then when company B asked for us to make a library for them, we're sampling from a much bigger pot and there's no overlap, which means that we're not having to remove hit sequences to enable to maintain the value for the first client. So depleting the value of what's on offer to the second client, that does not happen. So in fact, [ Dia Meds ] expecting the -- with the platform being used more and more, it will actually increase in value because people know about it a lot more. There's a lot less risk that's involved with it. So the library size is a very significant factor in the value of what it presents. It's compatible and synergistic with our other offerings. I've mentioned things like affinity maturation. Of course, we can do that. We can take those sequences once they're matured, we can put them into our stable cell line generation platform, so we can enable people to go all the way through in the offerings that we've got. It's also synergistic in the sense that if we engage with people to talk about OptiMAL, we also might well or are likely to pick up other projects, even if it's not OptiMAL work from that client base. There is further library potential within the scope of OptiMAL in its entirety, it entire sort of theoretical size, we could focus in on certain features. So it's possible for us to generate many OptiMAL libraries, some of them focused, some of them diverse as in the original formatting or indeed generating AI/ML designs in an OptiMAL way so that we're generating focused AI/ML libraries. The platform, as I said, it's proprietary. It's covered by the U.S. patent, which we recently had granted to us. So it can only be used through Fusion or under license in other laboratories. It's unique. So OptiMAL antibodies cannot be generated using other technologies. So in summary, considering things that Stephen mentioned, including that 73% revenue growth. FY 2025 was a really good year for Fusion. We had significant growth. We did very well also in the diagnostics sector. I haven't talked about that really today. We also had a key account win for a cell line development project towards the end of the process that we offer or processes that we offer, but of pretty high value for those sorts of projects. In particular, we made really significant progress with platform development. I've spoken a lot today about the OptiMAL platform and the work that we've done with the NCI, which has provided us with an awful lot of insight and learning points without having to spend more on R&D. And as Stephen said, we actually cut down on the R&D spend overall. In addition to that, though, we have also moved forward in developing more novel technology or more up-to-date technology in terms of how we can do cell line development. Some of that is coming through the FMI grant, the Future Medicines Institute grant that we had announced. And indeed, phage display and actually running the phage planning is also progressing through collaboration with the FMI and what that enables without us incurring building costs to do it in-house. So we reported on a significant increase in pipeline opportunities, particularly in the second half of FY 2025. We're very mindful that it remains a very tricky market. Certain things are moving through quite slowly, but we know that we're doing the right sort of things in order to identify those leads, business leads in the first instance. So we're continuing to progress our AI/ML design of antibody proteins, in particular, with the project with Oxford University under the auspices of Professor Charlotte Deane, who is a global leading authority on the subject. And we've seen some really interesting results coming from that PhD studentship project that we're running with them. I hope to have more come through from that as well. It's worth pointing out that post [indiscernible], we've now got the FMI grant funding coming into our account. Initially, we were hoping that it would come on tap quite quickly. It didn't. There were a few administrative teething problems with the process, but that now is incoming. We also started a further grant for the wonderfully named Death Receptor 5 or DR5 target, which enables us to do possible asset generation. And again, that DR5 grant was made possible really or we won that large part because we are collaborating with Queen's University through the collaborations within the Future Medicines Institute. Grant of the U.S. patent for OptiMAL library design and method is brilliant, and we are seeking patent protection in other authorities as well, but it's great to have that U.S. patent approved. And I think going for that accelerated fast-track approach getting that patent was a really sensible thing for us to have done and very pleased that it's paid off. We also recently announced significant follow-on projects from the cell line development, that key account win that I mentioned in the top of this slide has already paid off with significant follow-on projects for cell line development in this year and indeed possibly even into next financial year. So overall, FY 2025 was very good for future. And at that point, I'd like to close and we can maybe move on to the questions.

Fantastic, Adrian and Stephen, thank you very much indeed for your presentation. [Operator Instructions] While the company take few moments to read those questions submitted today, I would like to remind you that a recording of this presentation, along with a copy of the slides and the published Q&A can be accessed via your investor dashboard. Adrian and Stephen, as you can see, we have received a number of questions throughout today's presentation. And Adrian, if I may now hand back to you to chair the Q&A and read out the questions where appropriate to do so, and I'll pick up from you both at the end. Thank you.

