ICON PLC (ICLR) 2018 Q3 法說會逐字稿

Good day, and welcome to the ICON plc Quarter 3 2018 results conference call. Today's conference is being recorded. At this time, I would like to turn the conference over to Jonathan Curtain. Please go ahead, sir.

Thank you, Molly. Good day, ladies and gentlemen. Thank you for joining us on this call covering the quarter ended September 30, 2018. Also on the call today, we have our CEO, Dr. Steve Cutler; and our CFO, Mr. Brendan Brennan. I would like to note that this call is webcast and that there are slides available to download on our website to accompany today's call.

Certain statements in today's call will be forward-looking statements. Actual results may differ materially from those stated or implied by forward-looking statements due to risks and uncertainties associated with the company's business, and listeners are cautioned that forward-looking statements are not guarantees of future performance. The company's filings with the Securities and Exchange Commission discuss the risks and uncertainties associated with the company's business.

This presentation includes selected non-GAAP financial measures. For a presentation of the most directly comparable GAAP financial measures, please refer to the press release statement headed Consolidated income statements, unaudited, U.S. GAAP. While non-GAAP financial measures are not superior to or a substitute for the comparable GAAP measures, we believe certain non-GAAP information is more useful to investors for historical comparison purposes.

We will be limiting the call today to 1 hour. (Operator Instructions) I would now like to hand over the call to our CFO, Mr. Brendan Brennan.

Thank you, Jonathan. As a reminder, from January 1, 2018, the new revenue recognition standard ASC 606 became effective for ICON. Having adopted cumulative effects transition method, prior comparatives have not been restated under this new standard.

Instead, we feel that providing comparable Q3 '18 financial results under the previous revenue recognition standard is the best way to evaluate our performance during this transition phase. Whilst my comments may incorporate the impact of ASC 606, Steve will focus his comments in our performance excluding the impact of ASC 606.

In quarter 3, we achieved strong gross business wins, excluding the impact of ASC 606, with $716 million with cancellations of $111 million. As a result, net awards in the quarter were new record $605 million, resulting in a net book-to-bill of 1.27x. This means our trading 12-month book-to-bill is a very healthy 1.28x.

With the addition of these new awards, our quarter 3 2018 backlog, excluding the impact of ASC 606, grew to $5.3 billion, representing a year-over-year increase of 10.5%. Our top customer represents 9.8% of this backlog, down from 11.4% at the end of quarter 3 last year.

Reported revenue in quarter 3 was $655 million. Excluding the impact of ASC 606, revenue was $476.4 million. This represents year-on-year growth of 8.2% or 8.3% on a constant-currency basis and 6.6% on a constant-dollar organic basis.

Excluding the impact of ASC 606, our customer concentration with our top customer in the quarter represented 13.1% of revenue compared to 16.6% of revenue last year. Our top 5 customers represented 38.3% of revenue compared to 38.2% of revenue last year. Our top 10, represented 52.9% of revenue compared to 54.2% of revenue last year, while our top 25 customers represented 69.2% of revenue compared to 72.2% of revenue last year.

In quarter 3, reported gross margin was 29.9% compared to 30% in quarter 2. Excluding the impact of ASC 606, group gross margin for the quarter was 41.3%. This compared to 40.9% last quarter and 41% for the comparable quarter last year.

Reported SG&A for the quarter was 12.3% as compared to 12.6% in quarter 2. Excluding the impact of ASC 606, SG&A was 17% of revenue. This compared to 17.1% last quarter and 18% in the comparable period last year. This was a result of our continued leverage of our global business support model.

In quarter 3, we reported operating income of 15% or $97.9 million. This compared to 14.7% or $94.4 million in quarter 2. Excluding the impact of ASC 606, operating income for the quarter was $98.8 million and operating margin of 20.7%. This compared to 20.2% last quarter and 19.3% in the comparable period last year.

As reported, net interest expense for the quarter was $1.9 million and the effective tax rate for the quarter was 12%. Reported net income was $84.5 million or 12.9%. This equated to $1.54 -- sorry $1.54 in diluted earnings per share.

Excluding the impact of ASC 606, net income was $85.3 million, a margin of 17.9%, equating to diluted earnings per share of $1.55. This compares to earnings per share of $1.51 in quarter 2, which excludes the $0.03 tax benefit last quarter, and $1.35 in the comparable quarter last year, an increase of 14.8%.

DSO in the quarter was 49 days, which compared to 49 days last quarter and 50 days in the comparable quarter last year. Cash generated from operating activities for the quarter was $128.1 million and capital expenditure was 12 -- was $11.2 million. On September 30, 2018, the company had net cash of $142.3 million compared to net cash of $23.9 million at June 30, 2018, and net debt $56.3 million at the end of September 2017.

