Exact Sciences Corp (EXAS) 2011 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Exact Sciences Third Quarter 2011 Earnings Conference Call. (Operator Instructions) As a reminder, this program is being recorded. I would now like to introduce your host for today's program, Ms. Cara Tucker from Exact Sciences. Please go ahead, ma'am.

Thank you. Thank you for joining us for Exact Sciences Third Quarter 2011 Conference Call. On the call today are Kevin Conroy, the Company's President and Chief Executive Officer; and Maneesh Arora, the Chief Financial Officer.

Exact Sciences issued a news release earlier this morning detailing our third quarter 2011 financial results. If you have not seen it, please go to our website at ExactSciences.com or call 614-302-5622, and it will be provided to you.

Following the Safe Harbor statement, Maneesh will provide a summary of our third quarter financial results. Next, Kevin will provide an update of our clinical trial.

Before we get underway, I would ask everyone to take note of the Safe Harbor paragraph that appears at the end of the news release issued this morning covering the Company's financial results. This paragraph states that any forward-looking statements that we make, one, speak only as of the date made; two, are subject to inherent risks and uncertainties including those described in our most recently filed annual report on Form 10-K and our subsequently filed quarterly reports on Form 10-Q; and three, should not be unduly relied upon.

Except as otherwise required by the federal securities laws we disclaim any obligation or undertaking to publicly release any updates or revisions to any forward-looking statement contained herein or elsewhere to reflect any change in our expectations with regard thereto to any change in events, conditions, or circumstances on which any such statement is based.

It is my pleasure now to introduce our Chief Financial Officer, Maneesh Arora.

Thank you, Cara, and good morning, everyone. Our top priority remains ensuring the success of the clinical trial for our Cologuard product both financially and operationally. During the third quarter, we continued to invest in the success of the trial while closely managing our cash.

We closed the quarter with cash, cash equivalents and marketable securities of $75.4 million. We remain on track with our 2011 cash utilization guidance of $29 million. We expect to have a cash balance of approximately $66 million at the end of this year.

We are on track with our clinical trial enrollment and, importantly, we have pulled forward our automation program without a change to our cash guidance. We expect to complete work on our automation solution in conjunction with our FDA clinical trial submission.

It is now my pleasure to introduce Exact's President and CEO, Kevin Conroy.

Thanks, Maneesh. Let's review our 2011 priorities before discussing the clinical trial. This year we have focused on three priorities -- product development, clinical trial enrollment, and market development. The team at Exact continues to make strong progress on each of these priorities. Both product development and clinical trial enrollment remain on track. We'll talk about both in more detail in a moment.

Our market development efforts are also on track. Our efforts to reach GI opinion leaders and other key constituencies including primary care physicians, payors, and patient advocacy groups have been very successful. We continue to hear that there is a major unmet need for Cologuard and a tremendous amount of interest in our product and clinical trial.

Our pharmacoeconomic study has begun. This study will measure the cost-effectiveness of Cologuard. The study will help support our pricing upon launch. We believe we can demonstrate significant cost-effectiveness in the use of Cologuard in screening for colon cancer. We would expect publication of the study to occur in about 12 to 18 months.

The publication of scientific data is another important part of our market development efforts. You can expect to see a scientific paper published in the near term that compares Cologuard head-to-head with Septin 9 and another peer review paper that details the data from our November 2010 validation study.

We also anticipate providing additional validation study data by the end of the year. All of these efforts are focused on maximizing uptake of Cologuard after its approval and launch.

Let's turn to our product development update.

We continue to be pleased with the strong performance of Cologuard during our internal product testing. Our FIT or fecal immunochemical test assay has shown 99% analytical sensitivity and 100% analytical specificity in internal testing.

We presented data at an American Association of Cancer Research conference in October that demonstrated that our K-ras panel is 100 times more sensitive than sequencing in evaluating DNA in colorectal cancer adenoma in normal tissue. Our methylation markers continue to perform as or better than expected.

