Can Somaxon Pharmaceuticals find a bed partner for Silenor?
On March 18, 2010, Somaxon was granted FDA approval to sell Silenor 3 and 6 mg tablets for the treatment of insomnia characterized by difficulty with sleep maintenance. This approval was met with a stock price move from 4 to 10 dollars. Shares were only 30 cents just one year earlier. The stock has lost half its value since the day after FDA approval to just above 5 dollars. Can the stock rebound or will the price continue to drop?
Silenor is a unique dose of the compound doxepin (Sinequan) which is available as a generic in 10, 25, 50, 75, 100 and 150 mg capsules and a 10 mg/ml concentrate liquid. Silenor’s mechanism of action is likely due to histamine blocking. This is a different mechanism from Ambien, Lunesta and Sonata which work through the GABAalpha1 receptor or from Rozerem that works through the melatonin receptor.
Frequency and subtypes of insomnia
Based on polls and surveys, insomnia affects about 35% of American adults. Chronic insomnia, manifested by insomnia more than 3 days a week, is estimated to affect about 10% of the population and is linked to poorer quality of life and mood disorders. Chronic insomnia can be subtyped into four groups. Sleep onset insomnia involves 3 or more nights a week where the sleep onset latency or the similar latency to persistent sleep (SOL or LPS) is over 30 minutes and sleep is fine during the rest of the night. Sleep maintenance insomnia can be defined as more than 30 minutes of wakefulness after sleep onset (WASO) is achieved for 3 or more nights a week. Combined insomnia involves both delayed SOL and increased WASO the same night more than 3 nights a week. Mixed insomnia is either delayed SOL or increased WASO for 3 nights a week. Studies requiring subjects to fill out a sleep diary every morning are likely more accurate than the more common larger phone surveys. One such sleep diary study found the percent of chronic insomnia patients with sleep onset, sleep maintenance, mixed and combined insomnia was 23%, 34%, 16% and 26%. In general, the elderly are more likely to have maintenance or combined insomnia while younger adults are more likely to have onset or combined insomnia.
Studies with Silenor
Silenor was tested in several Phase II and 3 phase III studies in adults and the elderly with chronic insomnia and in one other phase III study in adults in a phase advanced model of transient insomnia (summarized in the form 10-K, for 2008). Briefly summarizing the studies, significance in wakefulness after sleep onset (WASO) and total sleep time (TST) was found in a dose dependent manner. However, improved latency to persistent sleep was generally only found for the higher dose on night one and not 5 to 8 weeks later. Because Silenor did not convincingly show efficacy for sleep onset, the drug is only approved for sleep maintenance. Ambien and Rozerem are FDA approved for sleep onset insomnia and Ambien CR and Lunesta are FDA approved for both sleep onset and sleep maintenance insomnia. Ambien (zolpidem) is only approved for short term use while Ambien CR, Lunesta and Rozerem are approved for any duration of use.
The drug was generally well tolerated with no significant increased adverse effect compared to placebo. This differs from the anticholinergic side effects commonly seen with the doses of doxepin used to treat depression (75 mg or higher) such as dry mouth, dry eyes, or persistent somnolence.
Opportunities
With over 25 million Americans suffering from chronic insomnia and more suffering from transient insomnia, the potential market is very large. Even if just the 8 to 9 million patients with pure sleep maintenance chronic insomnia are counted, sales could be huge. If one assumes a wholesale price of about $120/month and 10% penetrance, annual sales could break 1 billion. The side effect profile is excellent and patients taking Silenor are probably less likely to experience rare parasomnias such as sleepwalking or other sleep related activity that have been seen with Ambien and Lunesta. Additionally, unlike Ambien and Lunesta, Silenor will not be a scheduled substance with the DEA. This allows for more widespread product sampling. Most doctors prefer to write for non-scheduled drugs and Silenor prescriptions could be written for 12 months instead of only 6 months for Ambien or Lunesta.
Challenges
Doxepin is not a novel compound, though the 3 and 6 mg Silenor products forr the insomnia indication will have patent protection for a while. The sleep market is dominated by generic and branded products and the off-label use of many drugs, including doxepin is common. Generic doxepin 10 mg capsules are only 10 to 15 cents a pill. Liquid doxepin can easily be dosed at 6 mg with a dropper at pennies a day. Because the 10 mg dose has not been tested for insomnia, Silenor 6 mg can not claim superiority in either efficacy or tolerability against doxepin 10 mg. Indeed, their own data suggests that 6 mg might not be the optimal dose as 6 mg performed better than 3 mg which performed better than 1 mg. Could a higher dose have worked better? Obviously, Somaxon was not motivated to test the 10 mg dose as a new branded product since a generic already existed.
