Here is part 2 of my Guardant Health “First Impressions” post.
Can You Tell Me More About the Regulatory Framework?
Ugh. I was afraid you were going to ask me that! I’m not an expert, but I’ll do my best because it’s an important topic. Unlike drugs where approval is overseen by one organization (the Food and Drug Administration, or FDA), there are a few different governing bodies that can approve diagnostic tests, which fall broadly into the medical device realm. On top of that, many different groups have control over reimbursement policies. Even within Medicare, there isn’t just one group doing that! Private payers are also important.
Regulatory
Guardant Health’s laboratory in Redwood City California is accredited by the College of American Pathologists under CLIA (pronounced KLEE-ya), which is short for Clinical Laboratory Improvement Amendments of 1988. The site is also permitted by the New York State Department of Health (NYSDOH). These are top-notch certifications.
Guardant360 is approved by NYSDOH. It is my impression that other states take their lead from New York State, as I’m certain the test is prescribed elsewhere. FDA approval isn’t required to market a medical device, but Guardant thinks FDA approval could increase commercial adoption and improve reimbursement for their product.
The FDA designated Guardant360 as a “Breakthrough Device” in January 2018 and gave the same designation to GuardantOMNI in December 2018. This designation doesn’t imply FDA approval, although it does grant Guardant a more expeditious path toward approval. To get FDA approval, Guardant needs to file a PMA (Pre-Market Approval) application. Guardant filed a PMA for Guardant360 during the fourth quarter of 2019. Breakthrough Device designation typically implies a six-month window to an FDA ruling. That said, any time the FDA asks Guardant questions, the clock stops until Guardant responds. Furthermore, it should be noted that this is in several ways a “first of its kind” PMA submission, and it is reasonable to expect that the FDA will move cautiously and judiciously toward its ruling. During earnings conference calls, management has disclosed that its discussions with the FDA are moving forward smoothly. A PMA for OMNI has yet to be filed.
Payer Coverage and Reimbursement
As mentioned earlier, private payers are important in the U.S. Although claims can be filed with pretty much any insurer, there may be no, or minimal, reimbursement unless Guardant has a contract with the insurer as a “participating provider”. Among the best-known insurers, Guardant has a contract with Cigna, and many of the regional Blue Cross Blue Shield plans. These contracts currently cover NSCLC (non-small cell lung cancer), but Guardant hopes to expand that over time, as well as the number of insurers with which it has contracts. FDA approval will probably be instrumental in this.
The other critical component is Medicare. The Center for Medicare and Medicaid Services (CMS) contracts with MACs (Medical Administrative Contractors), which are private organizations. The MACs service a certain set of states, but also seem to have non-geographic responsibilities as well. To quote former TMFer Simon Erickson, “Medicare is a hugely complex beast…”. Palmetto GBA is the MAC responsible for administering Medicare’s Molecular Diagnostic Services Program (MolDx). Noridian is the MAC that adjudicates requests from California, where Guardant’s laboratory is located, and they are a participant in MolDx. The MACs issue “Local Coverage Determinations” (LCD) and these guide reimbursement levels and covered indications. MACs sometimes issue “Draft LCDs” before the LCD is finalized, and reimbursement can begin in some cases based on the Draft.
