All right. Good afternoon, everybody. Welcome to day three of the Stephens Conference in our last session here, but it's a good one. I'm excited for the conversation that's coming. I'm Jacob Johnson, the Life Science Tools and Services Analyst here at Stephens, pleased to be joined by Chairman, President, and CEO Dr. Stephen Dilly from Codexis.
Before we get started, it'll be a fireside chat. We'll do Q&A. I'll try to pause along the way if anybody has any questions. And with that out of the way, Stephen said no need for prepared comments, so we'll go right into it. So I think what would be good, Stephen, is to kind of talk through, we'll spend a lot of time on ECO Synthesis, then we'll talk on the rest of the business, but obviously that's a sizable opportunity, and you guys are awfully excited about it, and we'll talk about why.
But maybe I kind of want to think about it high level first. So the ECO Synthesis and the siRNA market, you've got a long history in the biotech industry. Why should we be paying more attention to this modality, and what does that pipeline look like today?
So as you say, I've been in this industry an embarrassingly long time, and I've seen a lot of waves come and go, and some of them really stick. And one of the most impressive things that I saw was the implementation of monoclonal antibody technology. And what you saw there was technical success, some things being worked out, which enabled a whole field to come. And they had advantages over the existing therapeutics. And we're seeing the same thing with siRNA now, where after really 20 years of hard work, we've now really addressed the challenges of targeting the molecules to the appropriate place, getting the durability of action, and getting clinical data to really show what they can do. And now there are six products approved.
There are 400 more siRNA products in development at some stage, about 50 of those in later stage clinical trials, addressing the obvious indications because the first targeting was to the liver, so things that you can do through the liver like lipid lowering and addressing hepatic diseases, but also now success in targeting places like the CNS and skeletal muscle, which opens up whole new avenues. So we're seeing this as very much like that monoclonal antibody revolution and driven by just what beautiful medicines these are, moving from needing daily therapy to sometimes two injections a year to control things like really difficult blood lipid cases and stuff like that.
Got it. So that's kind of the pond or maybe ocean you're fishing in. So from there, what is ECO Synthesis, and where does the double-stranded RNA ligase fit into this?
Right, so let's talk about the pond just for a second and then talk about ECO Synthesis, which is that at the moment, with the coming wave of siRNA medications, the global demand for these drugs in kilo quantities, just look at that sort of lumped number, is probably 700 or 800 kilos a year if you look across all the entities, all the territories. We see that growing by the end of this decade or early 2030s to about 30 metric tons, so 30,000 kilos. Currently, these drugs are produced chemically using a technique called phosphoramidite chemistry, which is an old technique that's been around for about 40 years. It's based on organic solvents such as acetonitrile and toluene, and it's very resource-intensive and very economically and environmentally toxic.
What we've done is looked at that and said, "It's going to be really hard for the chemical players to scale to meet the coming demand. Can we do this better?" What we've done is built a process that can build siRNA drugs using enzymes in aqueous media. So we take away the problems of the solvent and the solvent disposal. We think that using enzymes over time, we can actually demonstrate that we can produce siRNA that is purer, it's more scalable, it will be cost competitive, but really it'll be about competing based on quality and availability.
Maybe just one follow-up to that because we'll talk about the number, the finance implications for your customers and that value proposition. But the environmental piece, can you talk about just how the chemical process works from an environmental standpoint, especially when we think about, I'm guessing some of this is done outside of the U.S. because we don't want it in the U.S., so.
The process is onerous. One of the ways chemists and people looking at green chemistry measure this is the number of kg of components or inputs you need to get a kilogram of output. With phosphoramidite chemistry, it's about 7,000 to 1. That's 7,000 kilos of raw material to get. You can see that the overall process yield is pretty low. It's also that the organic solvent sounds nice until you realize that means acetonitrile, also known as methyl cyanide, which is a byproduct of the petrochemical industry. The way that is disposed of is burning it off. To make a kilo of siRNA by the conventional technique, you get about four and a half metric tons of atmospheric carbon. That really matters when you start scaling to metric tons a drug, multiply that by 1,000. That's a huge impact there.
