Some things to be getting on with

Like a few blogs out there, things haven’t been happening here for some time. There are a few good reasons for that, and some not so good. Like most folks I have been away on holiday for a while, then I was covering for a colleague which didn’t leave much time or energy to write anything meaningful and then life in the lab got pretty hectic for a while as well. All this has added up to a rather prolonged blog hiatus.

In the interim, Chris posted a comment about his site which I thought I would flag up: his site can be found here - link

I found this recent news from Tufts interesting - they looked at the success rates for oncology drugs through clinical trial, and found that only 8% of therapies reached the market. I haven’t read the full report (it costs $100), but reckon it probably makes interesting reading (especially for those equity analysts/portfolio managers/investment bankers that KinasePro tells us visit his site looking for the Next Next Big Thing) as it has obvious implications for the valuation of biotechnology companies working in the oncology arena.

Biotech valuation is sometimes seen as a bit of a mystery. These companies are quite often valued using a risk-adjusted net present value model, and the risk assumptions that lie behind such models are based on historical rates of attrition through clinical trial. For example, there was a summary in Nature Reviews Drug Discovery a while back of success rates in clinical trials with respect to therapeutic class -the failure rates in oncology were put at ~35% (combined Phases 1 and 2) and at ~65% for Phase 3, giving an overall success rate of ~22%. [LINK, LINK].

So if the real success rate in oncology is now downgraded from ~20% to ~10% there should be a significant downgrading of biotech stocks in oncology. The report also breaks down the analysis by therapeutic type - e.g. MAbs, small molecules etc. It would certainly be interesting to see at what stages the candidates are dropping out, but without access to this paper, I’ll have to remain ignorant.

EDIT - I found another analysis on the success rates of oncology drugs (also from Nature Reviews Drug Discovery, DOI), which puts the success rate at 5%. This seems to be a smaller sample (Top 10 Big Pharma, 1991-2000).

I have a couple of holdings in biotechnology companies - I find analysing companies interesting, not only for financial reasons, but quite often you pick up a lot of what is going on in the industry that you might otherwise miss (and those 10K filings sure make for some exciting reading!).

Anyway, as an aside, I though that I would bring to you an analysis of a company in which I have a modest holding. The company is Renovo (RNVO.L) - a British company, based on work that was done at the University of Manchester looking at scarring of the skin. The company was founded on the work of Professor Mark Ferguson and Dr Sharon O’Kane.

So what’s the big deal with scarring? I guess what most people initially think is one of two things - scars left by surgical procedures (whether for medical or cosmetic reasons) or for the victims of burns or other injuries. As it stands, there are no effective strategies for the prevention or reduction of post-surgical or burn scarring, and this is the niche that Renovo is hoping to fill. The company is positioning itself to exploit both the government/insurance funded markets, as well as self-pay markets.

Renovo had their IPO in April 2006, and raised 58 million GBP to develop their pipeline (they currently have 4 candidates in the clinic). The most valuable of these products is called Juvista - the majority of analysts assign half to three quarters of the valuation of Renovo to this product - so I will only focus on this. Their website is also a useful source of info (as you would expect) and it carries a number of useful summaries of the company, as well as info on their other candidates. They have plenty of cash for the time being, and their burn rate is pretty low.

Juvista

Juvista is a a recombinant protein, TGF-beta 3, protected by a method of use patent (expires in 2017). This candidate is being developed for use in the prevention of scarring, distinguishing it from currently available over-the-counter treatments for the improvement of existing scars (such treatments basically don’t work anyway). It is intended to be used in surgical scar revision (ie excision of existing scars, with reclosure to form a less obvious scar), as well as in the reduction of scarring in new procedures (trauma/injury, breast augmentation, caesarian sections, mole removal, keloid excision etc.)

The company recently licensed Juvista to Shire, but retained rights to develop and commercialise in the EU. The terms of the deal were pretty generous, and can be read about here. What I did was to hunt down a read-made valuation model on the web and then plug some figures in (model found here (credit to BioGenetic Ventures) so as to be able to arrive at a valuation of Juvista. I have broken it down into the two main markets, the EU and the US - Renovo will be due royalties from the US market, but wholly own the EU market, which makes the analysis somewhat more straightforward.

From company announcements, they anticipate the following

1) peak sales ~1.6bn USD per annum.

