Codexis Inc (CDXS) 2017 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q3 2017 Codexis Earnings Conference Call. (Operator Instructions) As a reminder, this conference call may be recorded.

I would now like to introduce your host for today's conference, Ms. Jody Cain. Ma'am, you may begin.

This is Jody Cain with LHA. Thank you for participating in today's call to discuss Codexis' 2017 third quarter financial results and the company's business progress. Joining me from Codexis are John Nicols, President and Chief Executive Officer; and Gordon Sangster, the company's Chief Financial Officer.

During this call, management will be making a number of forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. To the extent these statements made by management are not descriptions of historical facts regarding Codexis, they are forward-looking statements reflecting the current beliefs and expectations of management as of November 9, 2017. You should not place undue reliance on these forward-looking statements because they involve known and unknown risks, uncertainties and other factors that are, in some cases, beyond the company's control and could materially affect actual results. For details about risks -- these risks, please see the earnings release that accompanies this call and the company's SEC filings.

Codexis expressly disclaims any intent or obligation to update forward-looking statements, except as required by law.

Now I'd like to turn the call over to John Nicols. John?

Thanks, Jody. Good afternoon, everyone, and thank you for joining us.

This is a very exciting time at Codexis, especially given the announcement last month of our latest major strategic collaboration, this time with Nestlé Health Science, a business unit of Nestlé. It is incredibly gratifying to add Nestlé to our growing list of great companies that agree with us as regards to the power of proteins, proteins that only CodeEvolver can discover and engineer.

Nestlé Health Science follows our recent history of major partnerships with GSK, Merck and Tate & Lyle. Partnership deals like these are core to continuing to deliver growth in our financials, and you should expect more great companies to be added to this list over time. We have only just started to tap into the vast potential of engineered proteins as new value-creating materials.

The new partnership with Nestlé also highlights Codexis' creativity and flexibility in structuring major partnerships to fit the needs and strategies of different global -- globally leading companies. The partnerships with GSK and Merck centered around providing those pharmaceutical industry leaders with the ability to run our CodeEvolver protein-engineering platform technology on their own. In doing so, our protein-engineering capacity replicates, discovering more proteins as cost-saving catalysts (inaudible) could ever do by ourselves.

Now GSK and Merck are using their own resources to create a stacking stream of novel proteins from which Codexis will derive sustained growing back-end revenues. These back-end revenues are beginning to unfold, noting that we expect 7-digit revenues from the back ends of these deals to start in 2018.

We expect more major partnering deals of this platform license type going forward, evidence our recent CodeEvolver user forum that we held in October. I'm excited to say that representatives from 9 of the top 20 global pharmaceutical companies participated in this successful event. The forum featured the growing use of protein catalysts in pharmaceutical manufacturing and represents a solid pipeline of platform license prospects for Codexis.

The partnering -- partnership model with Tate & Lyle is different, but in no way less significant. In this partnering model, we collaborate on a specific breakthrough project basis. These projects aim to transform a targeted food ingredient product that Tate & Lyle envisions will enable accelerated growth in their downstream food and beverage markets.

If successful, as we have already demonstrated with Tate & Lyle, our partner gets control over the CodeEvolver-enabled, low-cost process technology for their very cost-competitive markets and Codexis patents profitable protein products that can commercialize rapidly with above-average peak revenue prospects.

We commenced development of our first commercialized product for Tate & Lyle only 3 years ago, and that product is now our second highest-selling product in 2017. Following that success, the second project with Tate & Lyle announced in March this year is progressing above expectations. If successful, this product could be a significantly larger revenue producer than our first successful product with potential for 8-digit product sales at peak.

Year-on-year revenue growth in 2018 with Tate & Lyle is secure. We expect more major partnering deals of this specific breakthrough project type going forward, both of as follow-ons with Tate & Lyle and with other great companies in a potentially widening variety of industrial verticals.

And now, our Nestlé Health Science collaboration secures yet another mode of strategic partnering. The core of this collaboration

(technical difficulty)

[Nestlé's] enthusiasm for Codexis' independently discovered biotherapeutic pipeline and also adds strategic access to our CodeEvolver platform for discovery of proteins that are new to both Nestlé and Codexis. I want to spend some extra time detailing this new partnership arrangement, highlighting its many features that make it exceptionally exciting for Codexis.