In July 2018, Palmetto GBA issued an LCD covering Guardant360 for NSCLC. Noridian followed suit soon after. In September 2018, Palmetto GBA set the reimbursement rate at $3500 per test and Guardant started receiving reimbursements in October 2018. In December 2019, Palmetto GBA expanded its LCD beyond NSCLC to all solid cancers of non-central nervous system origin but added several criteria. Noridian followed suit in February 2020 with a Draft LCD and reimbursements started in March. In my opinion, this appears to be a big deal. It appears that, subject to certain criteria, Medicare reimbursement for Guardant360 is now nearly pan-cancer. Since management withdrew guidance based on COVID-19 concerns, we’re not treated to any hard-and-fast numbers regarding their take on this news. One analyst asked about the effects of the expanded LCD on ASP (average selling price) and CFO Bertocci didn’t expect any spikes, but instead, a gradual increase. To my mind, the “spike” would be in the number of Medicare patients eligible for Guardant360 due to the broadened set of cancer indications, while ASP is a secondary consideration. We may need to temper our expectations as Guardant will need to market their product to a broader group of oncologists than they have in the past. Also, doctors must fully adjust to changed filing requirements for reimbursement – a process that seemed to take about six months after the initial LCD. Still, I think the increased number of eligible patients is – as I said – a big deal. Another potential effect of the broadened LCD in the Medicare space is increased pressure on private insurers to allow liquid biopsy reimbursement for a broader range of cancer indications. During the 2Q19 earnings conference call, CEO Eltoukhy indicated three “proof points” critical to accelerating the adoption of liquid biopsy: positive NILE results; pan-cancer FDA approval, and pan-cancer Medicare approval. With this news, we’re now basically at, “Two down; one to go.”
Before leaving this section, though, I should comment that one of the additional criteria added in the expanded LCD is that the cancers meet the “clinical criteria for complete genomic profiling with next-generation sequencing, or NGS, of tumor tissue to optimize treatment selection decisions, but have insufficient or unavailable tissue for molecular profiling.” (from the most-recent SEC Form 10-K). To me, this seems to be unfortunate wording in that it implies that tissue biopsy is preferred, and Guardant360 should be used when there is insufficient tissue for complete genomic profiling. Although I’ll maintain that the expansion in covered tumor types is a boon to Guardant, language like this makes me hope that an FDA approval leads to adoption of policies leaning toward “blood-first”.
Who Owns Guardant Health and What is the Ownership Structure?
The ownership structure is mostly straightforward; there is only one class of shares. Certain shareholders have an Investor Rights Agreement with Guardant. I’ll touch on that briefly at the end of this section because I don’t think it matters much. In terms of who owns Guardant, there has been a lot of flux over the past year – and I think the GH share price has shown that!
Throughout 2018, Guardant had five significant owners: Entities affiliated with SoftBank Group (32.0%), Entities affiliated with Sequoia Capital (9.0%), Entities affiliated with Khosla Ventures (8.2%), co-founder and CEO Eltoukhy (6.2%), and co-founder, President and Chairman of the Board of Directors Dr. Talasaz (6.0%). Each of these owners has had representation on the Board of Directors. The two co-founders are board members but, as executives, don’t sit on board committees. Dipchand Nishar represents SoftBank, Aaref Hilaly represents Sequoia, and Samir Kaul represents Khosla.
Fast-forward one year, and much has changed. SoftBank now owns 24.3% of Guardant, so they’ve liquidated roughly 4.9 million shares over the course of 2019. Of course, venture capital will always seek to exit after a successful IPO, but SoftBank is in a slightly different position. First, SoftBank has a Joint Venture (JV) with Guardant, where the JV is responsible for the sale, marketing and distribution of Guardant products outside of the Americas, Europe, and Turkey. Japan is the first country being emphasized. It is likely that Guardant will eventually own the JV because there is a put-call agreement where SoftBank has the right to “put” (i.e., sell ownership of) its half of the JV to Guardant under certain conditions, while Guardant has the right to “call” (i.e., buy ownership of) SoftBank’s half of the JV under other conditions. So far, the JV is intact. Second, SoftBank has experienced financial losses – and a “black eye” in the media – associated with its ownership of WeWork and the failed IPO of that company. SoftBank had to pour billions of dollars into WeWork to shore it up, and it sold off shares of its successful holdings, including Guardant, in support of that effort. Dipchand Nishar (SoftBank’s nominee to Guardant’s Board of Directors) has announced that he will not stand for re-election, citing increased responsibility to SoftBank during the COVID-19 crisis. 4.9 million GH shares is a lot of selling pressure for the market to digest, but it is only the beginning.