Also, the cost of disposal of the solvents in particular is really problematic. Now, there's also where the raw materials come from. The building blocks to do the chemistry or to do the enzymatic process also require significant, usually about a 30-step chemical synthesis to make the building blocks. And we built a process to do that enzymatically as well. So we're looking at the whole kind of ecosystem and affecting it.
Got it. So then as we think about this ECO Synthesis platform, I guess the way I think about it is in the near term, you have this double-stranded RNA opportunity, and then that should be—you can parlay that into the full-blown ECO Synthesis platform. Can you just talk about compare and contrast those two elements of this?
Yeah. So I can talk till I'm blue in the face about exciting economic opportunities for Codexis in 2029, and people will say, "Come back in four years," right? So we need to have an addressable market that will give us traction in the near term. And what we've learned is that using a single enzyme or enzyme family called the ligases, which stitch together short strands of RNA to make the complete drug, has applicability right now. Because even with the chemical process, if you make shorter pieces and stitch them together, you can improve your yield. Because chemistry is harder the longer you go with the strand that you're building, and the impurities mount up, and the costs mount up.
And this allows us right now to go into partners that are doing chemical manufacturing and say, "We can improve your purity, we can improve your yield, we can improve your cost structure by putting a single enzymatic step into your synthesis." Now, that's great because they do that, they adopt it, they kind of get enzyme religion and say, "Okay, can we do more of this?" We did a really cool thing last week, which was we showed the leading siRNA drug that's inclisiran, also known as Leqvio, that's produced by Novartis. We showed four different ways of making that drug. We made it completely enzymatically end-to-end. We made it in pieces enzymatically and stitched them together. We made it in pieces chemically, stitched those together. And then we did a hybrid where some were chemical and some were enzymatic, stitched them together.
The punchline, the mic drop moment was we got the same answer each time, so we're making the same drug. That's important because it allows us to take a chemical process at the moment and go stepwise to an enzymatic process. We're never asking people to make a leap of faith and make a huge step and put their supply chain at risk. We see the double-stranded RNA ligase as an important short-term opportunity and a bridge to the overall profitability of the company.
Yeah, it's kind of the first big building block of your platform. So maybe on that double-stranded RNA ligase, I think you've already had some initial success here. I think you had a customer adopt it in between phase II, phase III, if I'm not mistaken. Can you talk about that kind of effort, how that went? And then could you get installed, could that get installed into an existing commercial application?
So just a piece of background, why we're so excited about the ligase: Codexis' heritage has been in the biocatalysis field to produce small molecule intermediates where we will make unpurified enzymes, and we will charge maybe $3,000 or $4,000 a kilo for the enzyme. With the double-stranded ligase, we're talking about thousands of dollars per gram. So it's always a magnitude higher price point, but it's also got to be purified. It's got to be exquisitely optimized to the reaction we're trying to do. So it's really about using everything we've learned to optimize our game. Now, where we are is our first company, a big pharma company that we learned in retrospect had already been using a generic ligase, one that you can find in nature, tested our ligase family, found it was sufficiently better to make a switch from phase II - phase III.
So it wasn't that they hadn't been doing ligation before. It was that our ligase was better. We thought, "Okay, we're onto something here." They did a R&D order and said we liked it. Then they did an order to allow them to scale up, and we're fully expecting them to follow up with significantly bigger orders to supply commercial launch and then full commercial. We also signed up another partner, similarly in late-phase clinical development, to do the same thing and potentially adopt across a platform. So we now have two about to be three customers who've signed on with specific ligases for specific molecules.
Then we have a group of potential customers that are doing what we currently call our screening and optimization service, where you can bring us your molecule, and you can either bring us the chunks that you want to stitch together, and we can perfect a ligase specifically for your chunks, or you can bring us the whole molecule, and we'll go a step further of advising where to make the cuts and how to make it in pieces and optimize around that. We're expecting several of those to convert into real contracts over the next year or so. The difference is with one of our heritage enzymes, we'd be looking at annual revenues in the $5 million a year kind of range, be doing well. This is probably more in the 10, 20, 30 kind of range for an individual customer with a ligase.