As a reminder, I have dealt with the new revenue standard. We will focus his comments on our performance excluding the impact of ASC 606. And with all of that said, I'd now like to hand over the call to Steve.

Thank you, Brendan, and good day, to everyone. During the quarter, we continued to see a strong demand for our services in the market. Our backlog in the quarter grew nearly 11% year-over-year driven by another record net business award's level of $605 million. This delivered a net book-to-bill of 1.27x for the quarter or 1.28x on a trailing 12-month basis.

Our overall P&L performance was again very solid with revenue increasing by over 8% year-over-year to $474.6 million; gross margin improving to 41.3%, up from 41% last year; and further SG&A improvement to 17% of net revenue from 18% in the prior year. In conjunction to the 12% tax rate, this meant that earnings per share grew 15% year-over-year to $1.55.

In addition, since the end of quarter 2, we've repurchased nearly $38 million worth of shares. Year-to-date, we have repurchased $92 million worth of shares overall at an average price of $125.58 per share. Our intention is to continue to opportunistically repurchase shares as we move towards the year-end.

Looking to the future and the overall market for CRO services, the key growth drivers remain in place. Biopharma R&D spending continues to strengthen at an estimated 3% per annum, while increases in the number of trials and outsourcing penetration rates will affect further increases as customers look to improve the productivity and efficiency of their drug development programs. Consequently, we estimate the CRO industry will expand by around 6% per annum over the next 4 to 5 years.

Market share continues to shift towards larger CRO's that have the global footprint, breadth of services and patient access necessary to run increasingly complex global trials. ICON is well positioned to capitalize on this growth given our breadth of capabilities, scientific and therapeutic expertise, and global scale, operating as we do across 37 countries from 93 locations.

To enable this growth, our core component of ICON's strategy is our focus on partnerships. In becoming the industry's trusted partner of choice, we differentiate our services by listening to, understanding and providing solutions to our customers' key challenges.

To do this, we need to not only understand our customer needs but also the market environment in which we operate. In recent years, there has been an exponential growth in the amount of data available in clinical development. But the acquisition of and the access to large volumes of patient data by customers and competitors simply hasn't translated into better outcomes for clinical trials, particularly in key areas such as patient enrollment, patient participation and the overall speed of trials.

We believe that the transformation from data to actionable insights, and ultimately, improved outcomes will happen in phases through targeted analytics and an established infrastructure of receptive and supported sites and informed patients. We believe data will be democratized and ubiquitous so that the cost of acquiring data will not be a barrier to insights and opportunities. And we believe in leveraging analytics capabilities through people who truly understand the need for efficient drug development and patient centricity will drive this transformation.

ICON has a tripartite approach to its patient recruitment strategy. First, technology and data analytics. Our OneSearch platform helps us analyze a variety of key performance data to identify the right sites for the trial. In addition, our partnerships with a number of leading-edge organizations, including TriNetX, EHR4CR and Practice Fusion, allows us to access EMR data and real-world evidence to provide key feasibility and study information and help identify relevant patients in a confidential manner, which improves the flow of patients to sites conducting in our trials.

These sites form the basis of the second component of our strategy, which is our integrated PMG site network. These sites have dedicated trial support teams and their own patient databases to help maximize recruitment. Engaging and educating patients at our sites on the benefits of clinical trials as a care option is also a key part of the sites' role in achieving outcomes which are meaningful to patients.

Finally, our patient-recruitment services team uses traditional and nontraditional digital methods, such as advertising and social media to attract patients to our trials. We believe that as we scale up this integrated approach to the patient recruitment problem, we will materially impact the speed at which we can conduct our trials.

Over the past year, we have seen a 60% increase in the number of patients recruited into ICON trials within our PMG site network, and we expect such growth to continue. We are also starting to see progress in the reduction of site start-up times and improved data quality across our network.

ICON's partnership focus also encompasses mobile or wearable technology, enabling the move towards more virtual trial capabilities that enhance the patient's experience and help to build trusted relationships throughout the study and beyond. The use of wearables and mobile technology during clinical trials and observational studies helps patients to manage their medication and collect data by a structured instrument in a manner which is more convenient to the patient. The use of mobile health technology has been shown to improve the overall quality, consistency and availability of data between site visits, enabling greater compliance to trial protocol and leading to better overall outcomes.

Mobile health blends seamlessly into the goals of ICON's data strategy to provide patient-centric convenience and care that can improve patient retention while giving access to large volumes of real-world data for the trial. We aim to be a differentiator in the way in which we collect patient data and are partnered with Intel on their pharma analytics platform to enable continuous data capture from any wearable device or sensor.

This platform has already proven its use on over a dozen clinical trials, comprising more than 1.2 million hours of data collection and over a 1,000 patients. It should be noted, however, the collection of data is only one component of the mobile health data strategy. The data also needs to be visualized, analyzed and shared appropriately to maximize its value and provide actionable insights.