The current proxy statement shows Sequoia as no longer having a 5% position. Aaref Hilaly (Sequoia’s nominee) resigned from the Board of Directors in November 2019, noting that he was no longer employed by Sequoia. According to an SEC Form 13G filing on Feb. 13, 2020, Sequoia now holds only ~2.3 million shares, or 2.5% of Guardant, so they’ve sold roughly 5.4 million shares. [Please note: The timing and “as of” date of the SEC 13G filing is tied to Guardant’s fiscal year-end, not to the purchase or sale of the shares. Furthermore, since Sequoia is no longer a 5% owner, I don’t think they’ll be required to report the disposition of their remaining shares.] In the next paragraph, I’m going to go out on a limb and speculate that the market hasn’t yet seen the selling pressure from most of those 5.4 million shares.
Entities affiliated with Morgan Stanley now own 5.3% of shares, according to the 2020 proxy statement. While it is possible that some Morgan Stanley mutual fund managers adore Guardant, I think there is a more plausible explanation. Investment banks are often willing to provide liquidity to those wanting to sell a large block of shares without moving the market for those shares. The investment bank will offer the seller some percentage of what they believe they can ultimately get for the shares. They’ll take ownership of the block of shares and then sell them off slowly, so as not to depress the market for those shares. Through that process, the investment bank hopes to achieve much higher receipts than what they paid. My guess is that the Morgan Stanley holding is a block or blocks of shares from either Sequoia, Khosla, SoftBank, or some combination of them. I further speculate that Morgan Stanley will gradually liquidate those shares, and that we won’t see Morgan Stanley as a 5% owner in next year’s proxy statement. If my speculations are correct, I would argue that those owners who sold to Morgan Stanley have taken a noble path that minimizes share price disruption to their fellow GH shareholders.
Like Sequoia, Khosla no longer holds a 5% position. An SEC Form 13G filing from Khosla shows that they owned 1.9 million shares as of year-end, a decline from 7.14 million shares the prior year. Perhaps interestingly, Samir Kaul (Khosla’s nominee to Guardant’s Board of Directors) is standing for reelection this year.
If we assume that Morgan Stanley’s shares came from Khosla, Sequoia, SoftBank, or some combination of the three, then the net amount of GH shares that have flooded the market during 2019 from those three institutions is roughly 10.6 million shares, or roughly 11.2% of all shares outstanding as of April 2020. Again, that is a lot of selling pressure for the market to absorb, but there’s still a bit more to consider.
Drs. Eltoukhy and Talasaz have sold roughly 850,000 and 700,000 shares, respectively, during 2019. I don’t want to give the impression that either co-founder is bailing out – each owns over 4.5 million shares, and it has to be a significant portion of their net worth. But that represents another ~1.5+ million shares of selling pressure in 2019.
Finally, the outstanding share count in April 2019 was just shy of 87 million shares, while a year later 94.5 million shares were outstanding. To characterize the 7.5 million share difference as “selling pressure” would only be partially correct. In May 2019, there was a secondary offering that accounted for 5.2 million shares after underwriter allocations. These were new shares, sold by Guardant, not shares from existing shareholders, and they count toward legitimate “selling pressure”. The remaining ~2.3 million shares are probably from stock-based compensation (SBC). One would hope that many of these shares are safely ensconced in brokerage accounts of non-executive employees. That said, to think that none of the SBC shares found their way into the market would be naïve.
All told, I estimate that somewhere between 17.5-19 million GH shares created selling pressure in 2019. That’s huge, so it is unsurprising that the stock’s trading patterns that year were somewhat divorced from the company’s financial results, and more a function of supply and demand. What will happen in 2020? Your guess is as good as mine, but I think the Morgan Stanley shares will find their way into the market. I suspect that the co-founders will continue to lighten their holdings while still retaining significant ownership. SBC will continue. SoftBank is one wild card I can’t predict. Another is when or if the next secondary offering will occur (more later). Although the selling pressure experienced in 2019 was unpleasant for existing shareholders, the significant reduction in shares from the three largest pre-IPO holders does de-risk the situation going forward, at least somewhat. SoftBank’s holding is still a significant overhang.
