Description

Invitae (NVTA) is a genetics testing company that has been dubbed (by its cheerleaders as well as its management team) the “Amazon of testing”.  As a result, the stock price has rocketed from under $5 in March 2018 to around $19 today, despite escalating cash burn, deteriorating competitive positioning, and significant capital raises.  The short thesis is as follows:

1. Genetic (germline) testing is not an efficacious clinical tool, and will ultimately not be covered by insurance for most uses

2. Even if this is incorrect, NVTA has no advantages versus other competitors in the space, and significant disadvantages versus several of them.  The concept of “economies of scale” is virtually meaningless here, except as it pertains to the collection of data, which NVTA has promised to share with a public data repository

3. The easy gains have been made, with the BRCA1/2 business benefiting from a pricing umbrella provided by Myriad, whose long history in the space have given them a superior data set

4. Competition is set to accelerate

5. Medicare reimbursement is declining 15%/year going forward, and may face a much larger decline due to an October 29th coverage decision

6. The company burns cash in a voracious manner-- it is much more similar to Netflix than Amazon in this regard, if you must make a comparison

7. Management is promotional, misleading, and additionally enlists the help of questionable shareholders/promoters 

8. Valuation is disconnected from reality, given that we have a competitively-advantaged, cash flow positive comp in MYGN trading at roughly the same enterprise value

To take each of these points separately:

FIRST, germline testing refers to the testing of the human genome (or part of it) in order to deduce a correlation between a genetic mutation and the possible manifestation of disease.  If, for example, we notice that everyone with a mutation in Gene X gets liver cancer, then we can prepare accordingly. The best example of this is the BRCA1/2 gene, which predisposes its carriers to breast cancer.  This was the gene that Myriad discovered around 25 years ago, patented, and which was only recently opened to competition. 

However, this simplistic view of the world is not appropriate.  In fact, breast cancer is a special case, as:

1. We have a long and statistically valid sample correlating BRCA1/2 with breast cancer due to 25 years of testing

2. BRCA1/2 genes correlate well with breast cancer, so not too many false positives/negatives

3. There is a prophylactic action that can be taken (mastectomy) for this result that will extend longevity

The reality for most diseases though is much more complicated.  We don’t really know 

1. What genes lead to what diseases

2. Complications around other genetic mutations in the manifestation of the disease

3. Whether the disease will manifest itself in a timely manner such that it is a source of mortality or morbidity

4. Whether this genetic information tells us more than a simple family history would and

5. What to do even if all this were crystal clear-- do you remove someone’s liver, for example?

So, nice in theory, flawed (perhaps fatally) in practice.  Which is perhaps why many studies doubt the results. For example (there are many similar):

• In a recent 21 gene test for Brugada Syndrome curated for clinical validity, biocurators classified only one gene (SCN5A) as Definitive evidence, while all other genes were classified as Disputed with regards to any assertions of disease causality for BrS.

• The cost of indiscriminate expanded panel testing, aside from the financial implications, lies in the return of variants of uncertain significance (VUSs) to health care providers with limited understanding of their meaning. In the study by Beitsch et al,1 the 54% of patients who were found to have VUSs is unacceptably high. These results take time to interpret and explain to patients and may require follow up, additional testing, or review in case of reclassification.6 Of even greater concern, results are frequently misinterpreted, leading to inappropriate clinical management. Kurian et al7 found that many surgeons managed patients with BRCA1/BRCA2 VUSs in the same manner as patients with BRCA1/BRCA2 pathogenic mutations, and one half of average-risk patients with VUSs underwent bilateral mastectomy, which suggests a limited understanding of results among both surgeons and patients.”-Journal of Clinical Oncology 37, no. 24 (August 20, 2019) 

• Many tests have false positive rates in 20%-30% range yet there are no mechanisms in place for accountability. 

SECOND, economies of scale would come from two areas, in theory.  The first is operational-- bigger centers, more machines, etc.  The problem here is that once you get to a certain threshold-- each sequencer is running full out on the same type of test, say-- all the costs are the same.  Everyone buys the same sequencers and microarrays from Illumina, everyone hires the same lab techs, everyone buys the same vials from Orasure, and everyone uses the US Postal Service.  So… where are the economies of scale? NVTA has mentioned some automation, but this isn’t really proprietary, and I can’t imagine it gets you very far down the cost curve, anyway. And despite this statement from the CEO, their cost per test is ~50% higher than MYGN based on public disclosure ($250 vs $160):

“And in building that new business, building that new industry, the nice thing about going through the pain of all of this is once we build it and once we hit a certain scale, there is a network effect that will be put in place, where there will be no catching us. In fact right now we suspect we are already there.”- NVTA CEO NOV 2018 

More importantly, it appears as though their COGS/test is already hitting a wall, with 2Q19 being the only quarter with significant volume growth and no COGS/test decline:

The second area for EoS could be in collecting the data and using it to correlate genetic mutations/variations to diseases.  This has actually led to EoS for MYGN in BRCA1/2, as it has been operating in that space for so long. Basically, ~3% of variations in the gene are labeled “variant of unknown significance”.  So, it might be a problem, but it’s still unclear. The only way to get clarity is to have enough observations of this VUS to know if it clearly causes to breast cancer; since each one is so rare, you need a lot of data.  MYGN is the only company with this data, and they aren’t sharing it, thereby laying claim to having the best test for BRCA.  