With such a wealth of data available to ICON, it's processing must be efficient and provide intelligent analytics that enable better decision making throughout the life cycle of the trial and beyond. The ICONIK informatics hub plays a key role in ICON's data strategy by integrating a diverse range of technologies and data resources within one all-encompassing platform that includes EDC, Central Labs, ADDPLAN, goBalto, FIRECREST, OMR and eCOA.

ICONIK leverages clinical research, patient community and real-world data to enhance the development of new treatments from initial design of the protocol to delivery of the trial. The ICONIK informatics have also -- enables the delivery of feature-rich analytics capabilities and through leveraging our relationships with industry leaders, such as Saama Technologies, we continue to develop new ways in which we can enhance our engagement with patients and investigators as well as unlocking the real value of real-world data to provide deeper insights into how products can achieve better commercialization outcomes. We firmly believe that this integrated approach to data, in union with our patient recruitment strategy, provides ICON with a clear and differentiated position on the key industry challenge of reducing development times.

Before moving to Q&A, I'd like to thank the entire ICON team for all their hard work and commitment during the quarter. Thank you, everyone, and we're now ready for questions.

(Operator Instructions) Our first question comes from Robert Jones from Goldman Sachs.

I guess, just looking at the quarter across the industry, only a few CROs have actually reported so far so don't want to make too much of this, but there has been some divergence in what we've observed in bookings growth in book-to-bills. So I was just wondering, Steve, if maybe you could share what you're seeing as far as RFP flow? How the win rate for icon has been performing relative to your expectations?

Sure. Bob, thanks for the question. I think as we said, backlog's up 10% year-on-year. We're certainly seeing positive progress from an RFP point of view. On a year-on-year basis, we're up in the low double digits. But from an RFP, we've seen net new business, again on a year-to-date basis, be up at about 8%. So we're seeing a positive business environment, and I think at our trailing 12-month book-to-bill at 1.28 I think that's the way to measure these things. I think you can get some clearly volatile quarterly book-to-bills, but we've been fairly consistent at across that level, and across the last 12 months, we're at about 1.28. So I think all the indicators across the industry and the various segments that we work in -- large pharma, midsize and biotechs -- are positive. And we're seeing the benefit of that in terms of our win rate hasn't changed dramatically, but it's at a good level. I'm pleased with that level. And I think given the opportunities we're seeing, we'll be able to continue that.

No, that's great. And Brendan, if I could just ask one follow-up on gross margins. On a 605 basis, looks like you kind of expanded margins pretty nicely in the quarter. I know you guys have been talking about, thinking about gross margins in the near and long term more on a flattish basis. So just curious what drove the expansion in the quarter and anything in there that we should we think about as being more sustainable as we look forward.

Thanks, Bob. Yes, we were happy with our -- what our margin progression during the space of the quarter. Really, what you're seeing there, really, in terms of quarter-on-quarter progression was just good utilization of our existing headcount and resource. Again, I suppose we look at our headcount resource on a year-over-year basis, Bob. So we're kind of thinking in the mid-single digits for year-over-year growth, and we continue to think that'll be the way it'll be as we pan out into the fourth quarter. But yes, we were happy with our progression from Q2 to Q3 from a gross margin. And as I said, that was really around utilization. I think probably the level that we saw in Q3 is not a bad way to think about it for Q4, though, I think it's certainly something that we can sustain as we go in to that quarter. As we look out further, I think we've indicated that possibly down in a little bit from where we are at the moment but, certainly as we go into Q4, sustainable at the Q3 level.

Our next question comes from Ross Muken from Evercore ISI.

So maybe just in terms of the mix of business and momentum in some of the different segments, how would you kind of characterize -- I don't know if you want to base it on size or disease state, kind of what areas right now are the most active? And we saw some details recently just on the number of like CAR-T and gene therapy drugs coming through the pipeline and those obviously have pretty unique characteristics. So just give us a feel maybe in some of the areas, whether it's immunotherapy or gene therapy or CAR-T, where you're sort of seeing activity and how it's differing amongst the different types of players, I guess.

Sure, Ross, it's Steve. So I think as we see the opportunities come through the clock, obviously, oncology is going through something of a revolution scientifically with the PD-1 inhibitors and the whole immuno-oncology sort of, as I said, processes and new drugs coming through, a lot of trials going on in that segment, a lot of combination trials are starting up in those segment -- in that segment. And those are really coming out of the variety of companies. Obviously, some of the larger companies are during those, but many of the biotechs, of course, are focusing on the oncology space. CAR-T is a good example of where we've had some early success in terms of winning work there, getting some experience. These are very complicated logistical trials where you're taking a sample of blood from the patients, taking it away to a manufacturing facility, and then bringing it back and reintroducing it to the patient. So you can imagine the logistics around that. And we've been able to secure a number of wins in that area through experience that we've gained early on in that space, and that's been helpful. As I said, the biotech environment continues to be strongly funded. There are many opportunities coming out of that area. We've been successful in that space. And also, in terms of opportunities around partnerships, we're seeing a number of those playing out in the space with other numbers, but these take a while to secure, and then, of course, a while to ramp up. But there are a number of opportunities in various stages, particularly in the late stage area at the moment in terms of partnerships. So as I said, overall, the oncology -- basically strong vaccines, of course, is strong for us as well. But there's plenty of opportunity out there.