Our automation and integration program is on track for completion prior to the end of clinical testing in 2012. Our instrument is designed to be a user-friendly, high throughput solution. We will seek FDA approval for the instrument in parallel with the submission for the test.

Now let's turn to our clinical trial.

We are pleased to report that we are on or ahead of schedule in our clinical trial patient enrollment. We expect to enroll 10,000 or more patients in the study through approximately 60 sites. The enrollment sites include high-volume colonoscopy screening clinics and primary care clinics that refer patients to colonoscopy centers.

Our enrollment goal is centered on the number of patients in which cancer is diagnosed during the colonoscopy procedure, and we anticipate enrolling a minimum of 56 cancer patients and between 400 and 500 precancer patients. The number of cancer patients drives the total number of patients that we need to be enrolled and is dependent on the incidence rate within the study population. We are seeing the expected rate of cancer incidence in the patients we've enrolled in the trial, to date.

It's important to note, as well, that we have powered our study to at least 90%. So at 56 cancers, the study is powered to 90% success. We anticipate that enrollment will be completed in 9 to 12 months.

This map illustrates our enrollment sites in the DeeP-C study. The DeeP-C study is one of the largest prospective colorectal cancer screening studies ever conducted. We now have 49 sites across the country enrolling patients. We anticipate reaching our goal of 60 sites by the end of the year. We are pleased to be on or ahead of our patient enrollment goal.

I want to compliment our extended clinical affairs and operations team for the great job they have done in initiating the study.

Let's turn now to the test administered to every clinical trial patient.

Each patient receives three tests -- our Cologuard test, which includes a proprietary fecal immunochemical test, or FIT test; a commercially available FDA-cleared FIT test and colonoscopy. These tests support our primary and secondary endpoints, and the inclusion of the comparator FIT test is unique in the studies that other companies are looking at for the same type of product.

Meeting our endpoints for the DeeP-C study will provide two things -- first, the data we need for a successful PMA submission and the data to successfully market the product. The primary endpoint of the study is the sensitivity and specificity of Cologuard for colorectal cancer detection compared to colonoscopy.

A secondary endpoint of the study is the sensitivity and specificity of Cologuard for advanced adenoma detection compared to colonoscopy -- that's precancer.

Another secondary endpoint is comparing Cologuard to FIT. So the non-inferiority of Cologuard for the detection of cancer and the superiority of Cologuard for the detection of precancers.

The importance of including FIT -- a commercially available FIT tests in our study can't be overstated. While there are more than 10 million FIT tests performed every year, FIT detects only 65% of cancers and 20% of cancers based on the literature. We believe that despite the wide use of FIT, because the test is not very sensitive, it has minimal impact on colorectal cancer mortality rates and does not meaningfully reduce the incidence of colorectal cancer.

Our goal is to meet both the endpoints for cancer and precancer sensitivity compared to the FIT test. Our collaborators, in a review of the scientific literature, tell us that a clinical trial is one of the best studies of FIT ever conducted. It will generate head-to-head data that will be published in a peer review journal.

Let's talk about the Company's key milestones.

Our milestones include preparing for the clinical study, commencing clinical trial enrollment -- as we've noted, we've completed both of those goals; making an FDA submission for 2012; and launching the product after FDA approval. We're on track across the board, and we continue to expect to present additional validation study this year. We are running those samples now and will announce the forum for presenting the data in the near term.

In conclusion, we are on track in 2011. The product continues to perform well, the clinical trial has gotten off to a tremendous start. The trial will deliver the data for the FDA submission and for potential marketing of the product. Market development continues so that we'll be in a position upon launch to execute on this significant opportunity. And, as a final reminder, there is a major unmet need out there for detecting both early cancers and the precancers that develop into cancer. It's a significant opportunity in the market, and we continue to stay focused on this opportunity.

And, with that, I'd like to open the line for questions.