I promised to return to the Investor Rights Agreement and discuss why its impact is probably limited. The agreement runs for three years after Guardant’s IPO, so it will expire mid-2021. It covers pre-IPO owners (with some exceptions) giving them additional rights to financial information and “registration rights”. I am not an expert on this latter point, but I think it may relate to any secondary offering of shares to the public that Guardant might undertake. Should such an offering occur, these shareholders could, if I’m understanding correctly, make their shares available as part of that offering, while Guardant would be responsible for the cost of the offering. I don’t want to rule out the possibility of a secondary offering, although I think it could be years before Guardant needs one (see the next section). But, as I’ve described in great detail earlier in this section, many of the covered parties have already significantly reduced their holdings. Should SoftBank choose to push for reducing their position through a Guardant secondary offering in 2020 or early 2021, it is my impression that would reduce Guardant’s risk rather than increase it, although it would cause immediate “selling pressure”. If Guardant chooses to sell additional shares as well, that would dilute existing shareholders, but hopefully bring in enough cash that future secondaries aren’t needed. Hopefully that wouldn’t be done at depressed prices. The GH share price – currently in the mid-$90s – appears to have escaped the trading range it has been in for many months, perhaps increasing the attractiveness of a secondary offering; last year’s was done at $71/sh.
Is Guardant Health Financially-Solid?
I think that the overarching answer to the question posed in the section heading is “Yes,” but I should offer more detail than that and highlight areas of potential concern.
Guardant has more than $500 million in cash and short-term marketable securities. They have another almost $240 million in long-term marketable securities. Their liabilities total less than $120 million. They have no debt, long-term or otherwise. That’s pretty solid.
In Guardant’s history as a public company, it has yet to register one profitable quarter. Only one quarter had positive operating cash flow, but that quarter was free cash flow negative (i.e., capital expenditures were greater than operating cash flow). This, of course, undermines that solidity. That said, the rate of “cash burn” based on the last twelve months indicates roughly six, maybe seven years of cushion.
I am not sure, though, that using TTM free cash flow is conservative enough. Within the past year, as indicated earlier, Guardant has taken on two massive clinical trials: ECLIPSE and COBRA. Expenses will increase. Guardant has estimated the cost of the ECLIPSE trial to be in the $70-100 million range and we’ve probably only seen limited effects from it so far, given the way clinical trials typically ramp up.
Recall also that Guardant may choose to – or be forced to – buy SoftBank’s half of the Joint Venture. Having cash in reserve – or at least an undrawn revolving line of credit – for this contingency seems prudent.
I haven’t tracked this as completely as I want to yet, but it appears that gross margins are improving. From 1Q19 to 1Q20, they rose from 63.1% to 69.6%, and revenues are growing at an appreciable pace: 84% from 1Q19 to 1Q20. Guidance associated with the 4Q19 earnings conference call predicted the revenue growth rate to slow significantly, but the guidance has since been withdrawn and didn’t contemplate the expanded LCDs from Palmetto GBA and Noridian.
The basic tug-of-war that I see is that the existing products are nicely profitable on their own, but the co-founders’ long-term vision for the company is far more ambitious than what the current products support. Proving the worth of the LUNAR program will be expensive and will take years. Right now, it is my impression that there is still plenty of room for Guardant360 and GuardantOMNI to grow – Guardant360 seems particularly underpenetrated, given the completeness of its result set versus tissue biopsy, compared to standard of care guidelines. But I also don’t know what additional clinical trials Guardant will feel compelled to initiate to enable their vision of defeating cancer through early detection and proper treatment selection through thorough data analytics. Guardant felt as if it needs 10,000 patients for a two-year clinical trial for patients with average risk of colorectal cancer, due to the low percentage of trial participants likely to exhibit cancer in the trial’s time frame. How many patients would be needed to prove that testing the entire population as part of an annual physical is cost-effective due to less expensive, more effective treatments when cancer is detected early? How long would such a study need to run? What would such a trial cost? Guardant’s potential is huge, as are its ambitions. How well the co-founders can navigate the company toward its goals while maintaining a firm financial footing is unknown at this point. I think another secondary offering is probably more a matter of “when” than “if”. But with roughly three-quarters of a billion dollars of liquidity on the balance sheet, the need is decidedly not urgent. If Guardant can keep growing the number of tests sold, and the average reimbursement per test, perhaps they can avoid a secondary offering. Profitability is definitely possible, but the co-founders will probably have to throttle back the pace of their ambitions to achieve it and maintain it anytime soon. I don’t think they need to sacrifice their ambitions, just the pace.