So margins are really good. So it's a real driver of converting the company's profitability.
That 10, 20, 30, would that be full-blown commercial? Or any way to think about what it would look like for the double-stranded RNA ligase, what that looks like clinical versus commercial?
It depends on the structure of the deal as we move forward, right? And so there are different flavors for different customers. We've been learning. It's also whether we are simply providing them with the ligase to put into their supply or whether we're actually doing the ligation for them, right? It's that kind of thing. And so all the time with ligation, but also with the ECO platform, we're looking to move up the value chain. And one way to look at this is particularly in the early stage development, particularly with small companies, they don't just want the white powder. They want the analytical framework. They want the regulatory support. And so there's a much more full-service model that could be very attractive in terms of the downstream economics that could command.
Got it. And then, well, and just one other follow-up, just for clarity. For the double-stranded RNA ligase, expect revenues this year and more so probably next year.
More next year.
We'll stay tuned on 2026.
And what we're saying, because it's true, is by the end of 2026, Codexis turns profitable based on existing products with existing customers with expected future orders, right? So we don't have to invent anything new to get there. We just have to execute flawlessly.
So we've talked about that kind of first step, we'll say, in ECO Synthesis. So then if we think about ECO Synthesis, kind of the platform, which has a number of components to it, can you talk about what that ramp looks like? I think launching in 2026, maybe start there, kind of the next steps of that platform coming online.
Right. And launching in 2026 is a really interesting term that we use because it's a bit like SpaceX. There's a bit of work that goes on before you launch and there's lots of testing and all the rest of it, and so 2024 was really about showing that it worked. And so there was, as I say, the big sort of coming out party last week at the Tides Europe meeting showing full-length synthesis and all the other ways of doing it, plus showing the performance of our ligase and all the rest of it. In the background, we've now got two partnerships where we're moving from having proof of principle with the big partner to actually what we're calling a development partnership, and that's where we're getting ready for prime time, so we're looking at the challenges of scaling and how we address those.
We're looking at specific molecules and developing this process because ECO Synthesis generically is universally applicable. We've built a polymerase enzyme that we believe will build any existing siRNA if you introduce the right building blocks in the right sequence. But you can optimize for specific molecules by tweaking the process conditions and all that kind of stuff. So there is a process of optimization that we're going to spend 2025 doing with a couple of big partners across multiple products. So when we launch in 2026, it'll move from being they're working with us on the platform to they're working with us on these specific products.
Yeah. So then you're going to launch it in 2026, then what, right? Presumably, you have customers in the clinic. When could we look at a commercial customer from ECO if all goes well?
So it depends how far you sort of throw the rug over this thing called ECO. If we're talking about ligase as part of the ecosystem, then we expect drugs made with Codexis ligase to be in people next year, right? That's absolutely true. We expect some of those to be commercialized potentially as soon as 2025, but certainly in 2026 as they move forward. It's that kind of time frame. In terms of the overall ECO platform, it will take a little bit longer to get to commercial products. And that's why it's so important that the base business is healthy and growing and builds us the runway and the oxygen to actually get there.
And then maybe just you threw out the numbers for the ligase, but for the sake of people listening, can you talk about? I think the numbers get larger on the full-blown ECO Synthesis side of things.
Yes, and again, it depends on the nature of the relationship we've got, but we think that there are ways that we can get to very big involvement with individual products. Now, some of that will take time, but we're talking about $50 million and up per product annually, so if we can capture even a relatively narrow slice of this $30 billion or $50 billion dollar TAM, we're doing really well.
Yeah. So you've alluded to it a few times, but it's been a busy couple of weeks for your team, including Tides Europe last week. Can you just talk about, highlight any other kind of big highlights from that that you want to call out? And then also just kind of the reception and follow-through this week from those presentations last week?