(Operator Instructions) Quintin Lai, Robert W. Baird.

With respect to the enrollment, I think on the last call you talked a little bit about the parameters of what was required to get enrollment -- no colonoscopy within the last 10 years, et cetera. It sounds like that that's going nicely. And the math would suggest that you need to catch about four or five cancers a month. So is that on pace with what you're seeing now?

The first thing that you have to see is enrollment, because the data that you initially get is the data for the sample being collected and shipped to the Company. And then the colonoscopy reports and the pathology reports follow. We are on track in terms of the cancer incidence rate if not a little bit ahead. But I want to caution that we're still dealing with a relatively small [end].

When we initially kicked off this study, we started in a measured way with a relatively small number of sites, and now we have ramped that up to 49 sites. We targeted regions of the country, as you can see, a very broad geographic dispersion of clinical trial enrollment sites where we expect to see the incidence rate that you see across the country. And we're on track there, and we're very pleased to be on track both with enrollment and to be catching the number of cancers in the screening study.

Thanks for the update, Kevin. Recently, the US Preventative Services Task Force has come out with comments about other cancer screening modalities -- breast cancer, prostate cancer, HPV -- and they seem to be honing in on specificity and false positive rate. Could you refresh us all on your views of what specificity needs to be for good colon cancer screening tests and what you're hearing back from different key opinion leaders or CMS on what that parameter should be?

Yes. The specificity of our test we expect to be around 90%, and that is based -- we targeted that number [resident] in a higher number of 95% based upon feedback from key opinion leaders.

In terms of the USPSTF, they are looking at each of these tests and the screening modalities in a distinct way. As you know, the current screening tool for prostate cancer is one that -- a recent paper, I think it was two years ago, in the New England Journal, said that you had to screen -- you basically had to ruin 50 lives to save one because the PSA test is not specific and, in many cases, it's not sensitive.

Colon cancer is a totally different situation in that you have to remember these precancers are acceptable. They can be literally snipped out. That is not the case with prostate cancer, it's not the case with breast cancer. This is much more similar to cervical cancer, which has been nearly eradicated in this country with a test that is only 50% to 70% sensitive for precancers. And, again, because cervical cancer is easily accessible, the precancers can be ablated, and the patient goes home and continues on their week.

That obviously isn't true with breast cancer spotted by mammography or prostate cancer is picked up by the PSA test. So we think that we're in a really strong position to make a meaningful impact in the incidence rate and the mortality rates over a long period of time with this test combined with colonoscopy.

Thank you for that. And then with respect to that paper that you talked about, we actually found it online, and I don't know if you can really talk to it or if maybe you're embargoed on it. But in that paper, it does call into question the specificity for a blood screening test on a general population. If you can't comment on that paper, maybe you can comment on when you were evaluating blood screening was specificity an issue for the sensitivity that you need?

The answer to that question is yes. So the paper that you're referring to is a paper -- the lead author is Dave Ahlquist from the Mayo Clinic comparing head-to-head the Cologuard or a stool DNA test -- so the same format as the test that was run in November of 2010 -- head-to-head against Epigenomics' Septin 9 test in the commercial embodiment of that test run at ARUP, a lab in Utah.

The outcome of that was that -- and this is just the top line. We are embargoed from sharing the full report, but the top line is that the stool DNA test detected 82% of large precancerous adenomas, a medium size of 2 centimeters compared to only 14% by Septin 9. Septin 9 had 74% specificity. And we think the challenge with the blood test and, remember, we looked at developing our own blood test very closely, is that you need to have a test that is highly specific. It's probably in the same range is the current fecal immunochemical test in the 97% sensitivity range.

The reason being is if you have a test that has a 25% false-positive rate and that test is performed every year, that means every four years you'll have an unnecessary colonoscopy. Even if that test -- if an annual test has a 10% false-positive rate, again, you have too many unnecessary colonoscopies, and the cost-effectiveness of that test plummets.