In theory NVTA could also build a “data moat” by running very high volumes of tests and pursuing the same strategy as MYGN.  However, their corporate strategy is explicitly not to do this, and their CMO started the effort to crowd-source BRCA data into a public repository in order to work against MYGN’s strategy (and NVTA continues to send all their test outcomes to a public database).  So, this would be quite a strategy shift! Look at their corporate mission:

In other words, “don’t be evil”.  I applaud this approach as a citizen, but perhaps not as a shareholder.

Thus, despite paying lip service to growing into a large company in order to develop a moat, it is highly unclear what this moat is (maybe it’s around Trump’s wall?)

THIRD, it seems as though NVTA has worked through a reasonable business strategy that is now hitting a wall.  MYGN charges a lot for their BRCA test, justified (in their minds) by the better results. So, NVTA can charge a much lower rate (perhaps â…“ as much, using back-of-the-envelope calculations), capture market share from commercial payers and Medicare, and still make some money.  This is only possible in BRCA1/2 though. Other areas are either not reimbursed (since they are not efficacious) or highly price competitive (such as prenatal testing, where they are entering by providing someone else’s test on an outsourced basis) since there is no price umbrella of which to speak.  MYGN, as a competitor, has decided to treat BRCA as a cash cow and redeploy this money into diversifying their own business, such that it is currently only 25% of revenues. They have, in other words, allowed NVTA to exist… for now. It is important to realize that should MYGN decide that growing revenues at a negative margin were a good strategy, they could knock NVTA out of this market very quickly.  For now, that’s not their approach. However, other areas of genetic testing don’t have this benevolent market leader. They are highly competitive, massively price-aggressive markets, and NVTA (as discussed above) has no ability to compete and retain a reasonable margin. This may explain why as NVTA grows, it burns more cash, not less-- the new tests are less profitable!

FOURTH, competition is likely to accelerate from here.  NVTA’s recent move into the DTC space is surprising, not least of which because they said they wouldn’t:

However, it is necessary to get the growth, as doctors seem to have stopped ordering the tests (because the results aren’t meaningful!)  So, may as well appeal directly to consumers with a low-price offer, then have a staff physician rubber stamp the test… The problem is that the giants of the industry (23&Me and ancestry.com) are moving into more complex tests-- in other words, the same market (https://www.businessinsider.com/ancestry-reveals-expansion-into-healthcare-2019-8).  Who’s going to win?  Well, these companies are a lot larger, have great consumer brands, and know how to operate a consumer-facing business… so, I’m betting on them. 

By the way, this is just as the market is starting to slow.  Orasure, which makes the saliva tubes, warned that their revenues were down 7% as the market was saturated, and went on to say:

We continue to see a steady increase in the number of consumers in this area and expect the growth potential in future periods will be significant. In fact, during the second quarter, 15 of our top 20 customers based on a trailing 12-month revenues were in disease risk management submarket.

And more than half of the 45 new commercial genomics customers added during the second quarter were in this category. We are now seeing more customers in the disease risk space moving to patient-initiated model where the test results are given back to the patient through a medical practitioner.

For example, one customer recently announced a patient-initiated service to provide consumers with genetic testing along with telemedicine enabled clinical guidance and the ability to share the results with the consumers' personal physician. We expect other companies to pursue similar patient-initiated offerings.

Labcorp, which runs the operations for 23&Me, confirmed the slowdown in their results, and said it happened incredibly suddenly in November 2018.  In any case, the space is slowing, there are a lot of desperate companies trying to survive, and it should get pretty ugly.