Key to the collaboration is Nestlé's global licensing option for [these] therapeutic assets, the proprietary orally deliverable enzyme CDX-6114 for the management of the metabolic disorder phenylketonuria, or PKU. We are particularly proud of this agreement because it is our first revenue event for a Codexis-developed biotherapeutic. And even more important, we now have the strength of Nestlé Health Science to credibly validate that our CodeEvolver platform is indeed a drug discovery engine.

If CDX-6114 is successful, the deal has potential for total payments to Codexis above the $357 million plus tiered royalty of low double-digit percentages on future product sales.

We are scheduled to receive a $14 million upfront payment from Nestlé in this fourth quarter, which more than covers the investments we've made in CDX-6114's discovery and preclinical activities to date.

We remain on target to initiate a Phase I clinical trial in healthy volunteers early next year, which will trigger an additional $4 million milestone payment from Nestlé.

Following that trial's completion, which is expected around the end of (inaudible) Nestlé will have 60 days to exercise its exclusive license option. A favorable decision will trigger a payment of $3 million to Codexis.

(technical difficulty)

Nestlé takes responsibility for all expenses related to future development and commercialization activities for CDX-6114. Thereafter, Codexis can earn additional preclinical milestone payments through regulatory approvals that add up to $86 million.

If CDX-6114 hits the market, Codexis is then eligible to earn both sales-based milestone payments that add up to $250 million as well as tiered royalties on sales up to low double-digit percentages. The combined economic deal terms are remarkable for a preclinical therapeutic asset and show that a healthy percentage of the economic value from a successful CDX-6114 have been negotiated with Codexis.

With this excellent result from our first biotherapeutic discovery endeavor, our focus now is on following that up with another biotherapeutic assets in our drug discovery pipeline.

Today, that

(technical difficulty)

4 rare disease conditions that we believe our CodeEvolver protein engineering can address. In addition to the option for PKU, the Nestlé deal gives you a look into the quality of that remaining biotherapeutic pipeline. Allow me to explain.

Of our 4 pipeline disease programs, several target inborn errors of amino acid metabolism. Like PKU, where metabolism of phenylalanine is the central challenge, these earlier-stage pipeline programs address other amino acid metabolism deficiencies.

Nestlé Health Science is one of the leaders in providing nutritional products for patients with such diseases through their wholly owned subsidiary, Vitaflo. They know the patients, the specialist clinicians and the key opinion leaders in this category of rare disease.

While we have yet to publicly share specifics of our drug discovery programs, it is noteworthy that Nestlé saw our amino acid metabolism discovery programs confidentially, which drove their decision to establish a right of first negotiation as part of our partnership deal. In other words, they validated that we are on the right track with our drug discovery pipeline beyond PKU.

The third and final element of the partnership with Nestlé Health Science puts our CodeEvolver R&D capacity to work to discover novel proteins targeting performance attributes for Nestlé's portfolio that no currently available proteins can achieve. Here, the partnership aims to discover enzymes for Nestlé Health Science's established consumer care and medical nutrition businesses. Codexis receives separate R&D fees for these discovery projects, with R&D revenue generation already beginning in this fourth quarter.

Before switching over to financials, I'd like to conclude the strategic update section by sharing our continued enthusiasm for our efforts, launching new proprietary enzymes into the world's molecular diagnostics industry.

Our first product, a DNA ligase engineered by CodeEvolver for rapid and efficient conversions at ultra-low DNA concentrations, like what is needed in liquid biopsy, is receiving excellent market perception as we have moved through 2017.

In addition,

(technical difficulty)

engineered molecular biology enzyme, which has the promise to have a significant impact on the cost and efficacy of manufacturing for messenger RNA therapeutic products.

Our DNA ligase technology will be featured in a poster presented by our scientist, Dr. Matt Miller, at the Association for Molecular Pathology, or AMP, meeting on November 17 in Salt Lake City. The abstract will also be published thereafter in the November issue of the Journal of Molecular Diagnostics. AMP is the leading -- is a leading organization in the field of molecular diagnostics, making it an ideal book case for our technology.

We also enjoyed a very favorable reception at the 5th Messenger RNA Health Conference held in Berlin on November 1 and 2 in 2017.

Beyond these 2 products, we are now using the CodeEvolver technology to develop a third enzyme for use in molecular diagnostics. We are confident in generating meaningful market penetration and revenue from these high-performing novel enzymes in 2018.

Before I hand off to Gordon to go deeper, I'd like to highlight a few elements from our solid third quarter financial performance. Once again, the star goes to growing product sales, which were up 71% for the quarter and up 73% for the 9 months year-to-date. In addition, those product sales continued to be delivered with improving gross margin percentages, 43% for the quarter and 44% for the 9 months year-to-date, both well ahead of the 36% gross margin delivered in 2016.

