This article covers a potentially important part of the future profit prospects for two biotechs, Amgen (NASDAQ:AMGN) and Regeneron (NASDAQ:REGN). It is not, however, a review of each company. The focus is on whether their cholesterol reducers, given by subcutaneous injection, Repatha and Praluent, respectively, can rebound from weak product introductions and meet or even exceed expectations that were prevalent 1-2 years ago. Each product has 12 years from their Q3 2015 FDA approval dates during which no biosimilar can be introduced in the US; and probably 11 years in the EU. Patent term extensions, if granted, might extend protection, typically up to 14 years from approval date. Thus, there is plenty of time for significant growth to occur. I had been very bullish on these drugs relative to what I thought were restrained Street expectations in the year leading up to FDA and EU approval. But when the companies announced $14,000+ annual acquisition costs, I promptly criticized those as unrealistic and dumb, and adopted a much more conservative stance. I was certain that insurers were simply not going to pay anything like that price just to lower cholesterol levels beyond where oral drugs could take them. And so it has played out.
Now I'm hopeful that 2017 could see the beginnings of a change in fortune upward, and that each stock could benefit. The case for this possibility is made below - but without assigning a percentage chance that it will or will not happen. That's up to each reader to think about, if interested, and at the very least, I hope the following discussion provides a framework to interpret certain data presentations that each company should present next year.
The next 2-3 sections comprise more of the business and scientific background of these drugs than the remainder of the article, which is more focused on the business opportunity and the stocks. Feel free to jump down the article as desired.
Beginning more than two years ago, pharmaceutical industry insiders began acting bullishly about the prospects for a new class of injectable LDL-cholesterol reducers called PCSK9 inhibitors, or PCSK9is in my lingo. A number of REGN executives began accumulating the stock, a pattern which continued for some time as the stock rose to levels well above its current price. Most impressive was very large accumulation in and around August 2015 by REGN's chairman, Dr. Vagelos. He, while at Merck (NYSE:MRK), is widely held to be the person most responsible for development and successful commercialization of the first statin, Mevacor. This became a blockbuster and was rapidly superseded by MRK's follow-on statin, the mega-blockbuster, Zocor.
As the Age of Statins moved along into the 1990s and beyond, and Lipitor became the world's best-selling pharmaceutical, the industry and academia signed on to the "LDL hypothesis" that lower is better for prevention of first or recurrent atherosclerotic cardiovascular events, such as myocardial infarctions (heart attacks) or most strokes. That theory has yet to be proven for the PCSK9is, though as discussed below, suggestive evidence has continued to accumulate in favor of it.
When statins went generic, that left the industry looking for a replacement. It found the PCSK9is unexpectedly, and first AMGN and then REGN picked up on the idea to develop an antibody to block the activity of this protein, which works to elevate serum LDL through action on the liver, and the level of which is actually elevated via statin use. Thus, the logic of using these antibodies in patients on statins, including the large number of patients who cannot tolerate very high, or sometimes even moderate to low, statin doses, was quickly understood. That concept has been validated.
There are now two PCSK9is on the market, with no new competition coming from that or a related sector any time soon, if ever. One, developed by REGN in partnership with its lead partner on the drug, Big Pharma member Sanofi (NYSE:SNY), is Praluent. The other, AMGN drug, is Repatha. These drugs are extremely similar in dose, efficacy in lowering LDL cholesterol, and apparently, in safety and tolerability.
The discovery of PCSK9 as a therapeutic target
This protein has only been discovered relatively recently. Its discovery and work in elucidating its potential importance in therapeutics are covered engagingly in a Nature.com article titled Genetics: A gene of rare effect: A mutation that gives people rock-bottom cholesterol levels has led geneticists to what could be the next blockbuster heart drug.
More to the point is a Sanofi - REGN web page titled PCSK9 genetic mutations provide new insights into LDL cholesterol metabolism: Research into PCSK9 mutations provides genetic validation for the key role PCSK9 plays in LDL-C clearance.
I will quote from the page, and comment along the way.
People with proprotein convertase subtilisin/kexin type 9 (PCSK9) gain-of-function genetic mutations were found to have increased PCSK9 activity, familial hypercholesterolemia (FH), high low-density lipoprotein cholesterol (LDL-C) levels, and higher prevalence of cardiovascular disease (CVD).
Second point, in the opposite direction:
Conversely, people with natural loss-of-function genetic mutations in PCSK9 have lower LDL-C levels and reduced risk of coronary heart disease (CHD).
