Exact Sciences Corporation (EXAS) Earnings Call Transcript & Summary

Okay. Welcome back, everybody, to the life science diagnostics track of the Stifel Healthcare Conference. We're back on the public side with Exact Sciences. And I'm happy to say that we have CEO, Kevin Conroy, with us today. Kevin, thanks for joining and spending some time with us here.

Dan, it's really great to be here. Thanks for the invitation.

Yes, absolutely, my pleasure. Maybe just to start off the conversation with the quarter and where things are with the business. Obviously, these diagnostics businesses, there's a lot to be said about the environment and what you're able to do and what you're not able to do. So can we just sort of start off with a discussion on -- or a comment on doc access, volume trajectory and just sort of the most up-to-date thought on how things are trending coming out of the quarter on Cologuard.

Sure. And Dan, I wouldn't be doing my job if I didn't take us back to our mission, which at Exact Sciences, we have this ambitious mission to eradicate cancer by detecting it earlier, preventing it actually through early diagnosis and guiding treatment. And I'm tremendously proud of the team for how they thought through the pandemic and how we're coming out of it. Your particular question here about Q3, we saw a tremendous growth throughout the business. We grew 31% with Cologuard. And that's with some mighty headwinds that existed that are starting to become tailwinds as we head out of this year and into next year. I'll touch upon a couple of those tailwinds. First of all, we have more reps. We added 400 Pfizer reps to the Exact team in the field, nearly doubling our sales force. And more salespeople equals more sales. And we know this team is starting to have an impact on a combined basis. We saw a real impact in the latter part of October. And now heading into November, we're starting to see the growth. Two, we're getting more access to primary care physicians. We're still only at about 50% of what we were at prepandemic. While that went down to -- we lost nearly all access to -- down to like 20% of the excess that we had prior to the pandemic and that's starting to come back, and we expect that to continue to improve. Also, patients are coming back. So over the course of the pandemic, there were 30% fewer people going in for their wellness visits, for their annual wellness visits, which is where colon cancer screening starts. And those were the headwinds that we faced throughout the pandemic. We've seen tremendous growth, and we feel good about Q4 and heading into next year.

Okay. Yes. Great place to start. Maybe just on the Pfizer reps that you're putting into the field, 400 of them. Have their incentives changed now that they're Exact employees? And should they -- should we think of them as being more productive than your average? I mean it seems stands to reason that getting hired for the first time would have a longer lead time on productivity than coming from Pfizer and being plugged into your own sales force. So how would you suggest that we think about just productivity of those 400 reps that are coming into the picture?

Well, let me start by saying this team of Pfizer reps has an average of 16 years of experience in primary care, most of them in their own territories. So they know the doctor, they know the nurse, they know the administrator in the office and know their kids and birthdays and weddings. It is a team with deep experience and capability. That's why they were at Pfizer. And they -- now that they're part of Exact, they have one focus, and that is Cologuard. Over time, there will be other tests, including multi-cancer testing, that they will promote. And their incentive structure is 100% aligned behind Cologuard. They also have equity in Exact Sciences. Most important, they have gone from selling therapies to selling a test that helps prevent cancer, helps detect it earlier. And we believe that the fire that they bring to Exact Sciences is going to continue to lit all boats. And we -- that's a mixed metaphor at best, but that's the idea behind it.

I understood what you were going for there. Okay. And then maybe just on redeployment of your reps or their reps. I mean I guess starting with what you have as your own base of reps. How does that decision get made? I mean is it by region? Is it by state? Is it sort of ad hoc? Or do you look every 2 weeks or a month and say, where can we be redeploying? I guess I'm trying to understand what we should think about just how folks start getting filtered back into the field.

Well, they're back in the field now, by and large, 95% of the reps are back in the field. Everybody is vaccinated. Offices frequently are requiring vaccinated reps to get in. And so right now, the reps are still calling in the exact same territories that they were calling before. There is some overlap. We'll work that out here at the end of the year and going into next year in terms of making sure that everybody has their own group of docs to call on. So we're not -- there isn't too much inefficiency. And -- but that's a pretty normal exercise that one would go through when you're combining 2 teams. Now it's important to remember that this team of Pfizer reps have been promoting Cologuard for the past 3 years. They know it well. It hasn't always been #1 in their bag, but it is now. And so you combine our historical team that has been with us for up to 7 years with the Pfizer team, and it's a recipe for success and should fuel significant growth as we look into 2022 and beyond.