We are also reporting a solid quarter for R&D revenues. Excluding the technology transfer revenues recognized with Merck in last year's third quarter, R&D revenues for this year's third quarter were up 68%, year-over-year.

Now let me turn the call over to Gordon to provide more details on the financials. Gordon?

Thanks, John.

I will share some highlights of our financial results starting with the third quarter. Total revenues for the third quarter of 2017 were $10 million. This compares with $14.9 million for the third quarter of 2016, which included the recognition of $8.6 million in revenues from R&D milestones and related deferred revenues. As John stated, product sales for the 2017 third quarter increased by 71% from the prior year to $6.9 million, which reflected an increase in demand for our enzymes across a broad customer base.

R&D revenues for the quarter were $2.9 million. This compares with $10.4 million for the prior year, which included the recognition of an $8 million milestone payment from the completion of our CodeEvolver technology transfer to Merck and $600,000 in deferred revenue related to the transfer. As John mentioned, excluding the recognition of these items in the year ago quarter, R&D revenues increased by 68%.

Revenue from our revenue-sharing arrangement with -- for the argatroban injectable drug with Exela was $107,000 in the third quarter of 2017, down from $445,000 in the third quarter of 2016.

Gross margin on product sales for the third quarter of 2017 was 43%, up from 32% from the prior year period, with the improvement due to product mix.

Turning to operating expenses. R&D expenses were $8.1 million for the third quarter of 2017, up from $5.5 million in the prior year. The increase was primarily due to higher outside services related to CDX-6114 and the increased costs associated with higher headcount.

SG&A expenses for the third quarter of 2017 were $8 million compared with $5.2 million for the prior year. This increase was primarily due to higher legal fees and costs associated with our headcount.

The net loss for the third quarter of 2017 was $10.2 million or $0.21 per share. This compares with net income from the third quarter of 2016 of $1.4 million or $0.03 per diluted share which, again, reflected the recognition of $8.6 million in milestone payments and deferred revenue.

On a non-GAAP basis, adjusted net loss for the third quarter of 2017 was $8.2 million or $0.17 per share versus adjusted net income for the third quarter of 2016 of $4 million or $0.09 per diluted share.

Turning to our year-to-date financial results. Total revenues for the first 9 months of 2017 were $28.3 million. This compares with $38.9 million for the first 9 months of 2016, which included the recognition of $20.4 million from Merck and GSK related to R&D milestone payments and deferred revenue.

Excluding these milestone payments and deferred revenue in the prior period, revenues for the first 9 months of 2017 increased by 53%.

Revenues for the 9 months ended September 30, 2017, included $19.1 million in product sales, $8.3 million in R&D revenues and $0.8 million from the revenue sharing arrangement with Exela.

Gross margin on product revenues was 44% for the first 9 months of 2017, and R&D expenses for the first 9 months of 2017 were $20.2 million, and SG&A expenses were $21.1 million.

We reported net loss for the first 9 months of 2017 of $24 million or $0.53 per share, and this compares with the net loss for the prior year 9-month period of $3.3 million or $0.08 per share.

On a non-GAAP basis, adjusted net loss for the first 9 months of 2017 was $18 million or $0.39 per share versus adjusted net income of $4.5 million or $0.11 per diluted share in the prior year.

Cash and cash equivalents as of September 30, 2017, were $23.8 million versus $19.2 million as of December 31, 2016.

Let me now turn to our financial guidance for 2017. We are affirming our expectation for total revenues for the year to be between $50 million and $53 million. This guidance assumes significant revenue from our Nestlé collaboration in the fourth quarter. We also are reaffirming our outlook for 2017 product sales to range from $25 million to $27 million, which represents an increase of 63% to 76% over 2016. As I mentioned earlier, product sales for the first 9 months of 2017 were more than $19 million.

We are also reaffirming our expectations for 2017 margin on product sales to be between 40% and 43%.

As discussed on last quarter's call, we indicated an increase in operating expenses in the second half as compared to the first half of 2017. For the fourth quarter 2017 operating expenses, which had combined total of R&D and SG&A, we expect approximately $16 million. This reflects the additional cost for the accelerated development of the PKU drug candidate to move this asset more rapidly towards a Phase I trial, and additional legal expenses.

