Xenoport (XNPT), a small drug company with a recently-approved drug named Horizant, has a market capitalization of approximately $300m and an enterprise value of approximately $200m. Xenoport is in the business of developing “pro-drugs” – they link existing oral drugs to a transporter molecule that allows good absorption but then disassociates from the drug once it has been absorbed.
In this analysis we will focus principally on Horizant and one other little-known compound that Xenoport is developing.
Horizant was approved April 6, 2011 for restless leg syndrome (RLS). Horizant (gabapentin enacarbil) is a pro-drug of a widely used drug Neurontin (gabapentin). Neurontin was one of the biggest-selling drugs ever, but is now generic. A somewhat improved variant of Neurontin that is sold by Pfizer (PFE) under the trade name Lyrica (pregabalin) has sales exceeding $3 billion annually. Xenoport has partnered Horizant with GlaxoSmithKline (GSK) in the US for some indications and with Astellas in five Asian countries.
Horizant is a pro-drug of Neurontin that is much better absorbed and has better and more consistent dosing characteristics. It is transformed in the body to the same active ingredient as Neurontin (indeed it was approved by the FDA partly on that basis) and so one would expect it to be efficacious in the same conditions as Neurontin is.
Initial market for Horizant
RLS, which is the initial label for Horizant, is a little-understood disease that severely impacts sleep in some sufferers. According to Xenoport there were approximately 6 million prescriptions to some 2 million patients for RLS treatments in 2010, so it is a substantial market. The dominant existing treatment (85% of market) are dopamine agonists in the form of two approved drugs (Requip and Mirapex), both now generic. Before it became a generic, Requip was sold by GlaxoSmithKline and had sales in the neighborhood of $350m. Mirapex had larger sales, but it is unclear how much were for RLS and how much for its other approved indication, Parkinson’s disease. But in any event, dopamine agonists have an unfortunate long term side-effect – in about 50% of patients the drugs start making the condition worse after some years of use, a condition known as augmentation. Further, Horizant will be the only drug that is currently being promoted for RLS as all the other drugs are generic. GSK will initially use a sales force in the US of about 500 people – that is enough to reach physicians writing some 80% of the RLS scripts. It was recently announced that GSK has set the wholesale cost of the drug at $99 per month, with a launch set for July.
Estimates of peak annual sales for Horizant in RLS in the US vary widely, but $350m seems to be a conservative ballpark – that corresponds to only about 15% patient share, so certainly there is room for upside.
Subsequent markets for Horizant
Neurontin is a very widely-prescribed drug indeed. Its initial labeled indication was initially to prevent seizures, but then another indication was subsequently added for post-herpetic neuralgia (PHN). It is also widely used for off-label indications, including neuropathic pain of all types and prevention of hot flashes. These are both potentially larger indications than for RLS. It is generally expected that GSK will now go forward with an application for an additional indication for PHN for the drug, with other indications likely still to be studied. To date, Horizant has been successful in multiple Ph II trials for PHN, but failed an initial trial in diabetic neuropathy (DN), which is a very large market. It is quite possible that Xenoport will try again in DN given that the first trial showed a substantial trend towards efficacy at the highest dose used despite a substantial placebo response. It should be noted that Xenoport currently has rights in the US for the DN and other indications aside from RLS and PHN, although GSK will remain the commercialization partner. Thus Xenoport will have to decide whether to fund trials in such indications as DN or menopausal hot flash.
The challenge in indications such as PHN will likely be convincing managed care that the drug is enough differentiated from generic Neurontin to warrant premium pricing. In the meantime, there might well be some off-label prescribing for Horizant. GSK will, however, be very careful indeed about not promoting Horizant for any off-label use because Pfizer got into considerable trouble for their off-label promotions of Neurontin. Because of the large size of these other indications, peak sales in these indications could easily amount to $500m, but with a large potential variance in either direction. The fact that GSK is not paying for trials in other indications does markedly reduce the chance that the label gets expanded.
It should be noted that Horizant has an unusually long patent life – currently through 2022, with at least a two-year Hatch-Waxman extension likely. Thus GSK has every incentive to expand the label to PHN and heavily promote the drug.
A new pipeline entrant – a BG-12 prodrug
Recent news from Biogen Idec (BIIB) of a successful late-stage trial in multiple sclerosis (MS) of a new oral drug called BG-12 caused dramatic shifts in market capitalization for both BIIB and their MS rival Teva (TEVA). BIIB went up some 25 points (around $6 billion in market cap) and Teva lost about $4 billion in market cap. So clearly the market views the new drug, which is still some way from being approved, as a potential game-changer in the oral MS space. Now the drug itself is not new – a version of it is approved in Germany and is used for treatment of psoriasis. This has both good and bad aspects from BIIB’s perspective – the drug is unlikely to have unexpected new safety issues, but on the other hand the IP protection for the drug is likely to be weak.
BG-12 is a formulation of dimethyl fumarate (DMF). In the body, DMF is rapidly converted to methyl
hydrogenfumarate (MHF or MMF). MHF is the active drug. Unfortunately for Biogen, plasma levels of MHF are highly variable with the same dose of DMF – this will make dose titration challenging.
Recently Xenoport presented a poster on an alternative prodrug for MHF, Comparison of the Efficacy and Tolerability of a Novel Methyl Hydrogenfumarate Prodrug with Dimethyl Fumarate in Rodent EAE and GI Irritation Models. This prodrug, currently named XP23829, seems to have much better PK properties than does BG-12, at least in this animal model. It has better solubility and permeability, it produces higher MHF blood levels, and in a standard animal model of MS it produced substantially better efficacy. Significantly, it also produced markedly less GI irritation than DMF. GI side effects are a known issue with DMF.