This lower risk was medically significant:
Key findings in PCSK9 loss-of-function genetic mutations
In 2006, Cohen et al published findings on the effect of PCSK9 loss-of-function genetic mutations on the incidence of coronary events (MI, fatal CHD, or coronary revascularization) over a 15-year period in the Atherosclerosis Risk in Communities (ARIC) study.
Y142X and C679X carriers: 28% lower mean LDL-C levels, 88% reduction in CHD riskR46L carriers: 15% reduction in mean LDL-C, 47% reduction in CHD risk.
- Three different loss-of-function genetic mutations were studied, Y142X and C679X in African American patients, and R46L, which was found to occur almost exclusively in white patients
- All were associated with lower LDL-C levels and lower risk of CHD versus non-carriers
The above is dense, but it suggests that there is a decent chance that blocking the action of PCSK9 could lower the risk of CHD and other atherosclerotic cardiovascular diseases (ASCVDs).
This was confirmed (or at least well-supported) as the web page goes on to state:
Subsequently, Benn et al genotyped individuals 20 to 80+ years old from three large independent Danish studies
(N=45,699) representing the general population, and determined the association of the R46L loss-of-function mutation with LDL-C levels, risk of ischemic heart disease (IHD), MI, and mortality.
- Across the three studies combined, loss-of-function genetic mutations were associated with a 13% reduction
in LDL-C (P<0.0001) and a 30% reduction in the risk of IHD (P<0.001, OR: 95% CI: 0.58 to 0.86)
As summarized in a bar graph:
So far, so good.
Finally, the point is made that short-term treatment with PSCK9 blocking agents is not the same as lifelong exposure to low LDL levels due to genetic variations:
Higher than expected decreases in the incidence of CVD seen in those with PCSK9 loss-of-function mutations may be attributable to their lifetime exposure to lower levels of LDL-C.
In other words, short-term reduction of PCSK9 activity is of uncertain value.
In a follow-on web page, SNY/REGN go on to step on that message, which supports the LDL hypothesis. A number of people question it, so it's worth reviewing before going on to the quantitative market opportunity and latest research.
The case for the LDL hypothesis
Most experts and guidelines I have found are in agreement that despite anti-inflammatory properties of statins, the key is LDL lowering, which applies to other drugs. In support of this, SNY-REGN show some data, including the following:
This shows that whether in treating patients with known disease (secondary prevention) or preventing a first episode (primary prevention), the trends are similar, but much steeper for the former:
Adapted from Raymond et al.
The Cholesterol Treatment Trialists' (CTTC) meta-analysis of 26 studies suggests a linear relationship betweenFor every 1 mmol/L (39 mg/dL) reduction in LDL-C, statins produced a 22% relative risk reduction in major cardiovascular events at one year
low-density lipoprotein cholesterol (LDL-C) lowering with statins and cardiovascular (NYSE:CV) risk reduction.
I'm not a current expert at this point on the fine points for, and criticisms of, the LDL hypothesis, though years ago, I played a key role in the conceptualization and clinical development of a branded statin - so I was fairly knowledgeable once. My "take" now is that there is a good reason why the dean of American cardiologists, Dr. Eugene Braunwald, recently was quoted as saying that he was 99% confident that the PSCK9is are effective for prevention of MACE (major adverse CV events).
So - these drugs should work according to theory and observation of the effects of other drugs that focus on lowering LDL (as opposed to drugs that focus on raising HDL).
No guarantees! And note, as discussed later, there's a difference between limiting acute events in a shorter time frame of a year or two (or three), and limiting the insidious progression of atherosclerosis and the related arteriosclerosis that narrows arteries over spans of decades. It is possible that the PCSK9is could affect one without affecting the other.
The next points I want to cover before introducing the details of how I think these companies can surprise the Street in the wrap-up are involved potential market size, cost-effectiveness considerations, expectations of players in the market 1-2 years ago, and certain other matters.
What's the market size?
This is actually a gigantic question mark. That's because it can be defined narrowly or broadly. If one defines the market to only be the highest risk patients, such as those with certain types of high-risk familial high cholesterol states, plus perhaps very high-risk known patients with both acute coronary syndromes plus a major additional risk factor such as diabetes to go along with very high LDL levels despite maximal statin therapy, one can come up with modest number of potential patients. Those patients will be very needy, and high prices for a drug such as Repatha or Praluent can be justified both on cost-benefit rationales and humanitarian reasons.
However, there are giant potential markets out there for these drugs. Take the results from the US government's NHANES ongoing work; the acronym stands for National Health Assessment and Nutrition Examination Survey. From the CDC as of December 2015:
... Overall, 36.7% of U.S. adults or 78.1 million persons aged ≥21 years were on or eligible for cholesterol treatment, among whom 55.5% were taking cholesterol-lowering medication, and 46.6% reported making lifestyle modifications to lower cholesterol.
