We were wrong previously about the likelihood of Provenge showing good data in IMPACT but that’s fine because we believe in the long run our conservatism will save us more than it hurts us. There will be plenty of investing opportunities with better risk-reward profiles than Dendreon was presenting before IMPACT’s top-line results came out. That being said, now that the odds of future Provenge FDA approval look very high, shares of Dendreon are well undervalued at a $21 and change share price according to our model.

So what is Dendreon worth now? To be conservative, lets focus only on Provenge and largely ignore Dendreon’s other pipeline assets like Trp-p8 and Neuvenge (lapuleucel-T). These assets do have real value but for simplicity’s sake, in our model we gave these assets a value of $80 million and that’s all we’ll say about them for now. Since Provenge will be the driving force behind Dendreon’s share price for a long while, lets focus only on it right now.

There are several different models that we like to use when valuing drug developers depending on how mature their pipeline or commercialized drugs are. For Dendreon, lets present a relatively easy to display discounted cash flow model. Here are the inputs into the model:

• First full year of Provenge sales: 2011

• Number of initial eligible Provenge patients: in U.S 102,400 metastatic hormone-refractory patients (Dendreon’s estimate) and in European Union 80,000 (our estimate)

Slide taken from a Dendreon presentation showing Provenge’s market opportunity

• European Union Provenge royalty rate: 25% of sales

• Cost of goods sold: 25%

• Discount rate used for net present value: 14%

• Odds of Provenge eventual approval (and staying on the market once approved): 90%

Below is our model of how Dendreon’s Provenge U.S. sales could ramp up if it is approved. Trying to model sales of a new paradigm oncology immunotherapy such as Provenge in an indication such as late-stage prostate cancer with no existing rival therapies will likely lead to a wide variance between our model’s wild-ass guess of Provenge sales and reality but we are aware of that fact and have built in multiple layers of conservatism into our estimates.

We used our general knowledge and data from IMS Health about the prescription growth rates and trends for newly FDA-approved oncology therapies as the basis for our model. We adjusted the model as best we could for specific information Dendreon has disseminated publicly as well as for how prostate cancer treatment rates of targeted and other advanced therapies (non chemotherapeutics) have performed in solid tumors like breast cancer, lung cancer, and colorectal cancer. As we said just stated in the previous paragraph though, our model could easily differ significantly from the reality of Dendreon’s situation with Provenge but we still feel comfortable making investing decisions based upon it because of the conservatism we embedded into the model and the multiple sensitivity analyses we ran as well (we’ll add those into this article when we have time).

Here is how we modeled Provenge U.S. sales ramping up following a mid to late 2010 FDA marketing approval:

Provenge U.S. sales and expenses estimate (click image for larger version)

A couple of quick notes about this portion of the model:

1) Even after the approval of Provenge, we still apply a “90%” risk adjustment to all future Dendreon cash flows in case something unexpected pops up (safety issues, manufacturing issues, etc) and Provenge had to be removed from the market. This is just another way of saying that we aren’t 100% sure that the future cash flows that we predict will occur with Provenge will occur.

2) We assume that Dendreon will have few problems in scaling up its Provenge manufacturing operations but if there are issues in this regard, we feel that we have captured them in our risk adjustment to its cash flows. It’s worth pointing out that this risk adjustment is on top of the discount rate that we apply to our modeling of Dendreon’s future cash flows

3) We modeled in a linear $200 million a year in Dendreon research and development spending from 2011 to 2019. This estimate will undoubtedly be high in the early years but if Dendreon chooses to start up multiple drug development programs, it could easily be hit in the later years.

4) We assume that in 2018 (in the U.S.) and 2019 (in the E.U.) new competitors to Provenge will emerge that will eventually completely usurp Provenge either through the expiration of some of the key Provenge patents and the entrance of “generic” competitors or through competing branded therapeutics with superior efficacy profiles.

From an old Dendreon 10-k (not necessarily only discussing Provenge):

“Our issued patents expire on dates from May 22, 2007 through July 17, 2018″

5) We modeled in a relatively high 25% cost of goods sold for Provenge to account for its individualized manufacturing process. A small molecule drug would have a much lower cost of good sold (sometimes even around 10%) for example.

6) We priced Provenge at $35,000 for a course of therapy, which is below the wholesale acquisition cost for a course of treatment for most targeted therapies (depending on the indication) that Provenge will be lumped in with. We realize that Dendreon has stated that Provenge will sell at a premium to some biologics for a course of therapy but we just have no idea which biologics they are referencing and in which indications (for instance Avastin is dosed differently in different indications)

Dendreon intends to out-license Provenge outside of the U.S. Here is our model of Provenge sales and Dendreon’s royalty revenue ex-U.S.:

Provenge E.U. revenue and expense estimates (click image for larger version)

1) We penciled in a simple 25% royalty rate to Dendreon on sales of Provenge. This is highly likely to significantly understate what Dendreon will be able to get in a partnership deal for Provenge outside the U.S.