What percentage of Pfizer reps are now back in the field?

Most of them.

Okay. So more than 50% is the number we should get?

No. About 95% are back in the field.

Pfizer reps. Okay.

Correct.

Good to know.

Almost everybody is back in the field.

Okay. Maybe just on the electronic ordering. Can you just talk about how useful that has been for you? And then truthfully, I mean I guess when I saw the quarter, my first thought was that I would have thought that, that might have bridged the gap a little bit more than it did in 3Q. How do you respond to that? And how do you think that starts playing out for you once we hopefully enter a new phase in 2022?

Well, the reality is it did bridge the gap. The gap was probably wider than you would have expected because during the resurgence, the Delta variant impact during the quarter, what you saw was the entire Pfizer team was taken out of the field. Our team was out of the field in many locations and wellness visits were dramatically down. So you can look at this in one of two ways. The way that we see this going into next year is that the significant increase in electronic ordering capability, which leads to a near doubling of the per office order rate, which is a big deal, sets us up for a strong 2022, assuming that we don't have yet another massive wave of the pandemic. Because if you couple a test that is highly sensitive to promotion, with the ease of ordering -- of electronic ordering, which is now approaching 50% of all orders are coming in electronically, I know that doesn't sound great, but it takes a long time to bring a new test up in electronically through all primary care docs, the 250,000 docs who order Cologuard. And it becomes a pretty big tailwind next year.

Okay. And just so that I understand, the doubling of the rate once the doc visit situation opens up next year, that's assuming that the tele-ordering is completely additive to what it is that you're expecting?

Yes. So the doubling of the rate of ordering is physicians who order and offices that order electronically order about 85%, 90% more Cologuard tests than physicians who order by filling out a form [indiscernible] and faxing it. So it's really -- that's now about half of our orders are coming in electronically and half of them are by fax. Over time, we will bridge that gap and have everybody ordering electronically. That's going to take a little bit of time here.

Okay. And then just keeping with this idea of the drivers that you have into next year. I mean the rescreening opportunity is large. It sounded like you were actually a little bit ahead of your $100 million target that you had put out there for that population. Is that correct? And where are you finding that you're successful there?

Yes. Just as by way of background, Cologuard is recommended by the main guideline groups to be used every 3 years. So to be current with screening, patients should get tested with Cologuard every 3 years, every 10 years for a colonoscopy. That means that this is an opportunity to get people tested again. And we are out there in the field, educating physicians to this guideline. Many, if not most of our customers, still don't know that Cologuard is recommended every 3 years. So that -- this is part of the educating process of 0.25 million ordering customers. Patients, by and large, don't know that either. So there is an enormous push to get people retested. And presently, we have said we expect to do about 100 million of that rescreening revenue. This year, we are ahead of pace. And whereas up to date now, we have about over 1 million people who have been due for rescreening. Next year alone, there's another 1 million people altogether. So there are about 2 million people who are due for their rescreening test by the end of next year. And if you multiply that times $500 test, that's about $1 billion opportunity all by itself. Now we have not yet perfected the ability to get people tested on their third year anniversary. We're working on that. And over time, as we start to get those people screened automatically, we're working on programs with large health systems and physician offices to automate that process. We expect that to be another wave of growth into the future.

Okay. Yes, that was going to be my question is, what is the most effective way to drive those in the rescreening pool to the doctor's office and to get the test?