Very helpful color. And maybe just in terms of sort of the M&A or tuck-in environment, obviously, it's a good time to not have a lot of leverage on your balance sheet, although, I'm sure you'd like to be a little bit more active than you've been. I know you've been -- intend to do some tuck-in deals. It's been a bit now since we've seen anything of consequence. How would you kind of characterize that environment and then sort of your ability to execute on something in the medium-term?

Yes, I mean, we have been active in that space, as you know. We've been looking hard at a number of assets, really across the year. We haven't executed anything certainly in 2018 so far, although as I said, there's a few things that we're looking at. I would say at the -- certainly up till now with the recent volatility around the share market and the environment, a number of the assets we've looked at have been at shall I say very full valuations. Perhaps that's an understatement. And so we are pretty disciplined in terms of how we want to deploy that capital. We've got a very strong balance sheet. We recognize there's an opportunity there and an advantage there. We want to make sure we deploy that capital in a way that, obviously, adds clear value, but we're not going to pay crazy prices for assets that we'd like. We will certainly be aggressive when we find an asset that we particularly want and that really fits our strategy and tucks in to our organization well in a way that really adds really strong value. But as you say, given the volatile environment in rising interest rates, certainly in the United States, and very frothy valuations, we've maintained our discipline, and I'm happy that we've maintained our discipline. It means we do have some capital deploy, and we'll be looking to actively do that as we go forward. But we're going to be -- we're going to maintain that discipline as we consider those opportunities.

(Operator Instructions) Our next question comes from Donald Hooker from KeyBanc.

So question on Pfizer. I think in recent quarters, you guys have talked about that relationship stabilizing in terms of the revenues it represents for you. I guess, it was about 13% of revenues in the quarter, if I heard you correctly, but it's a little under 10% of backlog. So does that sort of just that maybe there's another leg down as they go into 2019? And I guess, with respect to maybe just when you're talking about Pfizer in your thoughts there, any kind of thoughts around the change in leadership there and how that might affect how you guys interact with them? Is there any update there?

Don, I might take the first part of your question, which was around the percentages, and I'll let Steve maybe talk more strategically around the further leadership piece. Yes, you're correct. It was 13% in the quarter. We're just under 10% now as a proportion of backlog. I suppose we're in the happy situation that it's not really because Pfizer's backlog is decreased. It's because the rest of the backlog is increased at that percentage, has decreased in absolute terms. So we still see that going forward being in that ballpark of 10% to 12% of our business, and that's probably not hugely off in terms of dollar amounts what we're seeing coming from that business on a quarterly basis at the moment. So still good strong customer, certainly, and will certainly be in our top customers for the foreseeable future.

And in terms of the change of management, Donald, with Albert Bourla taking over from Ian Read as the CEO, we're looking at that and monitoring that. We obviously have a number of key contacts within Pfizer. We're talking to them about any change in strategy. They've obviously recently announced some layoffs that, that could be opportunity for us. On the other hand, we're looking to continue to work with them to make sure we can -- as I said, listen to them, understand what their needs are and continue to deliver for them. So we're developing and continue to develop our relationships right across the Pfizer group but particularly with Albert and his new team coming in. We have our MSA in place. It's in place till the middle of next year. We don't have an opportunity to extend. We're just starting to get into those discussions. It's very, very early in that area. So again, we'll be meeting with our Pfizer colleagues over the next few months to start that sort of discussion. But we feel we're in a good place. We feel the relationship's very positive. And we think the new -- the change in management will not have a material impact on that.

Great. And I guess, speaking on the strategic relationships that you have and your emphasis there, I don't know if this is possible. But is there a way to think about kind of a percent of your revenues that are now kind of through these relationships versus kind of one-off kind of study opportunities? Is there a percent to kind of think about?

We don't tend to think of it in those terms, Donald. We have some internal targets with a number of partnerships and alliances we want to develop, obviously, from our point of view. It's helpful to have a good proportion of your sales and revenue coming in through these partnerships and to be able to drive the sorts of efficiencies we've been able to do over the last few years with the -- and the relationships, of course, with these partnerships and then investment and deploy capital, et cetera, in relation to improving the efficiency with those -- so but we don't have any particular target. We don't specifically disclose those sort of numbers, anyway.