And finally, we plan to introduce guidance for 2018 revenues in January. We'll provide 2018 financial guidance on additional metrics with our fourth quarter conference call in early March 2018.

With that, I'd like to turn the call back to John.

Thanks for that financial review, Gordon.

We are nearing completion of a highly productive (inaudible) for Codexis. Our fourth quarter is set up to be the highest revenue generator for the year, enabling us to reaffirm our 2017 revenue guidance of $50 million or above. Delivering on that 2017 revenue guidance will be an exceptional accomplishment by the Codexis team, especially given that over 20 million of last year's -- $20 million of last year's revenue were nonrecurring, relating to the completion of both GSK and Merck platform license technology transfers.

I'm proud to say that we are offsetting those substantial headwinds with multiple new sustaining revenue sources this year. Specifically, we entered into 2 new major partnership deals with Nestlé and Tate & Lyle, we continued to broaden the high-value markets we're reaching with our leading-edge CodeEvolver platform technology, and we're translating our growing pipeline into step-change growth in product sales.

It's equally exciting to report that our revenue sources show momentum as we continue into 2018. Year-on-year growth is expected from both of the Nestlé and Tate & Lyle partnerships, and we have visibility to low single-digit million dollar revenues from the back ends of the GSK/Merck platform license deals.

Add to that prospects for continued growth in product and R&D revenues from our pipeline's momentum, new sales of high-performing enzymes into molecular diagnostics and the probability of executing a fifth major partnership deal, we are confident in returning to double-digit revenue growth in 2018.

The strength of our progress on these multiple fronts further positions Codexis as a dynamic, thriving company with leverage to help us achieve and sustain profitability. These are exciting times here at Codexis with more to

(technical difficulty)

continue to accelerate, unlocking the power of proteins.

With these comments, I'd like to open up the call for questions. Operator?

(Operator Instructions)

While waiting for our first question, I'd like to alert you that we will be presenting at the Craig-Hallum Alpha Select Conference and at the Furey Research Partners 2017 Hidden Gems Conference. Both conferences are being held in New York City on November 16.

A webcast of our presentation at the Craig-Hallum Alpha Select Conference will be posted to the Investors section of www.codexis.com.

For those of you heading to San Francisco for the annual J.P. Morgan Healthcare Conference, we'll be holding one-on-one meetings across the street from the conference hotel.

(technical difficulty)

dates are January 8 through 10, and please give LHA a call if you want to schedule a meeting.

Okay, operator, we're ready for the first question.

And our first question comes from Matt Hewitt from Craig-Hallum.

This is Charlie on for Matt. First, I mean, can you talk about the length of the sales cycles for these partnerships? How long do they typically take to put in place? And are you seeing that time line shorten at all as you continue to grow the business?

Well, it takes a good bit of time to establish the partnerships of these magnitude, like with GSK, Merck, Nestlé and Tate & Lyle.

(technical difficulty)

Certainly, it can take as much as

(technical difficulty)

to establish partnership relationships of that magnitude. During those periods of time, we often will work on smaller projects, maybe specific

(technical difficulty)

for specific drugs in the case of major pharmaceutical companies. And that helps to establish a growing rapport between Codexis and these large companies and helps to build an understanding of the power of CodeEvolver within these great companies and also gives us time to really explore all the variety of different applications that our technology can supply. And so as you see, these partnerships affect many different potential parts of these great companies. So it takes time for us to establish them, but during that period of time, we're generating revenue, building relationship and being closer and closer to partnership models, like I highlighted on.

Sure. That makes sense. Now that the Nestlé deal has happened and you've partnered your first Codexis-generated biotherapeutic, are you having different conversations in the marketplace?

Yes. Right now, we're really focused on executing the Nestlé deal, of course. There's still significant economics right in front of us for continuation on that deal. And we're

(technical difficulty)

a bit more internally in the biotherapeutic discovery area, really just advancing those early-stage biotherapeutic projects that I gave you some color for on the call. We're looking into advancing those to a point where they are marketable assets like PKU, obviously, has become. And that's going to take more internal work for us to reach those stage against those 4 follow-on biotherapeutic targets beyond PKU. So not a lot of external conversations on those programs at this point.

Our next question comes from Brandon Couillard with Jefferies.

John, on the Nestlé deal, curious if you could just inform us why now was the right time to do this deal before moving in to Phase I studies? And when do you think you'll be in a position to discuss more broadly the types of assets that are in the pipeline? And what is the--your time line in terms of, I guess, progression for those projects?