So the first thing would be a kind of generic observation that we go to this Tides meeting every year. And last year, biocatalysis, enzymatic synthesis of RNA was kind of interesting to people, but it was fairly low-key. Whereas this year, there were a much higher level of interest, awareness, engagement in the field in general. But really, I think we surprised people with just how far we've come. And some of the feedback we had after our presentations was, "Oh my goodness, we thought this was five years off, but this is kind of now," right? And so an acceleration in the adoption. We showed the potential of the platform to make complete molecules, including for the first time ever, complete double-stranded molecule with the targeting moiety put on. We showed the sort of broad capability. I think that impressed a lot of people.
I think established that Codexis is clearly significantly ahead in this field. But then we did some stuff that was really grounded in today, which was showcasing our ligation platform and showing how the Codexis ligase is actually performing better than the available ligases across a variety of different real customer case studies. And so the ECO platform, the broad platform is about promise and tomorrow. And we were able to say, "We're taking orders now on the ligation platform." And that's kind of cool.
Yeah. And then, just you mentioned a couple of customers on the ligase side of things. It sounds like you're working on a couple of development contracts on the broader platform as well. Can you just talk about, I assume the first customers are the hardest to win and the second is probably the second hardest, but I guess, but you haven't named any names. As you start winning these, does that start generating further interest and kind of a snowball effect, or how does that work?
Yeah, it does, and where we are now is we're having to prioritize ruthlessly, so our problem is not sourcing customers. It is prioritizing which ones we work with, so we've talked about our ECO Innovation Lab, which comes on stream in the new year, and that allows us to produce about 100 grams of siRNA and can be run to GLP standards, so that enables people to take molecules forward into preclinical assessment and toxicology studies and all the rest of it, and with these two development partnerships that I'm talking about, we'll be at capacity, right, so we'll be topped out.
So we've had to be really sort of selective and said, "Well, what we're looking for is partners, not the ones that are going to pay the most, but the ones that are most committed to the field, have deep pipelines where the technology is applicable across an array of different assets," right? And so we play the long game because as a relatively small company, every partnership needs to be looked after and takes servicing and backwards and forwards. And you learn the different vibes of the different partners. We've already learned between those two. When they're setting out our objectives for 2025, one of them is interested in scale and saying, "Show us you can make 100 grams." And the other one is interested in reproducibility and saying, "Show us you can do the same 10 x and come up with the same answer," right?
So we can't do 100 partnerships, so we're sticking to two in the short term and then growing from there in terms of the full ECO platform.
You mentioned trying to partner people with people who have deep pipelines, which I think is kind of interesting. So when we think about these development, maybe the couple of potential development contracts, we're not talking about two customers, two drugs, presumably. They got something beyond that. And I'm guessing it's just you can confirm it or if you want to elaborate. It's probably easier to work with two customers with X number of drugs than X than 10 customers with one. Is that fair?
That's exactly right. And it's also that not all of those assets are going to win, right? And so you need to be able to account for attrition and they have big enough commitments, big enough pipelines that they're not going to sort of go cold on the technology because they have sort of the setbacks that are undoubtedly going to happen in their pipeline moving forward. Now, the other thing that we've got to work on really fast is the path to GMP. And there's two ways of doing that. One of them is that you can build it yourself and build ahead of demand. The other one is leveraging what we call a scaling partner. And we're in very advanced conversations right now with GMP scaling partners.
This would be companies with existing facilities where we can drop in our technology and get to the kind of 50-kilo scale very rapidly. That gets us there in sort of 2026, 2027, rather than taking three or four years doing it ourselves. That's an important stepping stone for us.
So we've talked about kind of business development and you've hinted at kind of capacity. So maybe it'd be helpful to kind of flesh out. I think ECO Lab comes on at the end of this year. Can you talk about that process? I guess one follow-up there too. If you're going to be at capacity, why wouldn't you add more capacity? And maybe we'll dovetail that into there is a relationship you have with Bachem. How do they play into all of that?
Yes. So absolutely looking to squeeze in as much capacity as we can, but we also don't want to create confusion by working with too many partners and being all over the place. We think Bachem is absolutely one of the potential candidates to be a GMP scaling partner. And some of that is because they have shown a commitment to innovation even within the chemistry field in terms of what they've been doing with solvent recycling and so on. And that's really important to us. And also, what we come across is the sort of fallacy or whatever it is of sunk cost. There are organizations out there that have literally spent hundreds and hundreds of millions, if not billions, on building the infrastructure to do this chemically. They are likely for all kinds of reasons to be relatively slow adopters of a new disruptive technology.