Our next question comes from Tycho Peterson from JPMorgan.

So Asia has been a big source of funding for U.S. Biotech, and I think the Trump administration recently issued new rules that will heighten the scrutiny of foreign investment in U.S. biotech sector. I'm just wondering, have you guys picked up any sort of customer thinking on how this might impact the funding environment recent activities in U.S. biotech going forward?

Sorry, could you repeat the -- just -- I think we caught that it was Asia specifically you're asking about and maybe new rules from government in China. Was that correct at how that's...

No, new rules proposed by the Trump administration that will heighten scrutiny of foreign investment in the U.S. biotech sector as it's considered technologically sensitive. Does any early customer conversations you've picked up on how might -- how this might impact the funding of all of the U.S. biotech given that it's Asia specific not the funding of U.S. biotech.

No, we haven't -- I haven't had any sort of communication from customers around any issues or anything related to trade wars or anything like that, that might impact the funding of U.S. biotechs. We've seen that funding be on a very good footing for really the last -- sort of the last couple of years now. And then, of course, it may not continue. Who knows how that will go? But at the moment, it's been good. It's been strong. And we've seen no impact from China. I mean, China represents, however, a significant opportunity for us. We've grown there around about 50% over the last year or so because of what -- some of the new regulatory changes that the China CFDA are making. So I don't know how much of that relates to your question, but there's certainly some opportunity out there, and we're seeing significant opportunities for growth. And we've been recruiting very actively out there in China to do more trials.

Our next question comes from Jack Meehan from Barclays.

I was wondering if you could just provide an update in terms of the life cycle of your backlog as it pertains to revenue burn and whether you think -- we've started to reach a point where that can stabilize as far as the outlooks for going into next year might -- it might concern that?

Thanks, Jack. Yes, obviously, it was down a little bit this quarter. We've set out a target of keeping it in the mid 9s. We are very active on really ramping a lot of new business that's come into the organization over the last year to 18 months, really. And our particular focus there on study start-up, on patient recruitment. I think Steve has been outlining a lot of it in some of his comments but also at our September 10 day. So we are very active in really focusing on how we can really mobilize sites and move our patient recruitment that much faster and be able to get that metric back into the 9s. It's not easy. There is a huge amount of our work now that is oncology. These are longer trials. They do take longer to recruit patients. So it's a tough task and a tough ask, but we're using both our technology and our site-specific organization to really help us on that front. Steve, do you want...

Yes, I think just to add to that, Jack, we do analyze this part of our business pretty active -- and really it's -- we've talked about those oncology trials. There's complex oncology trials that are impacting. And I do believe that is the case. That's still the case from a historical basis. But we've also got other areas of our business that we're working on as well. And then really if I look at the clinical group, overall -- actually, they've been pretty stable in terms of -- and we've been -- and even starting to sort of move up a little bit in terms of burn of backlog. So the large part of our group, although it's lower than our average, is starting to move in the right direction despite the overall dropping just slightly quarter-to-quarter. And some of the functional services business is stuff which is sort of pushing down a little bit and even in data management. So as we analyze that across our business, we're looking at the strategies score for improving it. But I was pleased to see that overall in the clinical side of things, it is pretty stable and even starting to perhaps track up slightly to get back towards where we'd like to be as Brendan said in the mid 9s.

Great. And as a follow-up to that as you think about ramping up some of this work, wanted to get your updated thoughts on hiring, and it seems like the entire industry has been doing pretty well in terms of the funding environment. So are you finding it -- the hiring environment's getting tighter at all? And what are the plans for you to ramp up some of the resources there?

Yes, I mean -- I think I'd be -- it would certainly -- there were certainly pockets in the organization and across the globe that is very challenging around hiring. I mentioned China. That's certainly been challenging to get experienced people in that part of the world. Japan, has also been challenging to get experienced people. We've gone to new graduate programs there. We're making investments -- very significant investments in new graduates in both those parts of the world to bring in new people. The U.S. has it's hot spots as well. CRAs are challenging. Project managers, particularly with oncology experience can be challenging to find. However, we have a pretty strong talent acquisition group within the organization, and we match our resource needs with our business pretty effectively. Over the summer -- I guess, the summer quarter, perhaps the headcount growth wasn't quite as much as we expected, but then you probably think that's the case with, as I say, through August -- July and August, we're ramping that up now, and certainly over the last month, we've seen some significant increases in our headcount where we needed to do the work that we need to get done. So overall, I'm happy with it. We obviously having to pay market salaries, and we're doing that. Our retention remains very strong at well over 85% of our organization. So while we need to recruit people, we're also retaining people. And particularly in our CRA ranks and our project manager ranks as well, which I'm, again, very pleased to see that we're able to keep those people in. So overall, there are hot spots. There are challenges. But I think we're doing pretty well.

Our next question comes from John Kreger from William Blair.