Yes. Okay, great. So 2 questions there. So why now? First, we've been moving closer with Nestlé on a variety of potential topics. And so why now with Nestlé is because we were able to put this much broader relationship in place. Of course, it's centered around the PKU therapeutic asset, but it also included other multiple elements, some interest in that pipeline, which is your second question, which I'll get to. And it also enabled us to put a platform access element in place to start working on some new proteins. Finally, I

(technical difficulty)

generated very significant economic deal terms from the PKU asset in this preclinical stage. I believe that we wouldn't have generated significantly more economics had we brought the asset into Phase I and generated positive safety data. I feel like we're very close to the economics of that kind of value inflection point even before having reached it. So we're extremely pleased with the economics that we were able to negotiate for the preclinical asset. And so those are some of the best reasons why now. So your second question, other pipeline assets. So I gave you a little bit of a feel probably the first measure of any detail of what other programs we're targeting in our pipeline beyond PKU. And so you heard that a few of the targets are also inborn error of amino acid deficiencies like PKU, so that's a measure of color. We're going to be careful about public disclosures about these programs. Certainly, we want to get our patent situation very tight and be clear before we provide publicity. And second, we don't want to provide publicity until we start to see real positive data in preclinical trials, real efficacy in living organisms, living animals. So those are kind of the 2 key milestones for when we'd be comfortable to start disclosing the details of the other biotherapeutic assets. I'm hopeful that as we move into 2018, maybe as early as the second half of next year or early in 2019 that you'll start to see what those

(technical difficulty)

And we'll start to be able to share our excitement around specific programs like we've already done with PKU. So hopefully, those give some nice answers to your 2 questions, Brandon. Any comments?

That's very helpful. And then one for you, Gordon, sticking with the Nestlé deal. There seems to be 2 elements of -- in terms of the near-term revenue impact, both from the PKU upfront and the milestone as well as the R&D program. Just to make sure, the $14 million up front, that will be amortized over the next 12 months? And then secondly, in terms of the R&D revenues that you alluded to coming in the fourth quarter, any chance you could break that out in terms of the relative size for us?

The collaboration revenues, the R&D piece that they're paying us are about $150,000 a month, so about $1.8 million to $2 million a year. And then the revenue recognition from the $14 million is much more complicated. That's something that we've been working on over the last month or so. So that's something that we'll be able to talk more about once we release year end results in March.

Then John, on the GSK and Merck back ends that you pointed to in 2018, I think you've talked about each of them having -- working on, I think, 5 projects for each of them. Curious as to how many of those you're banking on being contributors to the back end for next year?

Yes. So our understanding of Merck and GSK's protein engineering CodeEvolver teams is that each of them enable in the range of -- 5 seems high, but certainly, 3 or 4, maybe more parallel protein engineering teams. And that compares to here at Codexis, we have about 15 parallel protein engineering teams, to give you some color on that detail. As we start to see revenues generated from the back end, it will be a very few of the projects that have been worked on by GSK and Merck that start to contribute revenues as quickly as 2018. Most of the projects won't start to show back-end revenues until years later. So it's 1, maybe 2 products that will contribute to the low single-digit million dollar back-end expectations starting next year.

Last one, Gordon, do you still expect to book product revenues in the fourth quarter from the second Tate shipment? Or has that been pushed out at all? Just wanted to make sure.

No, no, we still expect to be able to ship that in the fourth quarter and recognize the revenue.

And our next question comes from RK from HC Wainwright.

Two questions from me. But the first one, in your initial commentary, you said expect increasing agreements with wider verticals. These are the comments you made when you're talking about the Tate & Lyle agreement. So could you provide some color as to what these potential verticals could be and what sort of conversations you are having within the food industry?

Sure. So yes, I think I alluded to us potentially starting to talk about other industries besides pharmaceuticals and food several quarters ago. And we continue to have really good progress to penetrate into other market arenas with other great potential clients. They would come in either of 2 different types. First, as you know, CodeEvolver is very competent at creating catalysts, protein-based catalysts that can enable low-cost manufacturing. That's been our core business for the whole history of the company for pharmaceuticals and has been a key part of how we're generating value for Tate & Lyle. So catalysis is fundamental to all chemical manufacturing. So another industry vertical could come from other great chemical industry verticals, and that could come from flavor and fragrances or specialty chemicals or other varieties of chemical industries. So that would be of one type of extension into -- of our technology into other industry verticals. The other type, we operate within -- we create enzymes with CodeEvolver, and there's a significant multibillion industrial enzyme industry dominated by very large competitors like Novozymes and DuPont. And they provided enzymes to a wide range of industries