So this is an important thing to keep in mind as you consider blood tests. We think that a single marker blood test will always be challenged by the inherent non-specificity and minimal sensitivity. But we'll continue to look at these tests as data comes out, and we'll continue to compare our product against them.

Pete Vitale, William Blair.

It's Pete in for Brian. I just had a question -- I wanted to confirm that the upcoming validation study, the November-2011-1, is not going to include precancers, just cancers, correct?

We do not expect that it will include precancers, correct.

And could you just remind us why that is?

The real goal here, the primary endpoint of the study is cancer detection, and that's been our goal here in terms of collecting precancers. Collecting precancers is a challenge to do except in a prospective way. So it is -- there just aren't a lot of precancer samples that are available. So we thought this was the most efficient way for us to get another look at the revised product before the end of this year.

William Hite, Lazard Capital Markets.

Hi, it's Steve Unger. I just have a couple of questions. Has there been any changes to your thoughts in the last couple of months on reimbursement? Have you started to discuss the tests already with payors?

Yes, so we have discussed the tests with approximately 30 payors. Our thoughts on reimbursement haven't changed. In fact, we've had some comments that our goal of $300 a test average reimbursement may be too low. The bottom line is there's a lot of work to be done. That work has begun, as we mentioned earlier on the call, around the pharmacoeconomic study. That will be the main driver of the value of this test. We know that payors are increasingly looking for pharmacoeconomic data, and we are approaching this in a very serious and thoughtful way in getting the input of key opinion leaders in the field of colon cancer pharmacoeconomic studies.

And as far as the existing models and so forth, they typically deal with cancer. But your test is including precancer and precancer detection. Do those models need to change? Or do you need to have separate studies done just to discuss precancer?

Steve, existing models do include both, and these models are highly complex, but they do include both cancer and precancer. And the precancer driver is an important driver of the overall cost-effectiveness of the product.

Good, okay, okay. And then as far as the clinical trial itself, you are able to, at least, as we move forward, to monitor how many cancers you are collecting? Is that correct?

Yes, we are able to see the number of cancers real time. Again, it slightly lags the sample being delivered by a few weeks, and, again, we are seeing cancers in the study, so far, and at a rate that you would expect if not slightly ahead.

Okay, and how does that relate to precancers? Are you also able to see the same sort of real-time data on precancer?

Yes, with the same lag. Since precancers are much more prevalent than cancers, there is very little question that we will be able to achieve significant powering well into the 90% -- probably approaching well north of -- maybe north of 95% power for precancers. Because you expect to have about 400 to 500 precancers in the study, Steve.

Excellent, okay, okay. And then just as far as the automation is concerned, is there any risk that -- because -- are you complete, I guess, and internally validated the automation to the point where you're confident that you'll be able to run these samples on the automation for the submission?

Well, let me remind you that the testing for all of the samples that we collect will occur at the end of the study. And so that should take about 30 days. We're targeting, let's call it, plus or minus the September 2012 time period.

And today the instrument is up and running here in our facility. It's in the beta stage of development so we continue to make slight modifications. Really, today, we're focused mostly on the software aspect of the instrument, and we have -- we are very confident that the instrument will not only be ready, but it will also help improve the results of the study because an instrument is inherently more precise than a human. That was actually an important driver for making sure that we had the instrument -- we pulled the instrument program forward.

Got it, okay. And then just lastly then -- does the FDA or some regulatory body need to sign off on the paperwork for the automation before you run the samples? Does that need to be cleared for you?

No, that's part of one submission, Steve. So you collect the data, and you submit volumes of documents that show the design and development process and quality processes all along the way of developing both the test and the instrument.

Jon Wood, Jefferies.

Hi, guys, this is Ashley. We're on the line for Jon Wood this morning. So just a couple of quick questions -- for the head-to-head study, what was the biomarker value? Was it the same one? Was it the final one with [NDRGPN PQS Vidactin]? Or was it the one that was used in the October 2010 study?