So the most effective way to do it is you go into a health system that say has 1,000 primary care docs. And let's say that they have 3,000 patients who are due for a rescreen test now. To go into their system and run a program that sends text, e-mail, letters and telephone calls to those 3,000 patients saying you are due for a colorectal cancer screening with Cologuard. We are going to send you a collection kit 2 weeks from now, unless you opt out. You probably end up with a 1% opt-out rate, and then you ship those kits and then you start following up with those patients to return them. That is ideal. That is not the matter of course today, but over time, with the power of our Epic platform integrating with the IT platforms that the customers are on, we have the ability to drive that kind of behavior. Today, reps are going into those offices and initiating a transfer of a list of patients. The reps don't see this, but the doctor sees a list of patients who are due for a 3-year rescreen. And then the doctor can go back into their system and reorder or fill out a new order form and reach out to the patient and let them know it's coming. So this will take time to work out, but it's another leg of growth as you look out over the next 5, 10-plus years. Over time, rescreening patients will become a larger part of our annual revenue than initial screens. There are 45 million people who are unscreened today. So this just keeps growing and growing over time. And it creates kind of a moat around this because patients become very comfortable with this method of screening.

Well, I turn 45 next year. So I'm -- I don't know if I could say I'm looking forward to joining that pool, but I will certainly be in there when the time is right.

Don't wait past your 45th birthday.

No, I will not. I will get it done. Okay. Maybe switch over to the blood side of the equation with -- well, the blood...

Can we touch upon one more thing since you just mentioned it?

Yes.

The 45-year-old population. That's almost 20 million Americans who, by and large, are unscreened. And I guess I would put the question back to you. Would you prefer Cologuard or colonoscopy?

Oh jeez, I guess it depends on what day -- how many drinks I've had and what day you're talking to me. I don't know. It's a good question.

So how does your work life allow you to take 1.5 days off of work away from your cell phone?

These days, it's easier than it used to be. I mean I -- we're generally unaccounted for a lot more than we were at Epic. But I can understand what your question is getting at, which is just from a lifestyle-y standpoint and the ability to not disrupt what it is that you're doing in your daily life. It certainly makes sense. Is that the #1 reason that you get folks -- I mean it's sort of a background question. But just when it is that you do your market homework and you understand the motivations of those that are taking the test, is lifestyle disruption the biggest factor that comes up?

Yes, it's discomfort with colonoscopy. And it is fear, embarrassment and then time. And 40% of Americans work for an hourly wage and taking time off of work, they can't make -- just make up that income. So it's -- Cologuard fits into the lifestyle of people 45 to 49. Then at 20 million times $500 a test, that's a $10 billion opportunity every 3 years, of course. So we're just scratching the surface here. That is yet another leg of growth. So you have these 45 million Americans who are unscreened. 20 million of them are in that -- this demographic. And we expect next year to tap into that more because we now have the top 5 insurers in America who all cover Cologuard for people 45 and older with $0 out of pocket.

Yes. So what does that translate to? Is that about 80% of the population that's not seeing an out-of-pocket cost? And can you bill that higher?

Exactly. Now to get this to be really ordered as a matter of course, you want to get it to 95%. Because if you can go in there -- into a doc's office with a message that 95% of people will not get a bill, they'll say, okay, we'll use this test as a matter of course. We're not quite there yet, but we're not getting much pushback from the payers. It's just a matter of contracting and adding this to their coverage policies.

Okay. So it's more logistics than policy issues or stances on doing so. All right. Okay. Let me move to blood and 2.0. There's obviously plenty of discussion of that these days just given all the things that are taking place in the market. And I'll start with BLUE-C, if I could. The enrollment for you guys is now up at 20,000. Guardant is talking about enrolling 13,000. And I know the difference is related to incidence of cancer within those populations, but it is a really big difference in the numbers. So are there other things that are coming into the equation as far as what it is that the population needs to be in order for a proper trial?

I'm not privy to other aspiring entrants into this space in terms of their clinical trial and the prevalence rate of cancer in the populations, their study. There is some degree of randomness to what sites you enroll and the incidence of cancer in those populations and who's being enrolled. But we expect our Cologuard 2.0 program to be -- we'll submit towards the end of next year. And that's the stool program. The blood program is going to take a little bit longer because interestingly, about 20% of people, even though they're being paid on average about $100 to participate in the study, about 20% of them will do a Cologuard test, will undergo a colonoscopy, but won't give a blood draw. This has always been the case. We've known this is a dynamic and -- so our blood test will be a little bit later than our Cologuard 2.0 test, which is our improved version of Cologuard. And that version of Cologuard, we believe, compares very favorably as a programmatic screening tool to colonoscopy, and we believe it's much better than the FIT test.