(technical difficulty)

can impart performance benefits for products that are needed in those industries. These are industry sectors, for example, enzymes are used in detergents. They're used in pulp and paper. They're used in animal feed. They're used in a whole variety of different industries. And so Codexis, with our world-leading protein-engineering capability, has been talking to customers in this variety of industrial enzyme sectors basically trying to see what kind of new enzyme performance needs those markets and those customers are looking for. And that could be a second category where we bring an improved enzyme into one of those industrial enzyme verticals. There are 2 different types of approaches. We're working on both of them. We're having some successes in both of them. So I'm hopeful that in the not-too-distant future, we're showing you material deals with great companies like those.

Fantastic. Actually, with all these various verticals, which have been doing well and which have also come to life, especially in the recent few quarters, we're talking about pharma, food, biotech, enzymes and, now recently, the proprietary enzymes, which you just did a deal with Nestlé. So with all this good success going on, are you facing any resource issues on one side? And then on the other side, does this success mean you and the board have to sit down and figure out where do you want to take this whole CodeEvolver thing so that you can forecast and also maintain sustained growth in the long term?

Yes. Certainly, a great question, RK, and of course, we work these issues constantly. So first and comforting is in the R&D arena, a -- like I said earlier, we have around 15 different parallel protein-engineering teams that are working on 15 new proteins at any given point in time. So to work one of those teams in one area versus another protein for another team in another area, the R&D teams are incredibly fungible. And so the need to focus in the R&D arena is mitigated as a result. So really, it comes down to the commercial choices. And here, the main constraint is our ability to access and build relationships with great, huge, complex companies like the companies I highlighted we have major partnerships with. And so yes, there are definitely some limitations and we're adding smartly headcount both in R&D and to enable more project work and our business development teams. But the need to focus at -- on one at the exclusion of other is also mitigated because essentially, each of the projects we work on generate return on investments. So if we add another R&D team and we had another good idea to create a protein, we're usually getting our R&D fees covered by the partnership, so Codexis itself is not digging an investment hole. And then if it works, we're generating back ends, where we have often.

(technical difficulty)

to protect our position. So we're really looking at expanding the applicability as opposed to picking one over another. So hopefully, that gives you some feel on your question.

No, no. That's good. I guess the last question from me, obviously, the success in generating this agreement with Nestlé is really -- I'm sure, is very encouraging for you and the team. And I know -- I do understand that you can't really relay too much of what is there in the pipeline at this point, but how much of an effort is it to maintain a pipeline of a constant stream of such products? And do you need to go out of your comfort zone to do these kind of [preplanned] and clinical studies, at least the Phase I studies, before you give it off to somebody else to shepherd the additional development that those molecules require?

Yes. So really good question again. So our bio

(technical difficulty)

pipeline, I referenced in my commentary about 4 projects. A couple of them are these inborn errors of acid metabolism, so we gave some color today. Very limited, I know, until we develop our patents and efficacy data. That requires roughly 3, maybe 4 of our protein engineering teams. So several times, I've referenced we have about 15 teams, so roughly, a healthy minority or 20% or 25% of our R&D capacity is dedicated to that pipeline. One of those projects is now part of this element of collaboration where we're creating new enzymes for Nestlé's consumer care and medical and nutrition. So one of them is covered by the Nestlé R&D fees that we're generating. So it's still a minority of our R&D capacity is focused on this pipeline of efforts. It does challenge us because generating a catalyst that has efficacy -- or a protein that has efficacy as a catalyst requires testing, improving the catalyst yield and stability, fairly straightforward things to do in the laboratory. For therapeutic, ultimately, we have

(technical difficulty)

in humans and for that in animals. So yes, that -- proving a protein is (technical difficulty) in living things has -- is something that we're investing in, building technology, understanding. But I will highlight that we actually pushed through that very successfully with the PKU program, so we've been building confidence on these kinds of technical challenges.

(Operator Instructions) Our next question comes from Steve Schwartz from First Analysis.

First question, have you disclosed how much you've invested into CDX-6114?

Not specifically, but I did say today that the $14 million upfront payment from Nestlé more than covers the investment to date in the discovery and the preclinical efforts.

And, oh, John, exactly. And you know when you make a comment like that to a bunch of finance people, they're going to ask for numbers. So anyway, okay. I understand if you don't necessarily want to disclose that figure, but I did catch that comment. A simple one, with respect to the Exela revenue, is this the new near-term run rate, where you're at this quarter?