2) Royalty rate payable to Renovo on sales was described by Mark Ferguson as being of an “industry standard” rate for Phase 3. The industry standard is 15-20% - (link) so I used a figure of 17.5% - I did not use 20% as OSI Pharma are due a royalty from Renovo for licensing of the composition of matter patent that OSI own covering TGF-beta3, which was described by the management as being “low single digit”(such that they said it could almost be ignored). I think the relevant patent is WO199200330 - which has a filing date of 25th June 1991 - and which is thus due to expire in 2011, so it seems unlikely that OSI will be due any significant years worth of sales royalties, if any at all. This could be the wrong patent, I am not entirely sure. Thus I am entirely comfortable with 17.5% figure as a compromise.

3) I presumed 1 further year of Phase 2, 3 years of Phase 3, and a further year till market (i.e. expected launch 2011-2012, which is consistent with Renovo’s own forecasts). You can adjust the number of patients in the various stages of clinical trial. I decided to go a bit over the top and enrolled 3000 into the Phase 3 trials, just to be on the safe side, but this does allow for 4 large-ish Phase 3 trials in mole removal, breast augmentation, scar revision and breast reduction.

4) Success in the phases of clinical trial. I set these at 80% for Phase 2 (which I believe to be too low in this case, given the successful completion of one P2 trial already) and at 67% for Phase 3 (which is the industry standard success rate), and left the FDA rate also unchanged at 81%. I have assumed peak market penetration of 15% - this may seem low, but I thought it better to err on the side of caution. This 15% figure is to be reached after 5 years on the market, with a total product lifespan of 11 years (11 years is the industry average, and is consistent with patent expiry in 2017, with a couple of years extension due to Hatch-Waxman provisions). I costed Juvista at $250 - I am sure I saw this figure on a Renovo presentation, but can no longer find it. Anyway, it seems a reasonable price to me.

First up is the model for the US http://www.box.net/shared/v4lav4p6kz

The royalty rate box is set at zero - I wanted to see the total dollar rNPV of the product, then adjust it myself [which is what the yellow-out boxes describe, with additional conversion into pounds and pence]. The discount rate is set at 20% (the default value from the source) although most estimate of discount rates for biotech companies are in the 9-12% range). The most relevant data can be found in rows 194 onwards.

The model comes out predicting peak sales of approx $1.6bn USD, which is consistent with the company projections. The current risk adjusted NPV under these circumstances for Juvista alone in the US is 139p….however I think the important factor here is the rNPV of the revenues, which for Renovo are worth 64pps

**[I am presuming royalties are based on gross sales not on sales net of costs]

Equivalent analysis for the EU market**http://www.box.net/shared/fx3ugy52dt which comes out with a rNPV for Juvista of 111pps

** This includes a 2.5% royalty to OSI - setting this at zero gives a rNPV of 119pps,
** Product cost of $200 (lower than in the US)]

These two combined give an overall value for Juvista alone of 175pps. It should be borne in mind that these figures were arrived at using a 20% discount rate! Using this model, and given the current share price is only 190p, and that these figures only consider one product (there are two other products in Phase 2, and one in Phase 3)

1) the market is either underestimating the true value of the company (BUY! BUY! BUY!) OR

2) my models are wrong. Take your pick!

Using these spreadsheets you can fiddle with the probability of success in Phase 2 and 3 trials, and you can analyse how the rNPV values change, and workout values accordingly. FWIW, Goldman Sachs have apparently put a value of 330p on Juvista alone, with a target price of 402p (RNVO is currently on their conviction buy list). The GS analyst is a chap called Stephen McGarry, who was rated by the Financial Times as Europe’s top biotech equity analyst, so hopefully he knows his valuations.

DISCLAIMER - I am not a qualified analyst, nor am I regulated by or registered with either the SEC or the FSA, if that makes any difference to you. And don’t buy any shares with money that you can’t afford to lose. Don’t gear up, wear sunscreen, and this is not a ramp. And like I said earlier I own some RNVO stock, although not enough to make me a millionaire. For a read about the general state of biotech stocks in the UK, this is a good read LINK.

[EDIT - I have updated the spreadsheets, both the US and EU are now contained within the one document. I have also tinkered with it a little. Email me if you need help - see the About page for details]