Steve, can you just talk a little bit more about the PMG site management business. Just curious how clients are -- how receptive they are to this type of offering? Maybe if you think about the study starts in the last quarter or 2. What percentage of those would you say are using the site management infrastructure that you've built?

John, I don't have that figure on top of mind, but I would say that we are, obviously, effectively promoting our PMG site network for all of our new full-service studies. However, as I've indicated previously, this is -- it's not a network that we do much oncology win at the moment. Although -- of course, the new DuPage part of that network will be, and we are ramping oncology and that way so. So it's -- up till now, most of the work that the PMG network have done, and as I said, it's increasing very nicely, has been in the non-oncology space. But with DuPage coming onboard, we will be ramping up the oncology space -- oncology capabilities there. So I would expect that all of our new work in oncology would have representation within the DuPage network -- the DuPage part of the PMG. I think across the board, I would say, around 60%, 65% of our work -- so the vast majority of our nononcology work goes into the PMG network. There are projects sometimes that they say they aren't able to do -- they can't participate in. That tends to be the exception rather than the rule because, whatever they do, it tends to be at least as good and usually significantly better than the ad hoc sites that we use. So as a segue, we're significantly ramping up the number of patients that we've recruited this year into our PMG network. It's really materially now contributing to our patient recruitment in the U.S. of course. And so we find our customers are increasingly interested in and satisfied by what they can get and what they can receive, and it's not just in terms of patient recruitment rates, which are significantly better, but it's in terms of start-up times, which there's of course always -- as always more work to do. But we're already faster -- much faster in starting up sites -- starting up our PMG sites and the ad hoc sites. We do believe, however, there's more opportunity we've improved on that as well. We have some fairly ambitious targets as we go into next year for not just patients recruited but the start-up times as well. So overall, I'm really pleased with the way that's really gaining some traction now with our customers, and of course, within our own organization.

Great. And maybe a follow-up on your data remarks from earlier in the call. I thought it was interesting how you said there's plenty of data out there but figuring out how to use it and translate into trial performance is the hard part. Where do you stand on that front? Do you feel like you're delivering meaningfully better metrics at this point around things like patient enrollment? Or is that something that you're hoping to achieve in the next year or 2?

I think -- to be very honest, I think we're making progress in that, but we're not -- we're clear we're not where we want to be. And I think what I was trying to outline in our strategies, we have those insights and we have, through our partners particularly in that access to data, but it's applying those insights into a receptive site network that is where we're differentiating. And we're starting, as I've indicated in terms of the numbers of patients that are increasing through our -- coming through our network, we're starting to see some real traction there. So I believe we're making progress in that area. I do as I say, think there's still some way to go there. But the access to data that we've had, and particularly, the insights that we're getting from those partners is allowing us to make sure we're getting the right sites and that we're finding where the patients are and, ultimately, we're moving those patients and giving those patients an opportunity to get to those sites to participate in the trial. So those insights -- and they vary across the various therapeutic areas that we work in. But those insights are already starting to come through, and we're starting to be able to apply. There's a lot of learning going on here as we roll this strategy out and as we do it. Every trial we do, we learn something from, and we can then apply it to the next one. And that's really the obligation we provide there, right. That's the expectation that our customers have as well as we move this forward.

Our next question comes from Juan Avendano from Bank of America.

Can you give us an update on the European pharma R&D spend in our sourcing environment? And what are your thoughts on any potential impact to ICON from Brexit given your exposure there?

Yes, we're -- we don't tend to look at European pharma as a separate segment, Juan. We tend to look at -- pharma is a global industry. So I don't know that I see significant differences between the segments. I mean, we have the global numbers. We talk about the business is a global business. Perhaps -- we see probably more decisions made within the United States, I would say. But in terms of spend in company, I mean, all these -- all of the major companies work in Europe and have decision makers in Europe and in the United States. And so we don't really differentiate the numbers across the European pharma segment and the U.S. segment. In terms of Brexit, we've looked at our business. We don't see any material change in terms of number of clinical trials being set up or being started over the last 12 months in the United Kingdom. So that would seem to indicate that our customers don't see Brexit as being a major issue. I understand through our Association of CRO, ACRO, that's also the case with 1 or 2 of our competitors. They haven't seen much too much of a decrease in the number of sites being set up for trials. So that would seem to indicate that Brexit's not a major -- now I'd say that our business is very much focused on services and clinical. Others go to moving samples in and out. We do that a little bit, but it's less of an impact on our business than perhaps it might be on others. So the movement -- the shipping of movement of drugs and samples may be impacted. But at the end of the day, if the U.K. steps out of the European Union, and even if it crashes out of the European Union, it would become another country, which would -- within which we'd do a clinical trial. Not every country we work with is in the European Union. We go to many different countries, and we work with them as they demand and as their regulations require. So we don't see Brexit being a major issue for us.