Now, not completely because we believe that even there the opportunity is present for the ligase to improve their yields, but in terms of building the whole ECO platform, less likely to do that in the short term. So we're looking for the people on the fringes, the smaller players that have the wherewithal to become big players and want to become big players very rapidly.
Then maybe some of this is a bit of a longer-term conversation, but the opportunity is large, obviously. So we talked about the ECO Lab kind of in the near term, and you talked about partners longer term. But if we look out five to seven years and this is going well, what does Codexis need more than what you'll have? Or how do you vision kind of your role in that piece of it?
So one of the things that I say to my team quite a lot is that the world looks simple in seven years' time. It's how we get there, right? Is the question. And we fully intend by that stage to be producing hundreds, if not thousands of kilos of siRNA ourselves and being a direct partner. We also intend to be a full-service provider to companies through their development stages. And as I say, by that, we mean the drug, we mean the analytical framework, we mean the regulatory support. So we need to build out some of that capacity and capability right now. We will be getting into scaling partnerships with current GMP players, but in parallel to that, we fully intend when we're secure in the demand curve and all the rest of it to build a right-scale GMP facility ourselves.
Now, the first step in that is likely to be what we call a kilo facility, which would be for phase I GMP. Because think about that. If an innovator has got an early- phase molecule they want to bring to us, we make them tens or hundreds of grams in the innovation lab and extensions of that. And then we could go into a kilo GMP facility, which would take them through phase I. And then the tech transfer of that to when it's already running at GMP is just that much easier. And we can control it for longer.
You capture a greater piece of the economics.
Absolutely. Yep.
Yep. Okay. So we've talked about some of the building blocks. The ones we haven't talked about are the ones I think you most recently kind of shined some light on, which is the NQPs and NTPs and that offering. I think on the latter there, there's an opportunity for that in the mRNA market. So could you just talk about the timing and size of that opportunity as we think about kind of finding additional ways to generate revenue from all of these efforts?
Right. And so one of my mantras with Codexis is we are moving from doing really cool science to building commercial products, right? And so we have enzymatic cascades now that can go from very simple raw materials to build NTPs and NQPs. And just to remind everyone, the NTP is the building block for mRNA. And then it's an additional step to put a phosphate group on there and make the NQP, which is the siRNA building block. So what we're doing there is working with partners that are already in the mRNA field around scaling that enzymatic process, but scaling it in the U.S. or Canada, yeah, North America kind of thing. Because we're also very aware of issues of supply chain security, Biosecure Act, deglobalization, all of that kind of stuff.
So the current process, or currently people get their building blocks from China that are made chemically using harsh processes very similar to the oligonucleotide manufacturer. We can again do this enzymatically in water. And the infrastructure you need to do that is way simpler and way less difficult to get permits for in the U.S.
So we walked through a lot of efforts on ECO Synthesis, and you've made a lot of progress, including the latest update from Tides last week. But inevitably, in a what have you done for me lately world, it's what's next. So what are the next kind of key proof points you'd point people to as we look ahead?
So let's look at the whole business first. And one of them is we've got this heritage business in biocatalysis for small molecule APIs, and we've returned that to growth. And I'm very proud of that because it was something that had been sort of unloved for a while, but actually it's a very nice, modestly growing business. And by modestly growing, I mean it's going to be growing double digits this year. And then we think low teens CAGR through the rest of the decade. It's about a high 30s top line at the moment. But if you do the sort of compound interest calculation, it could get quite interesting, particularly as margins are improving as we're introducing newer products.
We're also spreading from a heavy reliance on three enzymes with four customers to probably by the end of next year, more like seven enzymes with eight customers, which takes some of the lumpiness out of the business. So we're making some investment in infrastructure there so we can hold on to assets for slightly longer and we can scale them more effectively in-house before we tech transfer those. So look at news flow from that. There's also then the ligase, follow-on orders from the first customer, significant orders from the second customer, and so on. So really incremental, they're cementing that path to profitability. And then it's when we can start talking publicly about who we're working with in these development partnerships and showing success there.