You mean in the head-to-head Septin 9 study?

Yes, exactly.

Ken, there was the same panel that was used in the November 2010 with four methylation markers, seven K-ras mutations, and a fecal blood test.

Okay, so it had (inaudible)?

Correct.

And did you also look at the precancerous detection? I know you shared some results from the cancer detection. Did you have any precancer samples?

Yes. Well, in the head-to-head study, their test detected 14% of precancers are detected 82%. For cancer we detected 87%, and they detected 60% of cancer.

And then, lastly, for your final assay, is that going to be tied to a particular or artificial instrument or will the labs be able to use any artificial instrument they have in the labs?

Well, it will come with a package insert that recommends a specific real-time PCR instrument that happens to be already FDA cleared. We haven't announced that -- who that instrument manufacturer is, but at some point we will.

Okay and, lastly, can you go into the recent results by Epigenomics that, I think, were presented in September 2011? I think they showed 95% sensitivity and 86% specificity? Do you have any comments on that?

What I would say there is the way that we like to do things is just put our products into the hands of an outsider, let them run it head-to-head against the Epigenomics product. And we saw vastly different data, both sensitivity and specificity, in the hands of Epigenomics partner laboratory. I can explain to you the significant difference, although you may consider that the representative types of cancers that are tested. We do know that with blood tests -- blood tests have a very difficult time detecting these early stage 1 cancers call T-1 cancers, and those are cancers that have yet to invade the thick muscle in the intestinal wall.

And so I guess that's one question that you could probably pose to Epigenomics. We don't see strong sensitivity with the Septin 9 test for early stage cancers.

(Operator Instructions) Trey Cobb, Stephens.

Kevin, you mentioned in your remarks that you have almost 50 of the 60 client sites qualified and running, and it sounds like enrollment is going good, so far. Are you still targeting one patient enrolled per site per day once all 60 sites are up and running? And then how should we be thinking about when you hit that run rate?

We're at about that run rate right now, and we expect to exceed that run rate. When we started the trial, we were obviously a lot lower than that, because we just started with a handful of sites. So one of the nice things is that we have hit that -- already hit the hockey-stick curve that we wanted to hit. And we hit it sooner than expected.

Great. And then maybe if you could give us an update, if there is one, on potential lab commercialization partners for Cologuard and what your most current thoughts are there?

Our goal remains to partner with commercial labs across the country to sell a KIT and automation system. And we'll be able to give more updates as we get closer to FDA approval. And that continues to be the strategy, moving forward. So Quest and LabCorp have been the large labs as well as the regional labs and even hospital labs will be able to use this instrument, which is a high-throughput instrument generating significant revenue per shift.

Bill Quirk, Piper Jaffray.

First off, I apologize, I jumped on the call late, so I apologize if you've already touched on this, Kevin, but two questions. One, if you could just give us a reminder about some additional preclinical data that's coming up. I seem to recall the original target was AMP. I'm not sure if that's still the case.

And then, secondly, as we're transitioning internally from essentially a development company and start to think about some of the early stages of commercialization. Maybe if you could just give us an update on some plans you had had or KITs around[ rather two] to look at handling the front end of sample (inaudible). Thanks.

Bill, no, I'm sorry we can't answer those questions, we already did just a few minutes ago. I'm kidding.

We'll follow up later, thanks.

No, I'm kidding, Bill. So nobody has asked the question about presenting the data at AMP, and that is our goal. We continue to collect samples, so the samples continue to come in on a daily basis and process those samples. Bill, we don't want to put the scientific validity of this study behind rushing to get to AMP. So I would say there's a high probability that on the 16th at AMP in Dallas, we would be presenting data, again, on these normals and cancers as well, I think, as an update on the automation solution. But we think that there is a high probability that we will be there.