Even the lower doses have weak patent protection for the chronic insomnia indication. The in-licensed patent for the treatment of chronic insomnia, covering doses from 0.5 mg to 20 mg of Doxepin, will expire in 2013 (from the form 10-K for 2009). The in-licensed patent for the less commonly treated transient insomnia does not expire until 2020. Furthermore, they do not have patent protection outside the US.
Studies did not convincingly support a longer term benefit for Silenor in reducing the latency to persistent sleep. Therefore, the FDA only approved Silenor with the indication for the treatment of insomnia characterized by difficulties with sleep maintenance. Many insomnia patients have either sleep onset insomnia alone or both onset and maintenance insomnia. These patients would likely benefit more from other products. Sales representatives will not be able to market the drug for sleep onset, mixed or combined chronic insomnia. Doing so would risk penalties for off-label promotion. Also, because 6 mg was not tested against 10 mg or higher doses, Silenor reps will not be able to easily counter questions asking if the 6 mg dose is better (either in efficacy or tolerability) than the other off-label doses.
Conclusions
Silenor appears to be an effective agent for the treatment of pure maintenance insomnia that is safe and very well tolerated. These positive attributes are balanced by its inability to prove efficacy for sleep onset insomnia and the lack of data to distinguish itself from the off-label 10 mg dose of doxepin. All things being equal, doctors would prefer to prescribe an on-label non-scheduled product. However, insurance companies and prescription benefit managers have a nasty habit of taking cost into account. Paperwork is less when generics are prescribed. Hence, I seldom need to fill out additional forms for generic Ambien (zolpidem) or for the off label use of doxepin, trazodone, mirtazapine, gabapentin or other generic compounds. Indeed, many plans require that a patient fail zolpidem before being covered for one of the branded products that is FDA approved for insomnia. Additionally, many patients specifically ask for a generic compound over a branded one. The patients who may best benefit from low dose doxepin, the elderly, are also the patients most likely to request a generic compound and to be unable to afford their medication. Though Somaxon may offer coupons and vouchers that can help defray Tier 3 co-pays for privately insured patients, Medicare does not allow the use of these. In my experience, patients do not care whether a product is on-label or off-label as long as it is effective, and for most patients, affordable.
For all these reasons, I believe that Silenor will have difficulty achieving impressive sales. Marketing, manufacturing and regulatory expenses will take a big chunk out of sales. This has likely been realized by other pharmaceutical companies as no partner has come forward despite nearly 3 months passing since the FDA approval. Somaxon appears to still be counting on a partnership as they have not yet developed or rented their own sales force. Somaxon may be forced to take a less lucrative contract to cement a partnership. With no other compound in the pipeline, they must act soon as the clock is ticking with a potential patent challenge in 2013. If one is interested in taking a long position, a better entry may be at 4 dollars as this was the price before FDA approval and is just above the 200 day moving average. When it comes to investing in Somaxon, I will avoid taking a position and avoid the sleepless nights.
Acorda Therapeutics (ACOR): Ampyra gets FDA approval
The FDA approved Acorda’s Ampyra (4-aminopyridine, 4-AP, fampridine-SR, dalfampridine) to improve walking speed for MS patients. The 10 mg sustained release drug uses drug delivery technology developed by Elan (ELN) and is taken twice daily. A risk mitigation strategy will be utilized to educate about the risk of seizures, which ocurred in less than 1% of patients in the phase III studies. Unlike the immunomodulators, the drug can be used for patients with any type of MS who have leg weakness. Primary progressive MS and non-relapsing secondary progresive MS will likely use the drug in monotherapy. Those with realpsing remitting or relapsing sencondary progressive MS will be more likely to add the medication onto existing immunomodulatory therapy. 4-AP acts through a voltage gated potassium channel to prolong the action potential leading to improved conduction through demyelinated and poorly remyelinated axons and increasing the release of acetylcholine at the neuromuscular junction. It has been shown to improve walking speeds in a subset of patients with multiple sclerosis.