Okay. Maybe I'll pause there and see if there's any questions. All right. You kind of pivoted to pharma manufacturing for me and talked about the growth outlook. But maybe just talk about the capital allocation to that business and how you think about it. Obviously, I'm sure a lot of internal focus on ECO Synthesis and that platform. But to your point, you've got a pharma manufacturing business, which makes you some money. And I do think you're starting to maybe reallocate some or give it some more resources. So maybe talk about that.
Yeah. So some of this is about assessing the trajectory and how much we can impact that trajectory by careful investment. And the first thing that we noticed was that we've been quite successful in seeding the field with biocatalysis, including licensing our technology to some of the big players. And what surprised us was those big players who've been running our CodeEvolver technology for quite a while still come to us with challenges, but they tend to be the difficult ones. Because this is our lifeblood. We keep perfecting our ability to do directed evolution, the applicability of AI machine learning to that, getting better and better. But yeah, we're differentiated when people are trying to do really hard reactions. So yes, there is a demand from those big players.
But there's also a group of organizations that would like to do biocatalysis but don't have the capability right now. And that's opened up a field where we've said, well, you could either spend a humongous amount of time and effort in trying to bring our technology in-house, or we will do your R&D for you, right? And so we're looking at some different evolved business models there with that sort of middle tier of pharma companies where we can optimize enzymes for them, help them get biocatalysis religion, move that forward. And so we actually see that as a potentially quite significant market going forward. The other place where we're investing for pharma manufacturing, but also for the new parts of the business, is pure and simple sales, right? Because we have been a sciency company. We've been very technical.
We need to get some people in that are (and we are doing this right now) that are super excited about commercialization and selling stuff. That's part of a culture change within the organization that's actually quite pleasing to see happen.
Got it. Maybe just as we think about that double-digit growth, a couple of questions. I think you mentioned three enzymes with four customers, seven enzymes with eight customers or something. Are there some additional commercial approvals? Maybe talk about the pipeline for additional approvals on the pharma manufacturing side.
So we're looking at about 15 molecules currently in late phase II or in phase III right now that are using our enzymes. And so what we do as experienced drug developers is we put a probability of success. And now, when we're looking at the commercial potential of one of our enzymes, it's different from looking at the commercial potential of a drug because quite a few very lucrative drugs are for often indications where the price point is high. Well, that's a huge price per milligram. That does not help us, right? What we want is the opposite type of drug. We want drugs where they need tons of the stuff so they use a lot of enzymes.
So we have to look through that filter and say, which of the drugs in development using biocatalysis are pointed at big indications where they're going to need massive quantities of drug? We pass it through that. We then look at their chance of technical success, their chance of commercial success. And when we handicap that pipeline, we say over the next two to three years, this is the amount of success that we expect to see at a conservative level. We could actually do better than that. But yeah, it's a relatively sort of sophisticated calculation, but it's different from assessing which ones are likely to be commercially successful.
Gotcha. And then you mentioned maybe kind of getting better margins on some of this business. How much of that is just getting better terms? And how much of it is doing more complex work that's kind of higher value add for customers?
So you're always on a better stance if you're doing more complex things where you really way ahead of the competition and there's little alternative, right? So this is part of really understanding the pain points of our customers. And one of the things that we've been a little bit guilty of over the years is doing cool stuff, going ta-da, and then expecting everyone to come in. That doesn't work. You have to be in there in conversations with them. What's their problem they're trying to solve right now? Is it around an enzyme class? Is it about a particular step? What is it about that step that they don't like? And so really understanding in depth the challenges that we're trying to solve as opposed to just, as I say, doing cool stuff for the sake of it.
Gotcha. Maybe pivoting a bit, but I think the other piece of all of, I guess the way I look at it is you have ECO Synthesis, you have the sustainable manufacturing business, and then you have a number of assets that you've partnered off in the last couple of years. So I guess maybe the first question, do you feel like you've partnered off most of kind of the big things? I'm sure there's always stuff for maybe Kevin to be doing, but has he largely completed that? And then I'll have a follow-up.