If not, there is a scientific medical conference in Chicago in the first week of December, so one of those two dates we'll have the data. But we just want to make sure that this is done in a very scientifically valid manner.

And then your second question was about the up front processing. Bill, what we're thinking about there is to offer, as a service, a preprocessing of the container with the sample so that we can ship to the lab a tube with 2 milliliters of buffer and patient DNA. So that tube would go into a rack and into an instrument that then generates a result. It really significantly increases the ease of use, and we can have a couple of labs around the country that are very efficient, almost in an industrial processing way of getting to a -- going from a patient sample in a container to a supernatant with DNA in a tube.

We'll have the automation available for patients to do that up-front processing themselves, and this will be an option that we think will be attractive to some labs around the country -- maybe many labs around the country.

Very good. And then just one quick follow-up from that, Kevin. You mentioned that we -- you were also going to talk about the automated solution either at AMP or at this conference the first week in December. Should we read that or understand that to be that there will be some clinical or preclinical data, I should say, around the system and/or we'll actually get the chance to actually see the system, say, in the (inaudible)? Is that how I should interpret it?

No, I think we'll have a video of the system, and it's just so expensive and complicated for us to ship these beta units down to Texas. So we shot a video of the system. And we won't be presenting data from the instrument quite yet, but it is -- it's a nice development to show.

And I want to really emphasize that my hat goes off to the automation team. It's ahead of schedule and on budget in delivering this really easy-to-use system that radically simplifies the customer's job of getting a patient result.

So we're excited about that program, and we'll be pleased to be able to share that with the larger community -- the laboratory community.

Raymond Myers, Benchmark.

First I wanted to clarify the number of sites that have enrolled patients. I believe you said that 50 to 60 sites are online at this point. Have all of those sites enrolled their first patient?

Actually, we have 49 sites that have enrolled patients. And we expect to be at 60 sites by the end of the year. We also have a waiting list of sites that would like to join the trial. Presently, we don't expect to go over 60, but we do have a waiting list available.

Okay, sounds good, thanks for the clarification there. And then I want to also clarify regarding the presentation of data. No one wants us to present before you're ready, and the delay from November to December is not meaningful. However, then, that question might arise why might there be a delay in presenting validation data? Is something going slower than expected?

No, we started in September collecting samples for this study, and it's really just a question of making sure that you have a sufficient number of cancer samples. The normal samples, obviously, are a lot easier to collect than the cancer samples. And collecting the cancer samples in a short period of time is a challenge. We believe that we'll have enough samples by early next week to complete processing and analysis. But the devil is always in the details, and we expect to be there, we just are always cautious and don't want to surprise anybody in a negative way.

Sure, that's wise. Let me then further clarify, though -- it sounds like you're saying the clinical study is going on track to ahead of schedule in collecting cancer samples. However, perhaps the validation study perhaps is slightly behind your expectations. Why would there be that difference?

Well, no, we've been targeting AMP for the better part of the year, and we have a high likelihood of being at AMP. But until the data is in and analyzed by the third party biostatistician, I can't tell you with 100% certainty that we will be at AMP. And AMP is just around the corner, so I don't want to say one thing with certainty today and then tell you something else two weeks from now.

So let me reiterate -- we expect to be at AMP but, if not, we would be at a conference that we would announce in early December in Chicago.

Okay, that sounds good. And then I have one final point of clarification, if I might. You mentioned in your comments that you would be -- when you seek FDA approval for Cologuard, you would also be submitting for FDA approval for the instrument. Can you clarify -- is that a separate FDA approval? Are there two parallel approvals or is the instrument part of the overall single approval?

It's a part of the single overall approval.

Okay, great.

And I would note that this is an instrument that has gone through FDA approval for a PMA product in the past. So the OEM supplier of the instrument has the internal GMP system and all of the quality systems and design history files necessary to make that submission pretty straightforward. And I'd also again emphasize that an instrument is inherently more precise than a human, which we believe will help the underlying assay approval as well.