Thank you very much. And could I say thank you to those people who presubmitted their questions. It gives us a chance to be appear slightly more organized and hopefully give better answers than when they pop up. But also thank you to those people who have provided answers straight away. So I think most of these, Stephen, are for me to answer. I think you've covered your bit very well. So it's mainly for me. Question one is, what's your forecast for the share price over the next 12 months on the back of the financial situation? So we don't make forecasts for the share price. I think if I could do that and do that very accurately, I probably would be on a beach somewhere. But we are making very significant progress, particularly with the OptiMAL platform, which is due to come online within the next 12 months as the question is asking. So I would hope to see further improvements in the share price. OptiMAL should be very significant. It's a very significant platform. It should therefore be very significant for the business. And I think we are heading in the right direction and doing all the right things. Next question was, what is the earliest that we can get an update on the collaboration with the NCI. I do appreciate our shareholders, your patience with this, and we will provide an update just as soon as we're in a position to do so. But more than that, I can't say at this stage. Question four was you said you saw a significant increase in sales opportunity pipeline in H2. Has that trend continued into this year? So I mentioned this in the presentation. That uptick was very encouraging, and it showed that we -- the demand for our services is there. The market is definitely there. People may be being a little bit more cautious about how they move it forward. But that uptick was also attributed or can be attributed to our commercial activities, and it confirms that we are doing the right thing. We have the right approach. It generates leads. So that's really good. But as I said in the presentation, we are mindful of the economic conditions that our clients are facing. Indeed, the geopolitical conditions have been more turbulent than we and our clients would like. So we're certainly not taking anything for granted, especially with regards to the speed of progression or indeed the ultimate outcome of each of those potential sales opportunities. Next question was, can you outline the business models you are considering for OptiMAL, the scale of the contract value versus the current model and what sort of margins you are targeting for this service? What is your sense of the overall commercial value for this platform and what it should do to the company's market cap? That's a lot of questions in one go. So in terms of business models, as I said at the start, we can't overprice OptiMAL and price ourselves out of the marketplace. We also are selling into a very conservative audience. Their scientists, they like changing one thing at a time. It's usually the parameter that they're focused on. So we're going to have to work to get people to switch to OptiMAL. We shouldn't have to work massively hard because there's so many obvious advantages to it, but we have to work to make sure that those are noticed. Initially, I'd expect OptiMAL revenues to come from custom service contracts very much in line with the kind of work that we're doing at the moment and those screens being run in our own labs. We have a lot of confidence in OptiMAL and therefore, anticipate successful outcomes for quite a high proportion of those early screens. We may include incentives to encourage early adoption amongst our client base, and that might mean that we see an initial -- a lower initial fee than we might otherwise charge just to get them on board, balanced by success milestones. So if we agree a target product profile and we meet that target product profile, then we might get a success payment in that -- for that rather than charging them more upfront. Early success would encourage those more active partners or parties such as big pharma and the bigger biotechs, people who run many antibody discovery programs in a year to place orders for multiple screens. So they might want in year 2 of a collaboration, possibly three screens being run rather than one. So that obviously gives us a lot of potential for growth. And I would anticipate that, that would be done on a custom service basis also. But the next phase is the really interesting one where if they want more control over that or want to run more screens, than the budget would necessarily allow and they want to take that in-house, then we can enable that. And that's a real differentiator really for OptiMAL. They can bring it in-house, run it the way that they want to and pay us a license fee for the privilege of doing so. Margins-wise, a license fee is effectively as close to 100% margin as you can get. It's just some administrative charges. It also means that Fusion's capacity to run screens would cease to be a rate-limiting step or rate-limiting factor. We would continue to run them, but we'll be focusing on the next tranche of clients who will, in the fullness of time, some of those may want to license the platform itself. So I can see some very busy times building through for Fusion. And it's a proprietary technology with distinct advantages. So it does fit well with the client offering that time line and risk reduction and so being worthy of the premium positioning. Next question was what feedback have you had from your client on OptiPhage? And when do you plan to formally launch the OptiPhage service? I haven't mentioned the project that we ran with OptiPhage, which was done really quite ahead of time. So that existing OptiPhage library design project, which we ran, it's still with the client for review, which is within the scope of the agreement, how much time we anticipated they would want it for review. And they still have an option for an exclusive license on that particular library. That project was a little unusual in what we anticipated for OptiPhage in that it was just a library design project with the client running the actual screen themselves. They were running the actual lab work. So the formal launch of OptiPhage is really linked to us being able to offer phage planning as part of the process. And as I mentioned earlier, that's being enabled by doing it through the Future Medicines Institute. So as that comes on board, they're not the only people that we could send it to, but it would be our preferred way of running it just down the road. We've got the collaboration there, but it would be a great way to run the process. Once that's on board, then we can look to a launch process for OptiPhage to complement OptiMAL, et cetera. Next question was Fusion has referred in the past to agreements that could generate milestones or royalties and OptiMAL should increase this. Can you give us a sense of what the portfolio of future potential entitlement looks like today, recognizing that it's uncertain if they will come off? So we attempt to add to the portfolio projects with potential milestones and/or