I think so, yes.

Most likely. The product has faced an increasing generic competition, which has led to the numbers shrinking to pretty small numbers in this third quarter. So most likely, it will stay at this level, maybe even get closer to 0 over time. So it's become, obviously, a much less-important piece of the company's revenue.

Yes, okay. And then when -- certainly, you've talked at length about your penetration of CodeEvolver at large pharma, and most of these deals are coming from larger entities. What part of your business is with small and medium biopharma?

Gordon, it's a really good question. The reason we spend more time talking about big pharma is because -- large pharmaceutical companies is because they have so many more targets. So many more APIs can be affected by catalysts that CodeEvolver can create. And so the larger companies are better prospects for a platform license deal. So -- but we do a significant amount of business with smaller entities. We do a significant amount of business with generic companies, which are categorically much smaller than big GSK and Merck. One of our more material revenues in the third quarter

(technical difficulty)

from a much smaller pharmaceutical company. Typically, the small pharma relationship is 1 project, 1 drug because that's the character of many of these smaller pharmaceutical companies. But clearly, the biocatalyst cost reductions can help smaller companies, and it does.

Yes. So the biopharma world is all abuzz right now around the opportunity related to biosimilars. And is that a wave in particular that you can ride? Or is your world of opportunity so big that, that one particular area doesn't necessarily stand out for you?

Biosimilars is a very important, big trend in the world of pharma, as you just highlighted. But our technology, our CodeEvolver changes proteins. It is, in essence -- in its essence, it is creating a new novel protein, changing amino acid sequence

(technical difficulty)

[something] new. So

(technical difficulty)

technology creates new proteins. And so when you think about biosimilar, biosimilar is the same protein just in the generic chapter, which leads to great opportunities for manufacturing large -- larger volume protein manufacturers. CodeEvolver create new proteins. So the CDX-6114 is a new novel protein, has to go through clinical trials, et cetera. That's a better

(technical difficulty)

for CodeEvolver. And that's why we're not so much in the middle of the biosimilars wave. Does that make sense?

Sure. Yes, I totally understand. And that actually is a good lead-in to my next question, John, which is, I don't know that I have a good handle on exactly how many molecules you've been involved with and how many molecules have moved to various stages in preclinical or clinical development. And I'm just wondering if you've acquired enough data at this point to where, in very simple terms, if you do a CodeEvolver sales pitch to someone in pharma, can you -- at this point, can you say with confidence that you can reduce development time by x percent or you will improve the probability of approval at certain stages in drug development by using CodeEvolver? Is that information out there yet?

Not really. And the vast majority of our projects in pharmaceuticals is involved in the process to make the pharmaceuticals. We're helping them to create a process to make the drug. And the key for us -- by delivering a protein catalyst that makes the drug cheaper or the intermediate cheaper. And so the key for us historically has been to engineer that catalyst quickly enough to fit the time line of our pharmaceutical client. And so because the process doesn't drive the time line, the clinical efforts -- the proof through the Phase I, II, III drives the time line, but if

(technical difficulty)

job at

(technical difficulty)

our catalyst installed within those time lines so that they don't become critical path because the process change is going to drive the customer's time line. Only the regulatory and clinical results will drive the time line. So yes. And sort of not answering your questions, but us timing is to make sure we're not a timing issue that we can get our catalysts in their hands quickly enough. And as CodeEvolver's gotten faster and better, we've clearly been able to do that, which is part of our sales pitch.

Yes -- no, John, you did answer my question perfectly. In fact, you -- I think you made me realize it might not have been a good question to begin with. But no, so that was a good answer. John, when you talked about the Nestlé deal, you talked about a max of $357 million. And when you got to the point at which if the molecule goes commercial, you noted $250 million as a max, but then you also talked about a certain percentage, high -- I think it was a high-teens percentage of royalty. And that seemed to be uncapped, that element of it. Can you talk through that again?

Sure. You're absolutely right. Once the drug -- okay, so there are $107 million of potential payments from today through the approval of the drug, so that's $107 million. $14 million in the upfront. Okay, so once the drug gets approved, then Codexis

(technical difficulty)

the possibility of generating 2 different sources of revenue, which are additive. The first are

(technical difficulty)

of sales milestones. Those milestones can add up to $250 million. But in addition, in parallel, we will be generating tiered royalties on the product sales, and the upper end of the tier is low double-digit royalty rate once we generate both of the revenue sources in parallel.