Kevin Decatur, [Endenburg Dahlman].

Can you talk a little bit more about your plans for physician education and market development, specifically around awareness of the value of the screening for precancerous solutions. And when should we think about that process ramping up more aggressively.

That process, Kevin, won't start until after we get approval. As you know, there is a strict prohibition on marketing a product prior to approval. And we made a decision when we joined the Company that we would launch this in the right way, not as a lab-developed test but as a test with full-blown FDA study, a full-blown FDA clinical trial and updated approval.

Actually, that's not what I was -- I was more suggesting more in the context of working with third parties to more generally support efforts to educate the community. I wasn't implying that specific to Cologuard.

Oh, okay. So in terms of just general colon cancer screening education, the Company has limited resources and won't make a broader effort in that regard prior to FDA approval. But as you approach approval, you need to have your plan fully locked down. And, Kevin, we'll be in a position to talk more about what our plan is. We actually just presented our go-to-market plan to the board of directors last week at our board meeting. We will continue to refine that plan.

At this point in time, we don't want to talk a lot about our go-to-market strategy. But suffice it to say, the question that you are asking is a really important one -- how do you start to educate physicians that there is a better system for colon cancer screening that includes both what we believe our test and colonoscopy in a paired way. And we'll be talking about that more as we get closer to FDA approval.

Zarak Khurshid, Wedbush Securities.

On the OpEx side, just curious -- it looks like the sales and marketing expense is a little bit higher than we were thinking. Can you just talk about the trend there? What actually goes into that number? How should we think about it as we go into next year? And then on the R&D side, any additional color on the shuffling of R&D spend and how is that tracking relative to your expectations?

Sure, Zarak. From expectations, we are right on where we expected to be. You saw a little over $100,000 increase between second quarter and the third quarter. One of the things Kevin alluded to that we've been working on is this go-to-market plan and doing the market research with primary care physicians and engaging with outside parties. And that's really the key driver in the S&M line. It's a modest increase. We don't expect to see a spike in the S&M expense as you look forward prior to approval. It will continue to ramp modestly as we continue to do market research and prepare.

As you know, we've started to expand our efforts, outreaching to key opinion leaders, and we expect that to continue at current levels, modest increase. When you look at R&D, the real increase in R&D, you can see it ramp dramatically is that's where the clinical trial expenses are falling, and from my perspective, it's exactly where we expected to be at $6.1 million, up $1 million from prior quarter, and we expect it to ramp up in conjunction with clinical trial enrollment.

So as we enroll more subjects, that's what drives almost directly a lot of the increases. So you see this base that you saw over the last year, and that's base R&D. And then you see this uptick for the trial. I would expect to see that continue through the first three quarters and then the fourth quarter of next year will be, obviously, submission.

So -- does that answer your question?

[Edward Alpray, Equinix] Bank.

Thank you for taking my question -- I actually have two. One about your molecular monitors. I was wondering if they are -- for all along now they are stable in stool, and if they are influenced by diet and also by stool type. The second question will be more to better understand for your positioning. So are you positioning your test against colonoscopy?

Great questions. So the first question is are these markers stable on stool? And although we haven't really discussed this publicly, we have done extensive stability studies with our markers in stool, and these markers are very stable in stool. And we have worked really hard to make sure that we understand that scientifically, and that the product is designed to ensure the stability. We have done, again, extensive studying of that question, and the markers are, indeed, stable in stool.

And the second question was -- I'm sorry -- ?

Just a quick (inaudible) on the stability in stool. So are those marker incidents by either diet -- actually laboratory things or before they take the test or -- ?

We have studied this in a significant number of normal patients, and we do not see a correlation between diet or age with the current markers in our study. And we're pleased with that fact. So that is an important question that we believe we've answered over the last two years and, obviously, will continue to look at.

Okay, thank you very much. And the other one that I wanted to ask you was as to your positioning of the test -- are you positioning -- ?

