Wedbush analysts who had not taken VX-770 into their model for the VRTX valuation added only tiny $2 per vrtx share following the extraordinary Phase III data obtained from the VX-770 STRIVE trial(but allowed $7 for a potential value for VX-809). When you model after Soliris sales from Alexion Pharma, VX-770 for G551D should be worth a lot more than 2 dollars.
See the following and their sales figures.
Some serious matter was discussed in this thread. I brought it up here to have new investors to read the discussions.
The Vertex sales is initially targeting 2000 CF patients in the US and EU. This is just 3% of total US/EU CFer pool. The use of VX-770 will expand to treat young children under 6 at a low dose, and eventually, according to Ian Smith, to treat additional 5% of CFers who have gating mutations. A quantum jump would occur if VX-809 or -661 is successful or if any other correctors being developed by others are successful because the potentiator 770 has to supplement the corrector to repair F508del mutation. There are 40,000 people with this mutation in the US and EU alone.
Still 770 is an ultraorphan drug without a competing repair drug, a miraculous drug as well for 551 mutations, and it deserves an ultraorphan drug price.
Thoughts of whether 809/661 will get approved if it is only a partial fix? How much improvement in lung function? 5%? 10%? I would think a 5% improvement would merit FDA approval and I think that 809/661 will get at least 5% improvement with 770. But I doubt it will garner a better than 10% improvement.
We need a third drug and it might not be from Vertex. F summary of conference said:
"The ΔF508 has two problems: One is a folding issue; the second is an assembly issue. It is believed that this will necessitate two corrector drugs to effectively correct the problems."
They also said: "Our collaborative partnerships have expanded to include Pfizer and Genzyme, both of which will be focused on screening new compounds to address the ΔF508."
So in sum:
1) 809/661 isn't going to be the "cure" but a minor (5%) improvement might get FDA approval.
2) Vertex is working on other compounds, but so are competitors, but that makes 770 more valuable and potentially 661 and 809 if the other correctors are going to the assembly problem and not the folding problem.
A VX-809/VX770 combo valuation, just for fun.
Out of the 70,000 CF patients, ~68% have the dF508 mutation, so there are around 47,500 possible patients. I’ll estimate that 40,000 of those patients treat with VX-809 and VX-770 at a cost of $100,000/year. (I don’t think these numbers are unreasonable. Assuming the drugs work, they will significantly extend patients’ lives, improve their quality of life, and reduce or eliminate the need for other CF treatments & hospitalizations.)
So 40k patients at $100k/year gives revenue of $4 billion annually. Subtracting $400 million for cost of goods, $200 million for SG&A and $1.1 billion of tax leaves $2.3 billion of profit. A modest 10x multiple gives a value of $23 billion to Vertex, or an ultimate value of $112 per share, just for the CF franchise.
But that’s not a sure thing, and it’s not happening today. Let’s say there’s a 50% chance that the VX-809/VX-770 combo works as planned. Then you could say there is $56/share of probability adjusted value to the combo. Assuming we don’t get there until 2014, and discounting back to today at 10%, you get $42/share of net present value for the VX-770/VX-809 combination. Anyone think that’s built into the current share price?
More than a year ago, Tek_jansen posted a valuation model for VX-770+809. Since then, Kalydeco is approved for marketing and promising interim data for the homozygous 508del are out. It is the time to update some of numbers used by tek_jansen, and follow his model for the drug combo valuation.
Out of the 70,000 CF patients, about 48% have the homozygous F508del mutation, so there are around 33,600 who can benefit from the drugs. I’ll estimate that 30,000 of those patients treat with the drugs at a cost of $300,000/year (At this price, VX-809 is distributed practically free of charge). So 30k patients at $300k/year gives revenue of $9 billion annually.
Subtracting 1 billion for a royalty payment to the CF Foundation, another 1 billion for R&D and SG&A and $2.5 billion of tax leaves $4.5 billion of profit. A modest 10x multiple gives a value of $45 billion to Vertex, or an ultimate value of around $200 per share, just for the CF franchise (w/o VX-661).
But the approval for the combo isn’t happening today. Let’s say there’s a 80% chance that the VX-809/Kalydeco combo will be approved. Then you could say there is $160/share of probability adjusted value to the combo. Assuming we don’t get there until early 2015, and discounting back to today at an annual rate of 20%, you get $82/share of net present value for the Kalydeco/VX-809 combination.