Got it, okay. And then my last one is for Gordon. Gordon, in the press release this afternoon, where you address R&D and SG&A expenses, you talked about it being higher in the fourth quarter on legal, just getting ready for clinical trials. But can you give us an idea on what the run rate might be going through 2018?

(technical difficulty)

significantly for OpEx from point of view of PKU. But also legally.

Yes, and it will be offset by

(technical difficulty)

account. So we're seeing a

(technical difficulty)

new projects to put our teams to work, and so that's encouraging us to add to the R&D ranks. And I think it was RK, asked about constraints, and so we were incrementally adding headcount for business development. So I think next year, the PKU spend will be down, like Gordon said, but it will be offset somewhat by increased headcount. And I think you're right, Gordon, overall, that

(technical difficulty)

could come down on average off of the $16 million that we were pretty sharp to give you for the fourth quarter.

Yes.

Our next question comes from Greg Venit from Morgan Stanley.

So in the -- Tate & Lyle is considered, what, a food company or food ingredient type company, a flavor house. Would that be correct?

I wouldn't call them a flavor house. They're one of the world's leaders in food ingredients and things like fillers and sweeteners and functional additives to make food products more appealing or more healthy.

And those products are quicker to market, so quicker revenue for Codexis. Would you -- do you have this 2 year where you -- 2-year timeframe, every 2 years timing a deal. Do you think the next deal will be with a pharmaceutical company, which takes longer to bring something to the market? Or do you think it would be with a food ingredients company? Or do you think it would be maybe an industry that you're -- one of the other industries that you talked about that you're not in yet?

Since you put ours between all of that, I'll say yes. We're not foreshadowing which kinds of deals will be the next one. We're just making it clear that we have prospects for seats

(technical difficulty)

major partners across those 3 different categories.

Yes, I guess, I'm just trying to get a -- getting revenue faster in the company versus the long lead time it is for pharmaceuticals, except for the research and development money that you get or the upfront payment.

Yes. It's happening, Greg. First, we've been working on many projects over our 15-plus-year history, and those projects are requiring R&D sometimes and

(technical difficulty)

for commercialization of those. And those are commercializing -- accelerating in their commercialization, which is leading to our product revenue growth. And that's all -- you can kind of get a feel for that in our 2 years of showing you our pipeline snapshot now. So that's all happening very nicely. And then on top -- then we add these new partnerships. And these new partnerships create new short-term and long-term revenue opportunities. And you're right, with the food ingredients area or generally every other industry, it should be theoretically quicker for us to commercialize to ultimately get to that sustaining revenue place because every other industry doesn't require all the regulatory time and effort that pharmaceuticals do. So by growing our business in other industries, we're accelerating that time line to sustainable revenues.

With Tate & Lyle, with food ingredients, usually consumer -- CPG, consumer packaged goods companies take a long time, a lot of testing before they're going to change an ingredient in one of their products. But (inaudible) -- are you doing something with them that's a brand-new product that they don't have to do testing? Or is it replacing -- is Tate & Lyle providing an ingredient that's replacing an ingredient that's already being used where they may want to test it?

Yes. It's

(technical difficulty)

through our partnership with Tate & Lyle, we're helping to create new food ingredients that could be used in existing types of CPG products.

So there's a chance -- I'm just wondering the size of -- for the size of -- the existing, what they've got there, if they're slow in rolling these out? Or do you feel like these things, over the next year or 2, the markets define as far as what the volumes is going to be?

You know, you're right. There is resistance to change in big industries like the consumer products companies. But the fact that we've already -- that Tate & Lyle's already been able to commercialize something that we just started 3 or so years ago is a testament to the other process

(technical difficulty)

relatively quickly. And now, we're working on our second project with Tate & Lyle. And as I alluded in my prepared remarks, that could be even bigger than the first success we

(technical difficulty)

And we're hopeful that it also can commercialize in this kind of a timeframe. So...

And I'm showing no further questions from our phone lines. I would now like to turn the conference back over to management for any closing remarks.

Okay. [Thanks for the] questions.

Relentless focus on our core CodeEvolver protein-engineering platform technology, getting more customers using that technology and capturing more of the value we

(technical difficulty)

customer. That's our recipe here for success at Codexis.

(technical difficulty)

forward to providing ongoing progress updates to you. In the meantime, thanks, and have a great day.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program, and you may all disconnect. Everyone, have a wonderful day.