The write-up also reminds the reader that lowering elevated cholesterol is an important priority:
Cholesterol treatment for the reduction of ASCVD risk is promoted widely in the United States, including activities such as Healthy People 2020 (11) and the Million Hearts initiative (12). CDC-funded state programs use public health strategies for cardiovascular disease and risk factor management outlined in the Million Hearts initiative, including strategies related to improving clinical management of cholesterol.
So, the government is saying that about 40 million Americans are taking a cholesterol-lowering medication, elsewhere noting that the vast majority of them are on a statin.
But now that statins are inexpensive, they are used to some for primary prevention, not just secondary prevention; and willingness of insurers to pay for them and patients to simply take one pill a day of them runs high.
A narrow view of the market is how many patients had coronary bypass or angioplasty. In the US in 2011, that number was around 300,000. This is certainly a modest fraction of the number of identified cases of atherosclerotic vascular disease. Since it's a small portion of total cases of known vascular disease, the addressable market is in the millions of patients just looking at that very "hard" market.
So, one way or the other, however one looks at it, the addressable patient pool just in the US is very large. If a tiny fraction of the number of patients taking a statin were to take a PCSK9i, say 1 million, and the average annual revenue to the company were the $4,500 mentioned above, then revenues for the PCSK9I class would be $4.5B for the year just in this country; and it's a big world out there, so if the above occurred, the total global market could easily be $10B.
This happens to be the low-end of the range estimated in early 2015, per a report saying "some analysts see the entire PCSK9 market being worth $10 billion to $20 billion per year."
This leads to the investment questions. How important are these drugs to AMGN, REGN and SNY; what is the Street thinking; and, can/will either Repatha or Praluent surprise to the upside strongly enough to be a real benefit to shareholders?
I can't quite answer all those questions, but here are some observations.
Why the PCSK9is can be important swing factors for AMGN, REGN and even the large SNY
If these drugs can reach at least $10B in annual sales, which now looks like a pipedream, all stocks can benefit, even SNY. If SNY is going to earn $5B next year, even "only" an additional billion a year in annual profits would be material; and this is very possible. However, since it's difficult for me to get a handle on SNY as a whole, and I don't think much of its management, there's little more to say about it as a stock other than I would buy almost any other Big Pharma stock in preference to SNY.
That leaves AMGN and REGN. In theory, with annual revenues pushing $23B, an extra few billion dollars a year of revenue above that which is expected from Repatha would be very nice, but not transformational. My main reason for thinking that REGN has a better shot at seeing its stock revitalized if its PCSK9i becomes a big seller is that with another hit product, REGN can easily go to the next level and free itself from any further entanglement with SNY, its partner in any case in several products for years to come. A very big success with Praluent arriving by, say, 2019-2020 would also bring REGN's valuation to a much more "normal" level. However, there are unquantifiable risks for REGN and SNY vis-a-vis AMGN in this market:
AMGN has won a patent infringement case in the US against Praluent, and a decision is pending on its further request for a permanent injunction against Praluent sales. An appeal on the trial loss by REGN/SNY is pending, as well. AMGN has also lodged similar infringement suits in the EU. Just for sake of analysis, I'm going to take a middle ground and assume that neither side gets a complete win, which for AMGN would be exclusion of Praluent from all major markets, and which for REGN/SNY would be invalidation of the relevant AMGN patent claims and/or a declaration that Praluent does not infringe any valid claims. Instead, I will assume that AMGN wins on points and gains a royalty stream from Praluent sales that's tolerable for REGN/SNY. This assumption is just that, an assumption, and future events will be needed to allow for fact-based adjustments.
My tentative bull thesis has been in part revitalized by the failure of the much-touted Pfizer (NYSE:PFE) PCSK9i in Phase 3. This drug, bococizumab, was found by PFE not to maintain its therapeutic effect adequately for one year's treatment. This is a fatal problem for a drug used for many years, possibly nearly life-long. PFE also cited a high incidence of injection site reactions and PCSK9i market conditions in the PCSK9i field in announcing that it was dropping this drug. So, it dropped the project recently, which I take as a material positive for both REGN and AMGN and a small positive for SNY.
The only other direct competition that might be coming are Phase 2 compounds. These are a PCSK9i at Lilly (NYSE:LLY), which has been in Phase 2 so long that I'm ignoring it; and a Medicines Company (NASDAQ:MDCO) RNA interference compound in-licensed from Alnylam (NASDAQ:ALNY). The latter might work out, but the failure of a big company to do a deal with, or acquire MDCO, makes me watch it but not worry about it at this point.
