Valuing biotechs

FezHenry asks how I come up with the valuations for the biotechs I suggested. That is a fair question, and not an easy one to answer.

When you have companies that make products, have revenue, and (gasp) actual profits, then the task can be a bit easier. There are a number of metrics such as price-to-earnings (P/E), or price-to-sales, that you can use to compare a company against another within the industry and see which is a better deal. I am suspicious of these metrics, by the way, on the whole, because earnings are massaged quite a bit, and even sales can be "managed" from quarter to quarter. But at least there are real numbers to work with!

Biotechs that have no products available for sale yet are much trickier. Look at BioCryst (BCRX) which today announced its first order for the IV version of its swine flu drug peramivir. The pricing was MUCH more aggressive than the Street expected and the stock responded with a 15% gain on the day! At this point, forecasting more sales really depends on how much more the government will stockpile, whether doctors start prescribing it off its emergency-use label, if and when the other countries that have negotiated with the company (Israel, China, e.g.) start ordering, and perhaps most immediately important - whether Japan gives its emergency use authorization and they get an order for 500,00 to 1M doses. That would be a $1B order minimum, or $200M at its Shionogi royalty rate - one order alone that is half its market cap! Can you say hello $20s or $30s, from it's $11.39 today? (and yes, I have a little BCRX in my IRA - not a lot but I think this one, too, will do well in the next year)

Back to the actual question though.
Let's take Dendreon for an example, simply because I can do these numbers in my sleep. Provenge is not approved yet, so we have no real metrics. We don't even have a good price point on them - Dr. Gold has stated the one plant they have now, at full capacity, can produce $500M to $1B of sales. It however, is only 25% built out presently... and to complicate matters, they are planning two new plants, each of which will produce 3/4 of the NJ facility.
Now, you probably should also look at the number of patients a year who will get prostate cancer, at the phase at which this will be prescribed (post-androgen therapy) and make an estimate of the market penetration, and then figure the number of patients who actually get the drug multiplied by the cost per treatment. Suffice it to say I believe within 5 years, these three plants will all be at full capacity. (1+3/4+3/4) plants at $500M to $1B gives you a range of revenue of $1.25B to $2.5B ( you can usually also figure in modest price increases per year, but I'm ignoring that, for now). You can use the midpoint, but I like to assume the low point as a reality check. So $1.25B of sales is our first checkpoint.

Now you can apply some other metrics. Five-to-seven times sales is one rule of thumb for projected market cap. So using 5 * 1.25 gives us a $6.25B market cap, 5 years out. Our market cap now is about $3.2B so it's nearly a double - call it $55 from today's $27.78.
But wait!

We now need to discount back to the present, adjusting for risk. I have assumed approval and I have assumed revenues into the future (they will not be making $1.25B next year). At least I assumed the lower levels for the other choices! We can say there is a 10% risk that Provenge is not approved, or has a delay. There may be a 15% discount per year into the future revenue stream. All told that $55 in the future may only be worth $32 or so today. Still a nice 10 to 15% discount to today's price! And please note if I use the midpoints of revenue and times sales, we would arrive at a market cap of $11.25B - or a share price of $90 or so. Discounted that would be $60ish, compared to today's $27.78 - nice! I like to be conservative but I also like to see what might be in store.

You can also use the P/E ratio, once you estimate sales, to arrive at a price. At $1.25B (again, the low point in assumptions) in sales, the margins should be around 25% so there would be $312M in profit. I am going to assume they will issue more shares within 5 years and there will be 150M shares outstanding. That would be over $2 per share in earnings, and with an assumed x25 earnings multiple, we arrive at $50 per share - slightly less than my $55 from the times-sales estimate.

Note all these example all estimate ONLY United States revenue. Dendreon is actively seeking a rest-of-world partner - where they will collect a royalty on ex-US sales. These royalties will have an effect on the bottom line, perhaps as much as 20-25% of the US revenue.

They also have other immunotherapeutics in their pipeline that will follow Provenge, all based off the same method of action. They have learned HOW to construct trials for cancer vaccines, and their agents for breast, ovarian, kidney, colon, and lung cancers should move along more quickly through trials than Provenge did. This will become a growth stock once investors realize cancer may become a manageable disease through these immunotherapeutics.

Whew - lecture on Dendreon is now over. The take away from this lesson should be that there are a HUGE number of assumptions that go into any kind of estimate like this. Patient count, adoption rate, pricing, earnings, competition, production capacity - any of these could be off by factors of 50% or more! My usual method is to try to use the low points of most ranges to come up with a basement level price, and then start tweaking numbers, as not ALL categories will be at the lowest possible point. If the basement price looks attractive compared to today's price, though, you just may have a winner.

Too, the companies do not have to succeed in the end, for the stock to move in the next year or so. Each of these are "story stocks", where there is news coming out in a specific time frame. I actually think my 50-100% price rise will be accomplished BEFORE the actual news. I want to buy in to companies where there is some reason to believe the drugs work, let the pre-announcement excitement build, and then sell some-to-most of the position before the actual news. If GNVC hits $1.50 next spring, I will sell about half of my holdings and then let the rest ride into the interim results.

I have a lot more work to do on some of these models, but I will say that GNVC and SGMO look incredibly mispriced right now, based on potential sales. GNVC could be worth $30+, from today's $0.93 - but the trial itself still may go for 2 years, with a year plus from then for approval, and ramp up. Are you willing to buy $1,000 now, for $30,000 in seven years? SGMO could be a paradigm changer - where every company that wants to modify a single gene has to pay a royalty to Sangamo. In ten years, they may have 1,000 contracts for $10K to $50K each, as a yearly income stream along with their own drug sales from the trials they are running now.

Biotechs may be pie-in-the-sky, but several will pan out, into gold. There WILL be the next Amgen, the next Genentech... I think a couple of the names I've thrown out there may just be those companies.

Regards,
Trond