royalty elements. They are a long way downstream. There's an awful lot of work that needs to be done. And our clients are really undertaking some of the most difficult things in my view, that mankind has ever undertaken, fighting disease by harnessing biology. So it's very difficult also to achieve an agreement on a milestone and royalty because they are, by nature, fairly lumpy things. And it's particularly difficult if you don't have a proprietary position backed by a patent that shows that it's -- they can only come to you for it. Now OptiMAL does change that. And it is -- as it is patented in the U.S., and we are hoping that it will be patented, the patent will be granted in other jurisdictions as well. So that's where OptiMAL will improve the situation. It gives us a better chance of doing that really difficult thing of securing agreements with milestone and royalty. Those that we've already got in the portfolio that we have some agreement on already, we'll continue to monitor those projects, which may well lead to milestones and/or royalty payments, bearing in mind the length of time and the difficulty that is required to make that happen. We anticipate that some of these projects will terminate sadly before reaching those milestones, but we hope that some will make it through. Indeed, we have seen and reported on milestone payments in the past in the last few years, and I would hope to do so again in the future. Next question was in an answer to a question on the investor hub, you said several partners have expressed an interest in OptiMAL and you're progressing discussions. Do you expect to have any partners on board when it comes to OptiMAL to launch OptiMAL as a service? So I don't actually expect to have customers for a service before it has been launched. That said, we did see an opportunity to achieve exactly that with OptiPhage, which I mentioned a few minutes ago. Nevertheless, OptiMAL is very newsworthy and potentially impactful platform. So we find that people are very willing to engage with us to discuss things around that. That's very helpful, encouraging and positions us well to have traction when it is launched. And that includes the associated technologies as well as using OptiMAL with its diverse naive library, as we call it that 10 to the 9 library selected from 10 to the 45. What we also find is traction or interest in just mammalian display. Some clients may have a library or a library design or a library project themselves where they really need the mammalian display, and we will engage with those people as well. It positions the company to gain traction with OptiMAL with mammalian display, but also increase awareness and traction of our other services, too. So we're exploiting that in every way we can going forward. Next question is, when do you expect to hear from the NCI on progress of their validation work? We're in regular contact with the NCI, and we collaborate with them to best progress the OptiMAL antibodies as well as the platform. They will want to ensure that they -- that any viable hit antibodies that have been discovered are screened for what we call functional effects and not just that they bind to the target, which is kind of our responsibility, but that binding causes, in their case, anticancer effects. Now potentially, that could lead to patents. And therefore, before they publish any information specifically about those antibodies, they will want to make sure that they've got their patent applications tied up as well if indeed it goes that route. And we very much respect that and consider it the right approach. So it will take time. It is difficult to predict how long it will take. But to some extent, the higher the value of the antibody, the longer the time it will take before we hear about it. The next question was, can you tell us some more about the specific fusion projects that you're working on as part of the FMI, the Future Medicines Institute? Do you anticipate deriving new business directly from membership of that group? Or is it a benefit purely R&D at reduced direct costs? Well, I've mentioned several times now the OptiPhage platform. And I've spoken about the grant-funded DR5 project, which is being run in collaboration with QUB. Both of those OptiPhage and the DR5 are only really possible in the way that we're wanting to do them because of the collaboration with FMI. We also have two PhD studentships on the R&D side of things. So they're due to start soon. I'm very hopeful that they will develop some very skilled and talented scientists as well as getting some research work done, which will benefit Fusion. And they are fully funded by the FMI. So effectively, that couldn't come at a lower price point, I suppose. Some of our existing or pre-existing clients are also within the FMI partnership group as well. So we anticipate further commercial traction from those. The FMI is also looking to attract additional partners and new people to join the consortium, and they could well be clients. So I can see benefits from the FMI, not just -- definitely not just R&D, but commercial traction as well, be that internal projects such as DR5 or revenue-generating projects. It's all -- it's very good, whichever way you look at it. Next question is, in your annual report, you said the cash raise in March and the FMI grant positions the company well for the final financial year. Do you still think that you don't need to raise funds before breakeven? So we are working to a plan which would enable us to reach cash flow breakeven. We're still on course with that plan. And the auditors have in the annual report, confirmed their view that this plan is viable, albeit dependent, of course, on revenues being in line with our expectations. And as I've said, the market remains challenging, as everybody knows. The FMI grant does act as something of a buffer in this regard as staff not deployed on revenue-generating projects can be deployed on internal platform and process development work. and the associated costs claimed through the grant. So effectively, we have a very flexible system, extremely well devised for partners such as Fusion. Last question of the pre-submitted ones is, can you explain the decision to no longer remunerate Board partially in equity? So the Board have been extremely supportive, very supportive of the company. This included all of us making some compromises to reduce the cash burn of the business during what was a very difficult period. We're through that period, and we have growth again. So those special circumstances no longer apply. I might also point out that some of the executive remuneration, specifically bonuses is being made in shares. I am very happy to accept them instead of cash even at the higher share price. And so I'm getting fewer shares than would have been the case when the criteria for the bonuses were met. And I am hopeful that the share price will continue to improve and reward us and all of our shareholders for that faith in the business. So I hope that answers that question, too. Stephen, have you had an opportunity to look at some of the freshly submitted.