Other Random Musings
This is a section heading I’m borrowing from my earnings analyses. I use it for topics that don’t fall neatly into other subject headings or where I’m speculating rather than presenting facts. Since this is a “First Impressions” set of posts rather than an earnings analysis, you’ve already had to suffer through a lot of my speculations. But I did want to use this section to touch on an important topic that I’m not going to cover in depth, at least not anytime soon.
Competition
Guardant has been recommended by The Motley Fool in the past. I won’t say when, how often, or which newsletter(s). It was also a “Saul stock” for a while. When recommended, GRAIL was mentioned as a competitor. GRAIL was spun out of Illumina and has some very deep-pocketed, influential investors. Ray (imuafool) did some great competitive analysis of the liquid biopsy space back in 2018. It may be a bit dated today but has some great information. I really recommend reading it, especially since I’m not going to offer my own analysis.
https://discussion.fool.com/macro-view-of-liquid-biopsy-sector-3…
Conclusion
2019 was an interesting, if difficult, year to be an investor in Guardant. The GH share price started the year under $40/sh., spiked to about $98/sh. in March, and hit an all-time high of $103 in August. It tumbled to a bit above $60 in October and finished the year just below $80. I appreciated the doubling of the share price, to be sure, but the volatility has been difficult, especially when late in March 2020, the October 2019 lows were retested. My goal in analyzing the company at this point in time was to ascertain whether its place in my – and my family’s – portfolio is justified. I feel as if I am now better aware of its risks, have an improved appreciation of its potential, and have a sense of the hurdles it will need to clear to fulfill its ambitions. I am comfortable with my investment, but probably won’t increase it until the path to sustained profitability become clearer.
What Will Be My Role Reporting on Guardant Health Going Forward?
I cannot answer that question with certainty today. One of my official Coverage companies was taken private during the first quarter. That left me with some extra time to devote to analyzing companies. Sheltering-in-place to avoid contracting COVID-19 has given me additional time recently that might otherwise have been spent in fitness classes. Therefore, I chose to study two additional companies this quarter beyond my usual responsibilities. I kind of like the idea of shifting the time I spend analyzing companies toward those that interest me instead of ones that TMF has assigned to me. The most likely case is that I’ll cycle through companies that are important in my family’s portfolio, and put a bit less effort quarterly into my TMF assignments (they’ll get good coverage, but less than what I’ve typically offered). Learning a new (or semi-new) company is always harder work than updating the companies for which I have permanent TMF responsibilities. But I think I’ll benefit from a deeper study of the companies I own. My portfolio is a very diversified, so there is only so much time I can devote to each company. Hopefully Fooldom will also benefit as I cover a broader swath of companies than I have in the past. It is possible that I’ll return to Guardant with a brief update each quarter. But I’m not sure about that – you’ve noticed that I’m not good at “brief”, right?
I hope this was useful for you. If you have any questions or comments, please post them. Ideally, if you want me to see your post, please reply to one of my posts, not just a post in the thread. I am not “up-to-date” on every board where this analysis will be posted, but I do check for replies to my posts with some regularity.
P.S. Don’t expect negative commentary about Guardant Health from Morgan Stanley analysts anytime soon.
Fool on!
Thanks and best wishes,
TMFDatabaseBob (long: GH, AMGN, CI, ILMN)
Maintenance Coverage Fool
See what a “Coverage Fool” does here: http://www.fool.com/community/community-team.aspx
See my holdings here: http://my.fool.com/profile/TMFDatabasebob/info.aspx
Peace on Earth
Please note: I am not a member of any newsletter team. My opinions are my own and do not necessarily reflect those of the TMF advisers. I am a Maintenance Coverage Fool (please see the above link for details). I am not an investment professional, merely an investor.