In the year or so since I last looked at Exact Sciences (NASDAQ:EXAS), quite a bit has changed and most of it for the better. While top-line data from the company's pivotal study of Cologuard didn't live up to the most bullish hopes, I believe it was more than enough for approval and commercial adoption. The company has also decided to keep its tests in-house (meaning it doesn't have to share profits with testing companies) and laid out a logical marketing strategy to target the most probable high-volume users. Last and not least, the company should have a Medicare coverage decision in hand not too long after approval and in time for the commercial launch.
Even with all of these ostensibly positive developments, the shares have lagged the market over the past year (up about 12%). Some of this can be tied to excessive optimism and some of the normal pessimism that tends to hit med-tech stocks as commercialization looms larger on the horizon. Some can also be tied to concerns about reimbursement and whether the Cologuard data was good enough for real commercial uptake. I'm bullish on Exact Sciences at these levels, though I expect a great deal of debate about the ultimate level of reimbursement and market share for the company.
If It's Approved, Will It Get Used?
It's well worth noting that the Cologuard test is not yet approved by the FDA. The company has reported positive top-line data for the pivotal study (meeting the predetermined endpoints), and there will be an FDA panel meeting near the end of March 2014. I can see FDA panelists asking some sharp questions about the test's selectivity (disappointing, in that it was/is below 90%) and pre-cancer sensitivity (42% in lesions larger than 1cm), but I would be quite surprised if the panel does not vote for approval and if the FDA does not grant approval in time for a launch in the second half of 2014.
Approval has never been much of a topic of conversation or debate with Exact Sciences. The real question is what sort of commercial uptake the test will get.
A lot of the basic underlying data have been discussed at length before. There are between 75 million and 85 million Americans who should get regular colon cancer screening, but approximately half do not. The most accurate test (colonoscopy) is uncomfortable for the patient and expensive ($1,000 to $4,000 or more, once every 10 years), while fecal occult blood tests (FOBT) and fecal immunochemical tests (NYSE:FIT) are much less invasive and cheaper ($5 to $30, on the average), but are considerably less accurate and still have problematic compliance rates for take-home tests (the majority of the market).
Cologuard comes into the picture more or less in the middle. Top-line data from the pivotal study showed 92% sensitivity for cancer, 42% sensitivity for precancerous lesions, and 87% specificity. That's superior to FOBT, FIT, and Septin 9 tests, but inferior to colonoscopy. At a potential price of $300 to $500, it would be priced more cheaply than colonoscopy on a per-year basis (the Cologuard test would be recommended every three years), but more dearly than the other tests. It would also offer middle of the road compliance - it wouldn't be as uncomfortable as a colonoscopy, but would still have to surmount the compliance issues of take-home stool-based tests (many people aren't especially fond of dealing with their own waste).
Arguments For, And Against
I can see a couple of limitations to adoption. First, that sub-90% specificity could be an obstacle to use in patients who are otherwise considered at low risk of CRC. Second, it is a home-use test based around a fecal sample and that limits compliance. Management believes it has created a program designed to enhance compliance, but clearly nobody knows yet whether it will result in meaningfully better compliance than historically seen in these tests.
Finally there is the issue of its sensitivity to precancerous lesions. That 42% result was a significant disappointment given earlier studies that had shown sensitivities in the high 50%'s or low 60%'s. I think this will be an obstacle to adoption, but one that the company can mitigate with its marketing efforts. First, the sensitivity was a lot better in larger precancerous lesions (66% if they were 2cm or larger). Second, management points out that over a three-test cycle, the sensitivity would rise to almost 90%. I believe that's a relevant detail as it takes an average of 10 years for precancerous polyps to develop into cancerous growths.
Looking at two other notable novel diagnostic tests to come out in recent memory, Genomic Health's (NASDAQ:GHDX) Oncotype DX in 2005 and Sequenom's (NASDAQ:SQNM) MaterniT21 in 2012), both saw nearly 10% adoption in their first year, and nearly 20% adoption by year two. That makes a 10% adoption curve for Cologuard in years five to seven seem incredibly conservative in comparison.
I would offer a few rebuttal points. First, neither Oncotype nor MaterniT21 have the same compliance issues, as the first uses material removed in a prior procedure and the second is a simple blood test. Second, neither had especially effective alternative tests in the market to compete against. Third, not all tests have that sort of adoption curve, as the challenges with LipoScience (NASDAQ:LPDX) and its LipoProfile test clearly illustrate.
What Will Reimbursement Be?
Next to whether it is approved, how much Exact Sciences can charge for Cologuard is the biggest question still to be resolved. CMS is doing a parallel review of its coverage decision, which means that Exact Sciences should have a reimbursement decision close to the expected time of approval. This is relevant not only because Medicare covers more than 50% of the target test market, but also because many other payers will use the CMS decision as an anchoring point.
Cologuard includes three already-approved tests and a crosswalk methodology of existing reimbursement decisions suggests a reimbursement rate of $500 per test. That is probably the best possible outcome, and I instead go with an estimated ASP of $300 (similar to the company's own base case estimate) for my base case with a bearish assumption of $200 to $250 per test. It's also worth noting that Exact Sciences is running a pharmacoeconomic study, the results of which could be used to support a higher ASP.
Putting It All Together
There are reportedly 10 million fecal blood tests performed each year in the U.S. at present and a target market of 25 million patients per year. I am currently estimating that it takes five years for the company to achieve 6% of that 25 million market and seven years to attain 10%. That suggests almost $470 million in revenue in 2018 and $750 million in 2020 with that $300 assumption. At those testing rates, the company would exceed its internal testing capabilities late in 2017 or early 2018, but would be generating more than enough cash flow to fund an internal capacity expansion.
I feel that there is risk to the sell-side free cash flow estimates for Exact Sciences from 2014 to 2016, as sell-side analysts and investors routine overestimate initial gross margin and underestimate marketing expenses. I do not believe these risks really alter the long-term fair value, though, and I believe Exact's cost structure at $300/test could support long-term free cash flow margins of 20%.
Based on 2023 revenue of over $1 billion and a high-teens free cash flow margin, my discounted cash flow valuation (using an elevated discount rate due to the lack of FDA approval) is about $14, while my biotech-style discounted revenue-based model says $13.75 (due largely to some multiple compression in the diagnostics space). Those numbers go up significantly if the ASP ends up at $400 or $500; the revenue-based target would go to $17 or $21, while the DCF targets would move to $19 and $24.
The Bottom Line
While the pivotal trial data on Cologuard wasn't everything that the company or investors hoped, I believe it's enough to make for a successful commercialization story. There are plenty of unknowns left, including pricing and initial physician acceptance, but I think management is on the right track with its pharmacoeconomic study and its intention to focus on those hospital centers and physicians that make the most use of noninvasive tests today (1% of doctors account for 20% of FOBT/FIT tests).
As the above information makes clear, test ASP is a major driver at this point. Luckily, I believe Exact Sciences shares are still undervalued even on conservative assumptions for pricing and adoption (10% in seven years), leaving even more upside should the company execute well and see better real-world adoption.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.