I have, Adrian, there's -- I think again, they're probably mostly for you. I don't know if we want to take them in order. I know we're running tight on time. There is one on the cash, which I'm happy to deal with. Was the cash figure at the end of 2025, slightly low due to the second part of the placing and any FMI grant monies? It's really -- as I said, it's really a timing issue. If we look at what happened and when it happened, at the end of last year, we -- right on year-end, we had a raise, which substantially closed before March 31. So we had a cash raise right at the end of the year to push the year-end cash balance. The second tranche of that raise came in, in April 2025. So it's effectively been 11.5 months since we've had a cash raise in 2025 versus right on year-end at 2024. And then as Adrian alluded to on the FMI grant, while we would have liked to have the first payment in before year-end and reflected in the year-end cash balance, I think it was actually the 2nd of May of fiscal 2026 that we received the first payment on the FMI grant. So again, those two things are the reasons why the large delta in cash year-over-year.

I've had a chance to scan through some of these questions. One of them that leaps out to me because I can speak to it with some of my past experience was please explain the comparison between antibodies and aptamers and the pros and cons of each. So it's a bit of a scientific one. Antibodies are what we do. Aptamers are short strands of DNA, which can, in certain circumstances, be used in a similar way to antibodies. The question says, are they competitive processes and are aptamers lower cost to create? So I think the first thing in terms of are they competitors? Well, to my knowledge, there has been one aptamer drug approved and on the market. I think there may be a second one on the way, compare that to the number of antibody drugs that are out there in their various forms. And no, there's just no comparison. They are interesting molecules. The -- a lot of the advantages that they have are for the people discovering, engineering them, you can amplify up strands of DNA, whereas you can't amplify up an antibody. So they're easier to work with in some regards. But once the sequence is known and they're extremely easy to sequence, how do you protect them? You can apply for patents, yes. But really, they are extremely easy to copy. So I'm less enthusiastic about them for those reasons. I think they have a place. They're interesting binders. They have interesting properties. You can make them bad binders quite easily, which means that they could release the target, and that's quite interesting, can be applicable in certain circumstances. But really, if you compare them to antibodies, it's difficult to compare them. Antibodies are the benchmark the gold standards, the vast majority of the marketplace, the better applications, the more solidly founded. So I'm going to remain focused on antibodies, if you don't mind. Just one more, I think that we can squeeze in, and I know we're already a little over time. So one cuts right to the point at the end here. Are you now generating sufficient cash to be cash positive? Not yet, but it's a question of time, and we're doing all the right things to get us to that position. And we have some great assists in doing that, particularly with the FMI grant. I think that's one of the key things. But as it stands, as we've reported, we're not yet cash positive. It's just a question of time. So I think we best to draw a halt to the Q&A there, Alex, if that's okay.

Yes, that's great, Adrian, Stephen, you have covered a lot there today. And of course, the company can review all those questions submitted and we'll publish those responses on the Investor Meet Company platform. But. Adrian, before I redirect investors to provide you with their feedback, which I know is particularly important to the company, could I just please ask you for a few closing comments?

Yes. So as I said, FY 2025 has been a remarkably good year for the company. We want it to be better still. We are working towards being to cash flow breakeven. not quite there yet. But as I say, it's a question of time, and we have all of the components required, I believe, to get us there. Most notably, of course, is the OptiMAL platform. But at the same time, I really shouldn't allow that to overshadow the other things that we do, which are highly relevant to the client base. And indeed, some of those, one of our clients recently described us as being best in the world at doing. So it's definitely not a show with the other stuff that we do. But quite rightly, OptiMAL is new and gathering us an awful lot of attention, which we intend to exploit for the benefit of our shareholders. Thank you very much, everyone, for your attention.

That's great. Adrian and Stephen, thank you once again for updating investors today. Could I please ask investors not to close this session as you will now be automatically redirected to provide your feedback in order that the Board can better understand your views and expectations. This will only take a few moments to complete, and I'm sure will be greatly valued by the company. On behalf of the management team of Fusion Antibodies plc, we would like to thank you for attending today's presentation, and good afternoon to you all.