Okay. One of the things that you did also talk about was just this idea that in stage 1 patients, a lot of times, you have those cancers removed. And so you don't have the ability to detect with blood the way that you do with stool. Can you just sort of -- and I got a couple of questions after the call about that. Can you just sort of rehash that idea?

You bet. So the way to test a new diagnostic test is to go get a sample from patients with a condition, in this case, colon cancer. And so if you identify a patient in a colonoscopy procedure that they have cancer, you have to get that blood draw before they have surgery because once the tumor is removed, then that would obviously strongly bias your test ability to detect the cancer. And the challenge is that about 45% of cancers found in screening are stage 1. Half of them are really small early stage 1s that are hard to detect because you don't see much DNA methylation in the blood quantitatively, there's not much DNA. Half of those are removed during the colonoscopy procedure. So you can't get a blood draw because the cancer is gone. You can get a blood draw, but there's -- you're not going to see tumor DNA circulating in the patient's blood. And we know that from the prospective studies we've done, what the proportion of stage 1s, early stage 1s versus later stage 1s. Well, patients also who come into a screening center symptomatically, they're bleeding or they have pain, they're typically -- those tumors are farther along, which typically means that they're bigger, which means that they're shedding more DNA into the blood, which means they're easier to detect. So you can see the bias that occurs in a case-control study. We have seen some people with case-control studies not very carefully match the cases that people with cancer to controls coming from the same age, gender. You have to balance that and get more controls than you have cases so that when you run that algorithm, you aren't biasing it, which statistically you -- there would be some bias. And then the other thing is that if you keep retraining on the same cancers that you had earlier detected, you can improve the performance. You know the answer to the test. However, when you take that algorithm and apply it to a naive population or prospectively collected, in other words, blood draw before even the colonoscopy of asymptomatic patients, you end up seeing -- and again, we know this from the work that we've done over the last decade, you see a lower performance. And so that's why we're very careful when we do case-control studies to make sure that we do the proper balancing, get as many asymptomatic patients as possible, as many early cancers as you can. It's frequently just not possible to find those unless you do a prospective study. And that's how we look at things. Other companies look at things in different ways, but it certainly does perfect -- affect performance.

Yes. Okay. Very helpful explanation there. Maybe on the 2.0 assay, you mentioned data in January, so we'll be on the lookout for that. Will that -- will we get a comparison, a full comparison with Cologuard 1.0 by stage? And will we get an adenoma comparison there as well?

Yes. We'll carefully lay out that data, cancer sensitivity by stage, adenoma detection and the false positive rate. Our goal with the Cologuard 2.0 program, our primary goal was to reduce the false positive rate by 30%. Our secondary goal was to improve the precancer detection rate from 42%. We were hoping to get towards 50%. I'm not going to spill the beans here because I know the results. We're very happy with the results from the precancer detection sensitivity or ability. And then the cancer detection of Cologuard, we don't expect to change much because the cancer detection rate is incredibly high. It's 94% for stage 1 and 2 cancers. In fact, it was 100% for stage 2 cancers in the DeeP-C study. So as long as cancer detection remains north of 90%, we're pleased. And improving precancer detection and decreasing the false positive rate, that's the goal of Cologuard 2.0. And what does that allow us to do? It allows us to actively promote Cologuard as a precancer -- as a tool to detect precancer. And two, one of the biggest pushbacks about Cologuard is that its overall positivity rate of 16% feels high to some physicians. That means 1 out of 6 patients go on to get a colonoscopy. We would be able to lower that by almost 30% with improved specificity. And so we feel good about that. And that's going to set yet another moat around, I think, the 2 best screening tests, Cologuard and colonoscopy. And that's why we so fervently invest into these programs because we think it changes everything.

Yes. Okay. Okay. Let me ask you one on MRD, if I can, and then multi-cancer since we're rounding out the session here. On MRD, you had mentioned on the call that you have a pathway towards both a tumor-informed and a tumor-uninformed assay. Thoughts on the merits or drawbacks of those 2 approaches since you do have the unique perspective of what you can do methylation-wise? And then could you envision offering both of those?