Got it. Appreciate the color there. And a follow-up on PMG and your site network strategy. I'm just wondering how scalable that strategy could be and what are your plans for expanding your site network internationally at this point?

Well, we've been fairly upfront about that from the start. We've been actively looking for to expand that site network. We've -- DuPage is the network we've broadened into in the United States, and that's actively being brought into the family -- into the fold now. We would like to expand it to Europe, and we see on the horizon some opportunities. We've had 1 or 2 false starts in terms of getting quality assets in Europe. But we see there's some further opportunities on the horizon, and we would like to think that we'll be able to make some progress on that in the next 12 months or so. The challenge with this business is there aren't very many sort of large site management organizations that are available. And so you end up starting with a fairly small one, and then you have to grow it organically. That's certainly an option. The larger ones are at a premium and, as I say, not as available. But we continue to be vigilant. Our plan is to expand to Europe, and that would be particularly Western Europe. So Western Europe and North America would be really the focal point for our integrated site network going forward, and we won -- we are looking to make progress on that actively over the next 12 to 18 months or so.

Our next question comes from David Windley from Jefferies.

On the cost side, I wondered on 2 fronts, first of all, apologize if the question's already been asked. But labor environment and kind of cost of service-related -- direct cost-related trends and opportunities or pressures. And then on the SG&A line, there's -- you've talked, and I think been among the first to talk about robotic process automation and some things like that, that could help your efficiency -- further extend your efficiency in your regional service centers and wondered what -- is it still very, very early on that. Or you are actually implementing some things live via from an RPA standpoint.

Yes, Dave, let me take the SG&A and the RPA, and then maybe Brendan might comment on the labor service costs. You're certainly right. From an RPA point of view, robotics, we are pursuing a number of pilots in a number of areas around the eTMF, some of our IT systems here, help this carries information sort of gathering and providing. But it is pretty early, and I'd have to say, in terms of a reduction in our SG&A till now, it's been very little -- any of it has been due to our robotics. That is a program that is starting to roll out, as I say, piloting

(technical difficulty)

we do believe we can roll a couple of those programs, particularly obviously, the successful pilots out over the next 6 months or so. And so by middle of next year, I think we'll have better answer for you in terms of the impact that's having on our SG&A. We do believe it can be significant. We still -- as you probably are aware, have some opportunity and are moving through the offshoring side of things, and we're making progress on that, and we continue to make progress on it. So if I look at our SG&A, although it's come down nicely in terms of our proportion of revenue, actually, we are spending a little bit more. We're spending about 5% more in our SG&A year-on-year. So it's going up but obviously less than our revenue growth. And our headcount's gone up, I think, in the vicinity of 100 people in that -- mainly in an offshore space. So we've been able to leverage that model -- the offshore model pretty well, thanks to the capabilities of our GBS -- particularly, our GBS team and the management that they have in place. And so we've got more to go in that area -- I think, some more opportunity in that area. But our robotics is, as I said, at a fairly early stage, and I believe that, in 12 months' time, we'll probably have a better answer for in terms of that contribution. Do we want to talk about what labor is?

Yes, I think as you know Dave, in terms of a direct cost, the vast majority of the costs that are included there are actually salary costs. There's not a whole heap of other costs in there. There's a little bit of lab, but it's predominantly salary cost. So it is -- and as I think Steve alluded to earlier in the conversation, we do have pockets where we're looking to see -- it's a little more demand in the marketplace like North America, like China. And we're paying competitively in the marketplaces to ensure that we're getting the people in. So there is very -- you've got some hot spots in terms of salary inflation in those particular markets. But then we're also balancing that, as we said, we're on the onshore/offshore methodology even in our direct call space, and also, there are parts of the world where Western Europe's still at the epicenter of inflationary levels as maybe we see in North America and APAC. So overall, we try to balance it as much as possible. We try to keep that our annual basis, and again, we do look at these things from an annual basis in that 3% to 3.5% range. And that will be -- as we go forward, that will be our focus as well.

Our next question comes from Sandy Draper from SunTrust.

Most of my questions have been asked and answered. But maybe wanted to just follow up on the comments you made in the prepared remarks about some of the -- I'm not sure if it's digital marketing per patient or sort of going directly to patients. I'm just trying to get a sense of how you think about the relative impact of trying to market directly to patients for clinical trials versus getting leverage really working on sites and then getting them to then go out, and you market to one site. They're good. They're going to touch a lot of patients just the cost benefit, and I may have understood that you're not actually going directly to the patients. But just any further details on that, and is this like a Facebook digital marketing strategy there or just any comments would be great.