Moving on, there is affirmative new data suggesting - but still not proving - that the LDL hypothesis remains valid as applied to PCSK9i usage:
The successful GLAGOV study
Repatha underwent an interesting study, with the GLAGOV acronym, using coronary ultrasound. Patients who were found on coronary angiography to have a vessel with nonobstructive atherosclerosis were treated either with placebo or Repatha for 1.5 years on top of statin therapy as appropriate and as tolerated. They then had repeat angiography with intravascular ultrasound, or IVUS, of that one artery. As expected, the placebo group had no consistent change in plaquing, but the Repatha group had modest plaque regression associated with a drop in LDL cholesterol levels from 93 to 37 mg/dl.
As summarized by a writeup by the American College of Cardiology:
- Patients with plaque regression: 64.3% with evolocumab versus 47.3 % with placebo (p<0.001)
- Major adverse cardiac events: 12.2% with evolocumab versus 15.3% with placebo
Among patients with angiographic evidence of coronary artery disease and on chronic statin therapy, the PCSK9 inhibitor evolocumab resulted in a greater change in percent atheroma volume and a greater proportion of patients with plaque regression. Although not powered for clinical outcomes, major adverse cardiac events were numerically reduced with evolocumab. Larger studies powered for clinical outcomes are warranted.
The regression of plaque is consistent with that induced by statins at high dose, and plaque regression is felt to be associated with stabilization and thus less tendency for plaque disruption and blood clot formation causing artery occlusion.
Also, the decline in major adverse CV events is consistent with a meta-analysis of prior PCSK9i clinical studies, as was published last year in the Annals of Internal Medicine.
A highly-regarded cardiologist, Dr. Steve Nissen, said in a video associated with one of the linked articles above, the safety and encouraging results of GLAGOV suggests that "you can't be too rich, too thin, or have too low an LDL." He was smiling when he quipped that phrase, but he meant it.
And that could be representative of a big marketing opportunity for the PCSK9is if the companies would price the drugs to a level that the insurers believe is fair.
So, if outcomes results for Repatha and Praluent are positive as most cardiologists expect, here's how AMGN and SNY/REGN can turn things around.
The plan for PCSK9is to generate big-time sales momentum
My view is that these companies have to be aggressive to make these drugs big successes. Whether AMGN and the lead marketer for Praluent, which is SNY, actually are just too complacent to do that is something I cannot opine about. I would guess that REGN is young enough and small enough that it's not complacent. How much influence REGN has on SNY is unknown to me.
In any case, in my humble opinion, the two things that might take the PCSK9is from failure to success are:
- value pricing in line with established cost-benefit norms, set mostly by payors, not manufacturers
- an enhanced bifurcated, aggressive marketing approach to payors and prescribers.
The first of those is clear. The price needs to come down drastically from $14,000+ to vastly lower levels. No one cares that these are the least expensive antibodies on the market, which was an argument that one of the marketers for one of the drugs made last year in defending the pricing structure. If, for example, an antibody completely controlled an aggressive cancer, it could easily be worth many times $14,000 per year and still be considered fairly priced. Or, look at Soliris from Alexion (NASDAQ:ALXN), at $400,000-500,000 per year depending on the global market, for ultra-rare diseases.
Putting the case in the reverse way, Revlimid, the blockbuster small molecule from Celgene (NASDAQ:CELG), can justify a $150,000 or so list price per year in the US based on its life-saving characteristics. And it's not even a biologic, just a small molecule.
So what matters is value to the person or insurer that is paying for a product, as with all free market transactions.
Moving on to the marketing angle, the second point is that these companies need to create the perception that statins can only rarely truly "do the job."
If you look at the data presented above once again, that:
"Overall, 36.7% of U.S. adults or 78.1 million persons aged ≥21 years were on or eligible for cholesterol treatment, among whom 55.5% were taking cholesterol-lowering medication..." [largely statins], then the argument that the companies need to make as well as they can is that Dr. Nissen is correct. An LDL cholesterol of 70 is better for patients with ASCVD than 100, but 50 is better than 70, and 37 is better than 50.
And virtually no one can get to 37 with diet, exercise and a statin alone. So, in the US and then the developed world, the addressable patient population is gigantic. This strategy would then allow targeted deals with insurers to narrow the addressable population to, perhaps, the 30 million the companies were talking about 1-2 years ago, with an agreed societal goal to get a certain percentage of these patients (some of whom do not know they are patients yet) to LDLs of 50 or lower. The benefits would be short and long term. But the key is that the companies have to bend on price and then argue big picture health benefits for years to come.