Yes. I mean if a patient is diagnosed with cancer -- a friend of mine was recently diagnosed with kidney cancer. And I actually reached out to Steve Chapman at Signatera and said, "Steve, I need your help. And could you help get a Signatera test?" He said, "Absolutely. Here's a sample. Well, here's the sample requirements." And he was incredibly gracious. And that tumor-informed approach is a great approach because it theoretically provides for a really high level of sensitivity. However, what happened if you weren't able to get that sample? You still want to be able to test for recurrence. And so the approach that Guardant is taking makes a lot of sense. We're in a unique position to be able to someday, down the road, do both. And that's why we're working on this program. And it's important because it's a huge opportunity to be able to detect, a, after treatment, did the doc get it all? I mean my friend was diagnosed with pancreatic cancer. She wants to know an answer to that question after she's done with treatment. And then she wants to know if that cancer has recurred as soon as humanly possible before it would show up on a CT scan or a PET/CT. So -- and you can do it with the simplicity of a blood test. So being able to offer either, I think, is something unique that we can do. We licensed the technology from TGen, which provides for the ability to do a bespoke test. And then through our many years of work with the Mayo Clinic around identifying DNA methylation markers that are pan-cancer markers that span dozens and dozens of different cancers, we believe that we have the ability to pursue both. Now we're going to look at the data and see what the data show. But that's due to the 10 years of -- 12 years of R&D efforts that we have had with the Mayo Clinic that give us the methylation approach to do that.

Yes. Okay. And you mentioned that there will be data in 2022. But then someone asked about the CORRECT study and it turns out that it's not -- that's not the data you're referring to. But will we also get CORRECT study data in 2022, which means there's...

No. The CORRECT study, that was -- there is another study which we haven't laid out yet that we expect to have data in the second half of this year. We haven't identified publicly what study that is. The CORRECT II study is a prospective study. So that data will be -- I think the earliest data that we would see would be at the end of 2023 and the full data after that.

Okay. Last one for you, and then I'll let you go. On the multi-cancer side, I think there are a couple of data sets that will come there as well. Is the first one, is that going to be a set of data that's sort of akin to what we saw with GRAIL, where you present different approaches and you give performance associated with those approaches? Or will it be more performance on what is an established assay? In other words, should we have to wonder which approach you will take? Or will we see data on CancerSEEK as it was?

There are 2 types of studies here, a case-control study where you have a confirmed cancer and then you [indiscernible] a blood draw. These 2 studies will be case-control studies. The prospective study is the pivotal study which we have an agreement with the FDA and the design of that study, and we'll start that study next year, that will read out some point beyond next year. But the 2 studies that we will do will be large case-control studies that like in the [ colon ] paper for the Thrive assay originally showed performance by tumor type and by stage, the level of sensitivity and the specificity. So those are the 2 studies we'll see next year. The first study will be to actually deselect markers that don't add much sensitivity or have too high of a specificity hit to remove those markers. And then the second case-control study will be to lock down the algorithm that we will use in the pivotal study. So there's a -- this actually speaks to the rigor that you have to go through to develop a blood-based test. You cannot fool yourself with the samples that you're working with. You need to be really rigorous in terms of matching the controls with the cases. And then you need to do that and not keep retraining of the old samples that you already know the answer to. That's the process that our amazing R&D team is going through next year. And I've already seen some of the data from the work that they're doing. It gives us reason to believe that we are going to have a very high-performing test that will have the ability to impact cancer by shifting it to an earlier stage. I can't wait to see this next leg of human impact towards our mission of early cancer detection and just tremendously proud to be part of the team that is forging this path.

Yes. It will be exciting. No doubt about that. Okay. Kevin, great conversation. I found it really helpful. So thanks for spending the time. Best wishes to your friend, and I will talk to you after the Thanksgiving holiday. Enjoy the time off.

Fantastic. Thank you, Dan.

You bet. Take care, Kevin.

Appreciate it.