On June 21, 2012, Onyx Pharmaceuticals popped 37% from the prior close of $44.58 to $61.20 per share. On June 26, it closed at $67.12 after hitting a 52-week high of $67.62. The immediate cause was the June 20 vote of the FDA's cancer advisory panel in favor of approving carfilzomib (now Kyprolis) for patients with relapsed and refractory multiple myeloma who have received at least two prior lines of therapy [See Kyprolis Receives Positive Vote from ODAC, June 20, 2012]. This vote does not guarantee an FDA approval, but makes it highly likely.
I have been an Onyx optimist (with the usual caveats) since I first bought stock at $34.87 per share in May of 2008. Is today's price too high? Should I cash in or cut back my position? More importantly, is it too late to get in, or is this a good time for new investors to get in on what might be a very nice train of future profits? Either way it is a good time to make a new back of the envelope estimate of the future value of Onyx Pharmaceuticals. That done, I'll compare it with estimates from sell-side (Wall Street investment bank) analysts.
By way of background, at analyst conferences and in investor presentations, Onyx management (led by CEO Anthony Coles) has emphasized that while it thinks the Kyprolis data is compelling, it comes from a Phase ll trial. The FDA rarely approves drugs based on Phase II results, it usually requires Phase III trials, which are based on considerably larger numbers of patients. The company already has two Phase III trials of Kyprolis under way. These could provide data sufficient for approval even if the FDA turns the drug down in this round, and in any case would be necessary for approval by the European medical agency.
So why should an advisory board vote (ODAC, the Oncologic Drugs Advisory Committee), even if it was for Kyprolis approval 11-0-1 (the 1 is an abstention), cause such a large jump in the stock price? Most pharma analysts are pretty good at interpreting trial data. Everyone knew, from public presentations, that the Kyprolis data was pretty darned good and so highly likely to get approved when the Phase III data is submitted, if not based on the Phase II data alone.
There are two possible reasons for disapproval: lack of effectiveness, and side effects (adverse reactions, in industry parlance). The concern of stock analysts was mainly about side effects; some serious ones showed up in the Phase II data. That said, adverse reactions have to be taken in context. If a patient has no other treatment options and is likely to die in a few weeks of blood cancer, some side-effect risk is much more acceptable than if the same side effect occurred in a drug intended give long term to control weight or blood pressure, for instance. Kyprolis so far has shown a reasonable safety profile compared to other cancer and chemotherapy drugs.
This is a very risk-averse stock market in general. Just the possibility that Kyprolis data might not get FDA clearance was enough to dog the stock. There is always the possibility that Phase III data would come in worse than the Phase II data. It has happened to other drug candidates, and statistically it should happen every so often just because of the sampling probabilities involved. The ODAC vote did not really change the likelihood that Kyprolis would be on the market sooner or later, but it did give investors a higher degree of confidence in the outcome. Of course earlier market entry also means revenue and profits sooner.
Onyx already has a successful cancer drug, Nexavar, which is marketed by Bayer. The main reason that its profits have been minimal these last few years is that it has taken the Nexavar cash flow and invested in trials for further indications for Nexavar and in other candidates in its pipeline, notably Kyprolis, which it acquired from Proteolix in 2009.
Given that Nexavar revenue already covers basic operation expenses, and that Kyprolis is not a particularly expensive drug to produce in quantity, profit margins on any new revenue generated by Kyprolis should be high.
Multiple myeloma is a fast moving, deadly disease. While current therapies slow it down, they rarely cure it. If approved for third-line therapy Kyprolis could be given to most people who develop the disease. Sadly, the drop outs from the first two lines of therapy would be patients who die.
Between 14,000 and 15,000 new cases of multiple myeloma are diagnosed each year in the United States; worldwide the number is probably between 60,000 and 100,000. A safe, ballpark estimate is that if approved by the FDA, and with no new, improved competitor, Kyprolis could serve about 10,000 U.S. patients per year, and probably about the same number in Europe. Asia has a low incidence of multiple myeloma, and while Africa has a high incidence, the system there is not likely to deliver a significant number of patients in the next few years.
So 10,000 patients U.S. How much per patient? Lenalidomide (Revlimid) with dexamethasone would be a reasonable comparison, as would bortezomib (Velcade). Available cost data varies, but Revlimid appears to run around $8,000 per month, while Velcade is around $6,000 per month, but varies more because it is injected, has more variable time schedules, and is administered by body weight.
Since this is back-of-envelope thinking, I will use $100,000 per year as a guess at Kyprolis pricing. Then, assuming these very sick patients stay on the drug on the average of 1 year, annual Kyprolis revenue to Onyx would be 10,000 times 100,000, which gives us a neat $1 billion per year in revenue.
Keep in mind that it would take some time, several years, to reach a goal of prescribing to 10,000 patients per year. I could have overestimated, or possibly underestimated, the price, duration of therapy, or number of eligible patients.
BusinessWeek says five (investment bank) analysts estimated 2016 revenue at $523 million. Much of the ramp in the U.S. should be in by 2016. Being conservative, I'll use $500 million rather than my $1 billion guesstimate.
Again, given other costs are covered by Nexavar, I'll figure 80% of that $500 million will be profit. That is a nice round $400 million.
Giving a price-to-earnings ratio of 20, also conservative, that would mean a market capitalization of $8 billion due to Kyprolis sales in the U.S. alone. Doubling that for Europe would give us $16 billion, but I should note that national healthcare agencies in Europe are sensitive about the pricing of therapies.
Today Onyx Pharmaceuticals ended with a market capitalization of near $4.3 billion. Onyx ended Q1 with $620 million in cash.
By my back-of-envelope reasoning, today's stock price is still at the low end of its future range, depending on actual outcomes. If Kyprolis is not approved for some reason, obviously we are due for a fall. If it is priced high and data shows it is a compelling choice over other therapies, then with global marketing the $500 million per year estimate I used will prove to be minimal. Even at $500 million a year in revenue, and ignoring cash, the stock should roughly double in price again by 2016.
In addition, if the trials of Nexavar result in approvals for cancers in addition to liver and kidney, there is a lot of upside to the Onyx equation from that quarter.
On the whole I think Onyx is still underpriced, even given the risk of delayed approval and the other usual risks. I think we will know more after FDA approval (if it is granted) and we see how Kyprolis is priced. After a couple of quarters on the market we should also have a better idea of what profit margins will look like.
One last factor to look for in the future is R&D spend. I am not opposed to Onyx enlarging its pipeline, but R&D spend does reduce earnings. At some point investors will need to see solid earnings, or all the speculation about future profits will fall apart. Keeping R&D spend flat as Kyprolis revenue ramps would be a very nice scenario.
Disclaimer: I am long ONXX and will not trade the stock for 3 days after the publication of this report.