So in terms of the heritage stuff, I think we're largely done in terms of partnering that off. But then there's the new stuff where we have to look at the field and what can we bite off and what do we need a partner to really scale and address the market? And looking forward, there are some super exciting opportunities. I mean, we've already done this with mRNA, where we built a beautiful enzyme in terms of the HiCap and then successes to that. And we said, we don't have the infrastructure to move fast enough to capture that market. So we'll partner it. Hence the deal we did with Aldevron, which is a really good relationship for us moving forward. When we look at the applicability of ECO and its close cousins, well, there's siRNA, but there's also the field of antisense oligos.
There's also single-guide RNA, where some of our technology around sequential synthesis, but also single-stranded RNA ligase are really applicable. Will we do all of those things ourselves? I don't know yet, right? But there may well be opportunities for significant partnerships to extend our reach.
Got it. And then, I mean, I think the follow-up there, you've partnered a number of these things off, and you might answer what I'm about to ask in your last comment, but that includes a life sciences portfolio to Alphazyme, the gene therapy assets to Crosswalk, the DNA ligase to Roche, and then the RNA polymerase to Aldevron. Obviously, we don't have the most visibility into what those look like. Can you just talk about the economics you get from those relationships? And then which of those do you view as kind of most impactful as we think about the financials?
Right. And so there are stakes we've got in the ground where we're interested in the future potential of that market and we want to play. That would be an example of that would be Aldevron, where we are strong believers in the potential of mRNA. And so what we went for there was a relatively modest upfront, but significant royalties on their sales, which we're super excited about as they move forward. There are other fields where we said, we've built a beautiful enzyme, but we really don't understand this channel. We don't know how to get into it. And that would be, for instance, the transposases that we built and we outlicensed to a company in the field there. And we have a relationship where we'll get a royalty downstream. Now, with the double-stranded DNA ligase, we sold that to Roche, right? Because we didn't understand that market.
It wasn't something we were committed to. We said, right, on the barrelhead. That sets the sort of precedent for the future where if you have a field where you want a really cool enzyme developing, then we are perfectly willing to do the work for you and sell it to you, right? And that also streamlines my organization because the problem we have is when we're trying to maintain rights to the enzyme and service the IP and all the rest of it, it becomes a bit of a snowball where we've learned that we can't do everything. We're better off sometimes having a flexibility in terms of the transaction we'll do to optimize it for Codexis.
Yeah. That's helpful. So maybe kind of last question, putting in or wrapping up on the balance sheet. Just touch on where the balance sheet stands, the path, the cash flow, break even, and then maybe along those same lines, kind of the key areas for investment over the next couple of years.
Right, so I think we ended the last quarter with approximately $90 million in cash, which was good. We raised $30 million during Q3, which, we're going to half of it approximately, is going to stay on the balance sheet. Half of it, we're going to invest in a couple of key areas. We had a path to profitability before that, but my goodness, it was tight, right? So some of what we've done is given ourselves a little more cushion in terms of not being on fumes when we get towards the end of 2026. We expect the burn to reduce us all over the coming quarters and years as we flip to being a profitable company from 2027 and beyond. Our current runway takes us out into 2027, so that all hangs together nicely, as I say, based on current products and current customers.
We've already identified some areas to invest internally in terms of pharma manufacturing, that internal manufacturing capability, controlling more of the RNA ligase manufacture, and that's investment of a few million dollars in downstream processing and purification so we can really keep control of those assets before we need to farm them out, and then the big ticket item in terms of potential future investment is going to be our own GMP facility for siRNA, but the timing of that is going to be dependent on how successful we are in getting the right scaling partnerships quickly, how successful we are in getting traction with ECO, and also location cost, and looking to be opportunistic in terms of facilities that might be available where we can get them very cheaply and retrofit our technology.
All right. Well, Stephen, thanks for being here. A lot covered. Appreciate you being in Nashville with us.
Thanks for the invite.