This all comes into an already-crowded space.  The following graphic is from NVTA’s 2017 presentation; it has only gotten more crowded since then:

, under PAMA, NVTA’s Medicare revenue will decline 10%/test this year, then 15%/test for the next few years, on its way to an eventual 60-75% decline.  Here are historical results and a projection of rates going forward

This also begs the question, “Why are their Medicare revenues so large?”  Roughly 23% of revenues and 34% of GP is from Medicare, compared to 7% for MYGN.  Without casting aspersions, let me point interested readers to the onslaught of Medicare fraud in this space:

https://www.kolotv.com/content/news/Seniors-warned-about-scam-offering-free-genetic-testing-560072191.html

https://khn.org/news/genetic-testing-scam-targets-seniors-and-rips-off-medicare/

https://www.cbsnews.com/news/genetic-cancer-risk-medicare-scam-targeting-seniors-may-disqualify-them-from-legitimate-tests

https://www.fiercehealthcare.com/payer/doj-arrests-35-2-1b-medicare-scam-targeting-seniors-for-fraudulent-genetic-testing 

There is also a near-term event here.  Medicare has proposed removing coverage (following the decision of its contractor Palmetto) for germline testing, basically citing the same reasons that commercial payers have-- that it doesn’t help patients or the system as a whole (eg http://www.aetna.com/cpb/medical/data/100_199/0140.html ).  Palmetto recently revised a local coverage determination (LCD) for BRCA1 and BRCA2 genetic testing to restrict coverage for NGS panels when performed in individuals with early-stage disease. Palmetto's revision came at CMS's direction to align the terms of the LCD with a National Coverage Determination it issued a year ago for NGS testing for advanced cancer patients. 

CMS reopened the NCD for public comments in May, and the language used suggests the agency is trying to narrow the terms of its NGS national coverage policy to focus just on somatic and germline testing when done in advanced cancer patients to guide therapeutic decisions, and specify that all other types of NGS assessments, including testing performed on early-stage patients for cancer risk assessment, are not within the scope of the coverage terms. A draft decision is due Oct 29.

https://www.cms.gov/medicare-coverage-database/details/nca-proposed-decision-memo.aspx?NCAId=290&NcaName=Next+Generation+Sequencing+(NGS)+for+Medicare+Beneficiaries+with+Advanced+Cancer&ExpandComments=y&CommentPeriod=0&type=Open&bc=ACAAAAAAQAAA&

SIXTH, cash burn is really something here.  Despite this, management keeps promising that cash flow is just around the corner, and then breaking those promises.  To wit:

The company is currently losing money on each of the tests it offers. But by the end of 2016, Scott expects that the company will start making money. “Genetics is a volume-dependent business,” he says. “We hope to flip to being profitable.”- NVTA CEO INTERVIEW MARCH 2016

         Note: Company reported a $100ml loss in 2016

“We expect the cash used in Q1 and Q2 of this year will be roughly consistent, followed by an accelerating reduction in the back half of the year as payer contracts operationalized with the expectation that we will reach positive cash flow by the end of 2018.”-NVTA CEO, Q4 2016 CC

          Note: NVTA burn accelerated by 40% in H2 2017

“In the short term, we anticipate acquisition and integration related expenses to increase burn, longer term we believe that the acquisition of Good Start and proposed acquisition of CombiMatrix will contribute positively to cash flow, allowing us to reach cash flow breakeven by the end of 2018.”- NVTA CEO Q3 2017 CC

         Note: Company burnt through $92ml in 2018

“We're not going to need to do anything [new capital] for at least the next couple of years at this point given our current burn and current trajectory.”-NVTA Nov 2018

Note: In the 10 months since this statement, Invitae has completed a $185m  equity offering and a $340ml convertible offering as well as issuing 3.4ml shares for acquisitions and $90ml in RSU’s.

Graphically, we have this:

So, there’s that…  The company has doubled down on cash burn this year, moving from a projection of breakeven to $150 million of burn.  This wouldn’t be a big deal (to the market), except that post WeWork, that sort of thing seems to be frowned upon these days…

SEVENTH, management is quite promotional.  I think I already included some quotations that show this pretty well, and it will not surprise you to hear that they are significant sellers of stock (https://www.nasdaq.com/market-activity/stocks/nvta/insider-activity).  I would further add that during the 4Q18 call, more than halfway through a very soft 1Q19, management was not restricting optimism at all, instead talking about how they would reach “billions of people on the planet” one day.  However, in this case we also have their largest investor (ArkInvest) pushing the stock, as well as a strange Tumblr-based stock promoter called Capital Markets Forum (https://ophirgottlieb.tumblr.com/), who seems to have a direct connection to the company.  Several statements he has made seem to indicate a direct connection to the company, such as this one (issued recently):

EIGHTH, and finally, the stock trades at 10x revenues and will likely never be profitable.  More interestingly, it trades at roughly the same EV as MYGN, which has an advantaged position in BRCA, a large business outside of BRCA, 4-5x NVTA’s revenues, $100m of cash flow (vs. $150m of cash burn) and is buying back stock.  I’m not sure MYGN is a great long, to be honest, but it might make for an interesting pair trade.

Catalyst

October 29 CMS coverage decision

near-term weak earnings reports and competitive incursions