This calculation, although well intended and appreciated, assumes a probability of success way too high. 50% for a combination treatment, where VX-809 has shown so far marginal - if any - clinical benefit, is too optimistic. Although speaking about probabilities is always challenging, based on the historical data for phase II drugs, being new molecular entities with a new mechanism of action, the probability to get approved will be in the range of 20%.
If we assume 20% probability following the same calculations of sales, approval by 2014 and discounting back 10% annually (that is a low discount) the value of the combo today is in the range of $17/share. Probably this is partially built into the current price of $48/share.
Actually, the # are better than that.
1) The % with at least one d-508 allele is higher--around 75%.
2) And the # of CFers in the market will grow beyond the 70,000, because there will still be new births increasing the # of patients and the current patients will be living much longer.
$2/share for VX-770 is absurd. I'm not a professional analyst, but let's pretend.
I've seen numbers that 4-6% of the worldwide CF population (70,000) has the G551D mutation. Let's call it 5%, but let's be conservative and say only 4% treat (and that no patients take VX-770 off-label for other mutations). So 4% (2800 patients) treat at $180,000/year (again, I think this is conservative for an ultra orphan drug), giving Vertex revenues of $504 million/year. Let's say that's $400 million/year gross profit, subtract $50 million for SG&A (VX-770 will sell itself, IMO) and $140 million for tax, leaving $210 million/year in income from VX-770.
Put a modest 10x multiple on those earnings and you've got $2.1 billion in market cap, or just over $10/share for VX-770. (I'm sure Wedbush is doing some discounting for net present value and probability of approval, but I don't think any reasonable assumptions get you down to $2/share.)
And $7/share for VX-809... I'm not going to address that except to point out that it will have *twenty times* the number of patients as the $10/share drug VX-770.
What if VX-770 works on 10% of CF patients? I wasn't on the call, but Verity and thirdme say the CF foundation thinks the numbers could be ~14-19%. Let's do my numbers again with efficacy on 10% of patients, and 80% of those treating at $180,000/year.
That's $1 billion of revenues. I'll estimate gross profits of $850 million, minus $65 million of SG&A, and $275 million of tax. That leaves $510 million of annual profit, and a 10X multiple (Why am I being so conservative? I don't know.) gives a $5.1 billion boost to market cap. That's almost $25/share.
How much of that is built into the current share price? Maybe half? The other half may take another year (NDA filing in Q4, approval Q2, 2012), but it should put us around $70 (assuming no other catalysts between now and then).
An obvious holes in your reasoning: the great majority of the world's population is way too poor to afford anything like $180K/year for treatment. The patient pool is much smaller than you make it out to be.
TEK, your analysis is excellent, but very conservative. I would have put 35 for the multiple.
You said: [... Put a modest 10x multiple on those earnings and you've got $2.1 billion in market cap, or just over $10/share for VX-770. ...]
Analysts are not embarrassed to call a buy recommendation for Alexion Phama (ALXN) which has a PE multiple of more than 100 a few days ago. When 770 is approved, they should give a valuation multiple of at least 35. There is no competing drug for at least ten years, and 770 will also be used for the patients with F508del defect whether or not 809 is successful.
If the market gives a multiple of 35 for 770, the value of 770 should be $35 per share.
This means that the market is giving only $12/shr for the rest of the pipeline. This is absurd because Telaprevir alone can raise a revenue of 2.4 B, or $12/shr in the single year of 2012.
Keep on eye out for the NDA for VX-770--I strongly suspect Vertex will seek approval for more than the 551 mutation, i.e., for other mutations which have the protein at the cell surface. Given the nonexistence of other available drugs and the safety of VX-770, I think it's 50/50 then if FDA will approve for those mutations. That's another 3 - 7% of CFers and not off-label. But if not approved for those mutations, off-label use is pretty much assured for those mutations.
"Let's say that's $400 million/year gross profit, subtract $50 million for SG&A (VX-770 will sell itself, IMO) and $140 million for tax, leaving $210 million/year in income from VX-77"
Actually VRTX will pay no taxes for a few years until it has had profits of $3.5 Billion dollars due to its accumulated business loss to date.
if it's not a cure for CF, it's definitely the pre-cursor to a cure.
as a pre-approval bio-tech, VRTX s/b trading ~40... but with ~98% tute ownership, we might see 60 before teleprevir's pdufa date.
and wouldn't it be nice if VRTX trades on the day MRK's drug has its fda a/c.