2) We have E.U. revenue from Provenge starting up in 2012 and not 2011. This owing to the fact that Dendreon or its commercialization partner will have to not only go through a full EMEA regulatory review but also work out separate reimbursement in the E.U. countries.

3) We assumed a slower uptake of Provenge in the E.U. and a smaller overall market share in hormone-refractory prostate cancer in the E.U. owing to a different doctor practices and reference pricing and reimbursement issues in some of the E.U. countries.

Admittedly we are being a little lazy with the ex-U.S. portion of our model by not adding in any upfront cash and regulatory and sales milestone payments that Dendreon will almost certainly receive. This cash will help mitigate the need for more equity financings until Dendreon can become cash flow positive in 2011 .

We also did not model in revenue from Provenge outside of the U.S. and E.U. Places like Japan, Canada, and Australia are major pharmaceutical markets though. We don’t model in revenue from these places for two reasons: A) it adds in a little bit of extra conservatism into the model and B) as inaccurate as our Provenge U.S./E.U. sales models may be, we at least have some defensible figures backing our estimates. Trying to estimate sales and royalty rates for Provenge from these places just isn’t something that will add much to our model at this time and our estimates are likely to be even more inaccurate than the rest of our model may be.

Below is our model of Dendreon’s combined U.S. and E.U. Provenge revenue and risk-adjusted discounted cash flow.

Full Provenge rNPV Cash Flow Model

Ok, lets explain this model a little fuller and then go into some of the alternative and informal analyses we tried to see if our Dendreon (Nasdaq: DNDN) valuation model sounds reasonable…..

1) With Dendreon bringing in revenue from Provenge until 2019, we come to a $2.3 billion risk-adjusted, after-tax, net present value of Dendreon’s cash flows.  We add in the $80 million that we believe an acquirer would pay for the rest of Dendreon’s assets and then ignore the $85 million worth of convertible notes Dendreon has because we assume that these notes will convert into shares.

2) At the time of this cash-flow analysis Dendreon had a market capitalization of around $2.1 billion and with the convertible notes ended 2008 with around 98 million shares outstanding. Dendreon was trading at around $21.65 when we recommended buying shares (April 27) and had an intrinsic value of approximately the same figure.

3) Most profitable biopharmas that are still growing rapidly have share values roughly around 8 times their trailing annual sales. At year five in our model (2015), we have predicted Provenge pulling in a little more than $2 billion in revenue and at an 8x P/S ratio this gives us a $16.4 billion valuation for Dendreon. That sounds high to us considering the market capitalization of other biotechs but we’re using a 8x P/S ratio considering the peer group comparisons below:

P/S ratio of profitable biotechs

The fastest growing peers have P/S ratios of 8x or more but the reality of the situation is that most of Dendreon’s closest potential peers (Millennium Pharmaceuticals, ImClone Systems, Pharmion, PDL BioPharma, etc) have been acquired or broken up over the past five years so there are few close peers to compare it to for valuation purposes aside from Alexion, Regeneron, and BioMarin (and some of these peers are not profitable yet). Nevertheless, we feel that this 8x P/S ratio gives us a very rough approximation of what Dendreon could be worth in 2015.

4) Alternatively, if we tried valuing Dendreon based upon our prediction of its 2015 cash flows and applied a 15x multiple to its 2015 risk-adjusted, after tax cash flows of approximately $330 milion, we get an approximate $5 billion valuation for Dendreon.

5) Our P/S and C/F valuation methodologies have obviously come up with widely disparate values for Dendreon. We are fine with this as we know that no valuation model is going to have much forecasting accuracy so many years out with such a unique treatment as Provenge.

If we take the midpoint of our $16.4 billion 2015 P/S valuation for Dendreon and our $5 billion 2015 C/F valuation for Dendreon this gives us a $10. 7 billion 2015 value for Dendreon. If we assume an average of 5% share dilution per year until 2015, this gives us 125 million shares outstanding for Dendreon at 2015.  Therefore, if our estimates are a reasonably accurate model of what will happen with Provenge then this would mean Dendreon shares could be worth approximately $85 per share around 2015.

If we require a 13% minimum annualized rate of return to make investing in near-commercialization stage biotechs like Dendreon a profitable endeavor, then we would be happy selling shares of Dendreon at around $46 per share if its shares spike that high in the coming months. The reasoning behind this is that even though the difference between $46 a share and $85 a share is quite large, once Dendreon shares are worth more than the mid-$40s, then we couldn’t achieve our 13% annualized rate of return goal with Dendreon shares anymore.

Of course we will always be updating and refining our model as new information comes out but if there is no new information that gives us the opportunity to re-evaluate our price target for Dendreon shares, in the near-term the $40s is where we would start to consider recommending a sell of Dendreon shares if there are alternative good investment opportunities out there. Otherwise we hold until Dendreon shares get closer to the $85 range.

continued….