Relative to the colonoscopy? No, we're not. It's really important we think that patients that are willing to undergo colonoscopy to undergo colonoscopy. We believe for those patients someday our test could be used in between those colonoscopies. But 65% of the patients in the US and well over 90% of the patients outside of the US don't undergo colonoscopy. And we're seeing a capping of the rate that people will undergo colonoscopy on a regular basis every 10 years.

So we think that if you pair Cologuard with colonoscopy, you end up with -- ultimately maybe end up with a result like you have for cervical cancer. But that remains to be seen, and we have an important clinical trial here we need to complete.

Thanks for the answer. And if I understood correctly, so are you targeting about 300 test price -- $300 per test price? And --

Yes. That is our current assumption. It could end up being more than that, but that is what we're currently using in our models.

And according to five-year interval, right?

No, approximately -- the interval that we have talked about is -- our FDA study is a point-in-time study with sensitivity and specificity, and that's the most important thing. So our FDA study is not an interval study. We think that the GI societies and the literature and studies afterwards will look at the question of interval.

Okay, thank you. I was just wondering if you are planning to run some health economic studies as well, and -- ?

Yes, those studies, which we talked about a little bit earlier, are in process now, and we expect to have something published in the next 12 to 18 months.

Jacob Price, Edison.

Yes, thanks for taking my questions. Forgive me, I am quite new to the story here, but if you can just remind me -- there was a change in the sampling protocol. Remind me of the thinking behind the change from the sampling device to the correction, the container device, please?

Yes, it was interesting. Before we launched the clinical trial we wanted to make sure our assumption was correct. We looked at -- we sent the small scoop device and the larger cup-like collection device to 300 patients, and those 300 patients, 87% preferred the container device, where there is less interaction with the patient sample and only 13% preferred the scoop device where there was more interaction with the patient sample.

So that was pretty stark data, and we decided to go with what patients clearly preferred. The positive thing about that, too, is that most of the data that we have seen -- in fact, all of the data for stool DNA testing has been collected thus far in a container device with a stool DNA preservative over the sample. And so we -- that gives us very good confidence in terms of the performance of the test in the clinical trial.

Right, okay, thanks. And just to pick up on the previous question about health economics -- I was again slightly confused about this issue about whether the test is being positioned against colonoscopy or not. I noticed in the previous presentation you'd made a statement saying that it was being positioned to avoid colonoscopy. So I was a bit confused by that.

But I guess the key question is does the -- does that $300 price still stuck up in health economic terms if you were to include colonoscopy in the mix as well?

This test will be targeted for people who are -- the main use of this test, we believe initially, would be for patients who don't -- who are unwilling to undergo colonoscopy. And in order to assess the value of the test as an interval to colonoscopy, we think there will be future studies done. The important thing, though, with colon cancer screening is that you get screened.

And so we just don't want to be in a position to advocate that people shouldn't receive colonoscopy, which is a proven method of reducing colon cancer incident and mortality rate. And we'll continue to take that position because we think, at the end of the day, you have to be focused on what's best for the patient. And some patients will prefer colonoscopy, and for those patients who don't, we would expect that our test, like FIT today, is a strong alternative.

Thank you. This does conclude the question-and-answer session of today's program. I'd like to turn the program back to Kevin Conroy, President and CEO, for any final remarks.

Well, thank you for joining the call. In conclusion, we'd like to say that we are very pleased with the DeeP-C study, the trial being on track, getting kicked off in very efficient and rapid manner. I'd also like to highlight that the product development team and the automation teams have just done a tremendous job. And a broad thanks to all of Exact Science employees who have been incredibly focused and dedicated over the last couple of years. And as we continue to make progress, we will need to see that dedication continue.

So thanks to all, and if anybody should have any questions, please don't hesitate to reach out to us directly. Thank you.

Thank you, ladies and gentlemen, for your participation in today's conference. This does conclude the program. You may now disconnect. Good day.