Sure, Sandy. I think we've been going directly to patients in terms of marketing clinical trials for many years now. Television, advertising, radio advertising, internet has been a relatively common part of how we've made patients aware of trials and then directed them to sites. So whether it would be the site doing that, and we've often provide them with a budget to do it or whether our patient recruitment group, which I mentioned is the third part of our patient recruitment strategy. We have an in-house group who can use the various media to make patients aware, if you like, of clinical trials, particularly -- and that's much more effective, of course, when we know where those patients are having used some of the technology through TriNetX, et cetera, to find out where these patients exist, in what parts of the -- particularly the United States those patients exist. So we can then target that advertising to those sorts of areas and drive them to the sites that we have in that area. So that's been a pretty common -- I think as we move into more the social media, that's probably the newer area, and that's an area where we're starting to become active in. We have some people who have some expertise in that through our patient recruitment group. And so the digital marketing directly to patients, again, through -- for our traditional trials is starting to happen there. What I was also kind of alluding to is that as we move forward, the virtual trial where we have less involvement from a site and there's more direct to the patient in terms of providing them with medication, doing electronic informed consent or else -- we do see that as an important component of the evolution of clinical trials going forward. But predominantly, I think in the late-phase area

(technical difficulty)

but increasingly as we go forward, we'll have, I think, opportunities to deploy those types of trials in the Phase

(technical difficulty)

there'll always be some involvement with a site. I'm not suggesting for a minute that we'll be doing trials directly to patients without the involvement of investigator or site. But what we're looking to help to do is as we make trials more patient centric, to make the burden for them to be in a clinical trial to much less than it has been up till now, where they've had to visit perhaps every month perhaps even more often over a 6, 12 month period. If we could take that down to maybe a visit at the start, visit in the middle and a visit in the end and connect with them through, as I say, social media and through the various apps and wearables that they have, it would make the whole trial experience, I think, much less of a burden, and we believe would make patients much more amenable to being part of a clinical trial. And we can all relate to the busy lives that we lead, and to be part of a clinical trial is a burden. Many people I think need to work and very difficult. We're trying to make that burden less, and we'll be leading the way in doing that.

Our next question comes from Daniel Brennan from UBS.

Just a high-level question on -- there's news out today regarding renewed efforts by the administration to bring down drug prices. I'm just wondering regardless if the administration will have any success, do risks around drug pricing show up during discussions with sponsors today. And how should we think about the potential impact on R&D activity going forward?

Drug pricing doesn't come up in our discussions with customers, Daniel. It's something that we -- that they obviously deal with, with the payers and with the regulators and the governments particularly, not just in terms of the United States but particularly in Europe and other parts of the world. However, having said that, we do recognize that, when our customers are under pricing pressure, it means that, that they have to be more careful with the way they spend their money. And so it behooves us to be more efficient and more effective with the development dollars that they spend. And so that's -- while we -- as I say, we don't get directly involved in drug pricing discussions -- we do feel their pain, and so we do recognize that we have to be more effective in the way we spend our money, and we provide value to them. So that's, as I say, I guess the indirect impact of drug pricing for us.

And then maybe just a follow-up just on trends in the market or even with your results this quarter regarding full-service versus FSP kind of. Can you just give us an update on what trends look like in the quarter? And just remind us how you think this mix of ours going forward? And how does that impact ICON?

I think the question was particularly around the FSP versus full-service model.

Exactly, yes.

We still see strong demand for both, to be honest. We've seen, over the last -- I say for the last 1.5 years, we've seen a good, really strong market for our full service work. It's been where we've been doing really well. And the last little while, I think the FSP market has shown that it waxes and wanes a little bit, that marketplace. But it's still a very valid part of our marketplace. I think we put it, in growth rate terms, in a similar position to the full service. So we see 6% to 7% annual growth rates in our full-service business but not a dissimilar number in our functional service offering. And it is definitely a very valid part of the marketplace for being around $4 billion of the total market. So I think both are still -- our customers are certainly both interested in both of those service lines, and we continue to see good demand. So yes, very happy with the mix that we're seeing there at the moment, though.

I think Daniel, just to add to Brendan's comments, as we get engaged with partners and with customers around new partnership opportunities, our ability to bring different solutions in terms of an FSP solution and a full-service solution, and -- I know some of my guys hide it, but a hybrid -- a more hybrid model is also, I think, an advantage that we have because we've been used to working in that model, and we do deploy that sort of model for customers as well. So it's -- I think we see opportunities in the marketplace, not just around pure FSP or pure full service, but this more integrated hybrid model is increasingly something that customers want to talk to us about.

It appears there are no further questions in the queue. Mr. Steve Cutler, I'd like to hand the call back over to you.

Thank you, everyone, for listening in today. We're very pleased with our performance in quarter 3, and we're looking forward to working hard for the rest of the year as we continue to enhance our position as the CRO trusted partner of choice in drug development. Thank you, everyone.

This concludes today's conference call. Thank you for your participation. You may now disconnect.