While it is very unusual to give much weight to any preclinical compound, this may be an exceptional situation. Normally a preclinical compound has less than a 1% chance of ever being approved.
But because we are dealing with a prodrug, many of the normal drug development risks are not present – basically the main risk is that the animal PK model is not predictive, with only a small chance of some unexpected toxicity. I would characterize the approval risk here as being in the 50% range (assuming that BG-12 itself gets approved first). Likely Xenoport could go the same 505(b) (1) route it first took for Horizant. If the Xenoport compound is indeed what it seems to be, it would have a number of advantages over BG-12 – better PK and tolerability and also much better IP protection. In that case it could easily be a multi-billion dollar drug. Thus either BIIB or Teva would be well advised to license this compound early, or simply buy XNPT.
While in 2010 Xenoport restructured and in a cost cutting move deemphasized their early stage pipeline, in the aftermath of the far better than expected BG12 data it would seem Xenoport would be wise to prioritize XP283289 and expedite development given BG-12 shifted market caps in the two giants, Biogen and Teva, in sum, over 30-times Xenoport's current market valuation. This is surely not lost on Xenoport management, and indeed in their recent quarterly call they announced the intention to take the compound into clinical trials.
Xenoport has two other mid-stage programs, to which the market is currently awarding very little value.
The first is arbaclofen, a pro-drug of an isomer of baclofen, a fairly widely used generic antispasmodic (the company claims over 2 million scripts annually). The Xenoport version clearly has a dosing advantage (twice a day instead of four times a day) but it remains to be seen whether they see an efficacy boost that will be enough to make managed care pay branded prices. The drug has been in a decent-sized Phase II program for another indication (GERD) that has now been abandoned for efficacy reasons, so safety is likely not an issue. The likely risk here is thus commercial rather than regulatory.
The second mid-stage program (currently in Phase II) is XP21279, which is a pro-drug for L-Dopa, which is widely used in the treatment of Parkinson’s disease. L-Dopa is poorly absorbed and has a short half-life, leading to fluctuations in drug levels that have very unfortunate consequences for Parkinson’s patients. In an open-label Phase IB, the Xenoport drug (which is now formulated to include carbidopa) showed both better PK and increased “on” time accompanied by decreased “off” time compared with the standard of care, Sinemet. This seems to be a moderately low-risk program, both from a regulatory and commercial perspective. If favorable, the results of the current blinded Ph II trial expected before year-end should substantially de-risk this program. If the initial results are replicated, then this drug may by itself justify XNPT’s current market capitalization.
XNPT’s current enterprise value is in the $200 million range. Although the stock popped on approval of Horizant, it promptly came back down again. The presumed bear case is that RLS is a dubious condition and GSK will have trouble getting managed care to pay for treatment, and similarly GSK will have trouble selling for other conditions against generic Neurontin. But it should be remembered that GSK has expertise in this RLS market, and they and Astellas have already paid Xenoport nearly $250 million in upfront milestones, so they clearly believe this will be a substantial drug. Xenoport has the option of selecting either a profit-sharing or royalty deal for Horizant. Currently they have selected the profit-sharing option, which awards them a profit share of between 20% and 50% depending on sales, with the 25-30% range being most realistic. (Note that the first $10m of initial losses are forgiven, and losses beyond that are deferred, so there will be no cash outlay for Xenoport during the initial marketing of the drug). Xenoport can switch to a royalty-based payment, with rates between 15% and 30%, with 20% being a reasonable target.
As of the end of Q1, the company had some $93m in cash and cash-equivalents, with $37m in milestone payments from GSK and Astellas not yet included. XNPT’s guidance is that cash plus anticipated milestones will fund operations through the end of 2013. A financing in the next year or two is certainly still possible though, particularly if the company wants to accelerate development of some of its pipeline.
If it becomes clear that peak sales of Horizant will be in the $350m and up range (corresponding to royalty payments of some $70m a year), then XNPT is a likely double from here based only on this one drug with its long patent life. In the less likely scenario that the drug becomes widely used in other conditions, with total sales in the $1 billion range (corresponding to royalty payments of $250m a year) , then 5x from here is possible. Note that in other indications there might be competition from Depomed (DEPO), who recently gained approval for a long-active version of Neurontin for PHN. This asset is currently unpartnered.
If the drug is a commercial bust, with sales in the $100m range, corresponding to annual royalties in the $15m range, then this drug by itself would not support Xenoport’s current market capitalization. But it would certainly be enough to prevent any precipitous decline in the stock price.
Thus there is a good case to be made for considering XNPT a good buy based purely on Horizant, with little financing or FDA regulatory risk and further upside prospects with downside support given Arbaclofen and their L-Dopa program. The BG-12 prodrug is thus all pure upside, and it is a dramatic one. If the BG-12 program played out in the most favorable manner, then this is one of those very rare stocks with a 10X-20X upside potential.
The bottom line then is that Xenoport is a stock with good support on the downside because they have a substantial interest in an approved drug partnered by a big pharma, but with substantial upside potential from either good sales of Horizant, success with arbaclofen or their L-Dopa pro-drug, or from their new pro-drug of BG-12. It also presents a very good acquisition target (with Teva perhaps the ideal acquirer) given it has a good pipeline and an interesting platform that can be applied to other drugs. So it presents an unusually favorable risk/reward ratio for a small biotech.