AMGN may have certain advantages in this strategy. First, it owns the GLAGOV data, which will go in its label. The SNY/REGN marketing effort will have to assume that doctors have no trouble generalizing from one PCSK9i to the other.
Second, its CV outcome trial, or CVOT, should have data ready to be presented within a few months, at least top-line data. It looks as though the Praluent data will take until late next year to be final. Again, since the general view is that one PCSK9i is like another, this will not be at all fatal for Praluent, but so far, so good for AMGN competitively.
Third, AMGN's Repatha CVOT covers a broader population than that enrolled in the Praluent CVOT. AMGN's enrolls at-risk patients, SNY/REGN's acute coronary event patients (within one year of enrollment).
However, these two molecules are so similar that I assume that, side effects aside, the therapeutic effects will be identical.
As molecules, Praluent has a longer half-life and is much better suited for once-monthly administration. Also, for unknown reasons, AMGN refused to produce a starter 70 mg half dose of Repatha, whereas Praluent comes in a 75 mg starter dose and then a full 150 mg dose. Many doctors find it persuasive to tell an apprehensive patient that they can start on the low dose of a drug, especially something potentially troublesome such as a protein, and then if needed try the full dose.
So, overall, if both Repatha and Praluent have well-regarded CVOTs, I'd model a 50-50 market share, assuming that Praluent stays on the market. The key is that they will need the Dr. Nissens and other key opinion leaders of the world to agree that what he just said on videotape is what they believe - whether or not it's in the FDA label.
If that's the case, the companies will have a double-barreled marketing message. On the one hand, the (hoped-for) successful CVOTs will show that in the relatively short term, PCSK9i therapy prevents CV events, presumably by stabilizing plaque and thus also decreasing the amount of thromboses that form on the plaque. Then, in the long term, by not only stabilizing plaque but reducing it, the multi-decade burden of progressive atherosclerosis to individuals and society as a whole can, at least in theory, be substantially reduced. This second point will be most persuasive to the world ex-US, which has one-payor systems that can figure long-term health benefits into their cost-benefit analyses; the US system is deficient in this matter due to often-frequent shifts in insurance coverage.
The biggest issue I see is that large pharma companies are hidebound. The industry tends to attract cautious souls due to the regulated, litigious nature of the business. Whether the executives at SNY, which is in charge of the relationship over REGN re Praluent, and at AMGN, have the vision to lower their selling prices to make their products fit within well-accepted cost-benefit schemes is a very open question.
What do I think will happen, assuming positive CVOTs?
My best guess is that just as has happened with Gilead (NASDAQ:GILD) and its HCV drugs, where GILD has been willing to drop its list price by 50%, these companies will do the same. So I do think they can then do the dance with the insurers that the higher-risk patients, say with ASCVD and diabetes (and so on), are worth, say, $7,000 per year.
This article is long enough that it would be overkill to speculate even further on how the stocks might react to strong CVOT data next year, with AMGN's Repatha up first. I would expect that if we get persuasive data, REGN would react as much as AMGN upward, given REGN's less diversified nature. But it would be interesting to consider going long either name coming into the announcement.
Now for a brief summary.
The hope and expectation from the cardiology and investment community is that the CVOTs first from AMGN and then from REGN/SNY will show clinically meaningful benefits, as statins have, from Repatha and Praluent.
If that's the case, then the case is made here that in association with the GLAGOV data, health outcomes of at-risk people can be beneficially affected by these drugs both in the short run and in the intermediate and likely long term as well.
If that's the case, then substantial benefits could accrue to society as a whole and to numerous individuals. These benefits could be measured financially and would indeed include averted healthcare costs and greater work productivity, in addition to the subjecting and highly important improvement in personal well-being flowing from having cleaner arteries.
With PFE and its marketing muscle (and sometimes genius) out of the picture, it is incumbent upon AMGN and REGN/SNY to look at the very large possible market for their drugs, lower their selling prices appropriately, and engage in a fact-based set of arguments that PCSK9is are, in effect, statins 2.0 and deserve widespread use.
If all that happens, the chances of which are unknown, the effects on the corporate P&L statements and on the stocks could be quite favorable. On the other hand, if it does not happen, it is possible that expectations for these drugs are already low. Thus, it is possible that risk-reward considerations could favor taking an optimistic view of the future of Repatha and Praluent as we near the first release of data on Repatha's CVOT.
Note that my personal stock preference is REGN over AMGN for reasons that go beyond the topic of this article.
Thanks for reading and sharing any comments you might have.
Disclosure: I am/we are long REGN,GILD,CELG.
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.
Additional disclosure: Not investment advice. I am not an investment adviser.