Patients with mild and moderate weakness are more likely to benefit than those with severe weakness or those with severe spasticity. Many MS patients have no significant weakness. Thus, one may assume that about 50% of MS patients are potential candidates for treatment. In the trials, 35% of subjects with MS significantly benefited. Hence, 17% of MS patients could potentially become long-term users of the drug. Of course, penetration will be lower as some patients will have side effects, some will decide not to continue on therapy and some will never start due to fear of seizure. Twice daily dosing will have minimal impact. Others will not start due to cost, insurance barriers or be under management from a conservative physician. Taking all this into account, I would expect a peak use of 7 – 9 % of all MS patients in the US, Canada, EU, Australia/NZ and Japan (1.2 to 1.5 million total patients). This would be in the 90,000 – 135,000 patient range. There will be minimal off label use for Lambert Eaton myasthenic syndrome, some neuropathies and possibly mild spinal cord injuries.
The drug will likely be priced around $8000/yr. Complicating the calculation, Biogen Idec (BIIB) has the rights to the drug outside the US. 4-aminopyridine is very inexpensive. However, with 16% royalties to Elan, marketing expenses, and other overhead, the ultimate margin may average about 30-40% at steady-state but be far lower initially. Hence, when steady state is reached in 4 years, profitability should be about 200-300 million (2500/yr x 100,000 patients). There is no other agent to offer short-term competition, though 3, 4 diaminopyridine has a similar profile and is in clinical trials.
FTY720: Topline Data from FREEDOMS released today
Novartis released topline data from the Phase III FREEDOMS study comparing 2 doses of FTY720 (Fingolimod) to placebo. Previously, the Phase III TRANSFORMS study comparing rthe same 2 doses of FTY720 to Avonex showed that the novel oral agent had a 38-52% better relapse rate reduction than the injectible interferon.
The current study showed a 54 to 60% relapse rate reduction compared to placebo and also showed 30-32% reduction in disability progression. There was little difference in efficacy between the 0.5 and the 1.25 mg daily dose. As would be expected with this significant clinical data, MRI parameters were significantly better with FTY720 than placebo, though details were not released. There were three deaths, all felt to be unrelated to treatment, 2 in placebo-treated patients and one in a high-dose FTY720-treated patient.
The combination of the TRANSFORMS and FREEDOMS data supports the use of this drug for the treatment of MS. The efficacy places it between the interferons and Tysabri, similar to the efficacy shown by oral cladribine in the CLARITY study. Combining both studies, there have been 3 deaths in patients treated with 1.25 mg and no deaths in those treated with 0.5 mg (out of about 860 patients for each dose in the 2 studies). In TRANSFORMS, tolerability was a little better for the 0.5 mg dose compared tot he 1.25 mg dose.
I expect that the NDA will be submitted to the FDA by mid-January for the 0.5 mg dose of FTY720.
Novartis: Powering into the MS Market
Most pharmaceutical companies fuel their growth by creating more product lines within their core strengths. Entering highly contested therapeutic areas is a risky endeavor, but one a large pharmaceutical company is capable of funding. Novartis currently has only a moderate presence in Neurology, with modestly successful products including Exelon for Alzheimer’s and Stalevo for Parkinson’s. Novartis (NVS) is employing several strategies to enter the field of Multiple Sclerosis therapeutics. Strategies employed include in-house development, early to mid-stage partnering and development of bio-generics
In-House Development
Novartis is developing FTY-720 and BAF312 as sphingosine-1-phosphate receptor agonists for MS and other indications. FTY-720 (fingolimod) is furthest along with a second Phase III trial (FREEDOMS) completed in July. There is a near term catalyst with topline results likely to be released to coincide with presentations on September 10, 2009 in Düsseldorf at the ECTRIMS Congress (European Committee for Treatment and Research in Multiple Sclerosis). FTY-720 has already reported details of a successful Phase III study (TRANSFORMS) in a large study comparing 2 doses of FTY-720 and Avonex (BIIB). The oral FTY720 showed a 38% reduction in relapses compared to the interferon and did well with all secondary clinical endpoints. However, safety issues were raised with 2 deaths due to severe viral infections and the presence of more skin cancers (including melanoma) in the FTY720 treated groups. In a clinical update in late April 2009 at the AAN meetings, no additional fatalities from other studies of FTY-720 in MS were mentioned.
Novartis is also developing BAF312 as a follow-up compound to FTY-720 with a similar mechanism of action. A Phase 1 study to investigate safety, tolerability, kinetics and dynamics was started last year and is likely complete. No details have been announced. Some of the perceived risks of FTY-720, might be class-related, and hence about the same in both agents. Other side effects, including bradycardia and shortness of breath might be improved in a follow-up agent as there are five known SIP receptors and the two agents likely have a different profile. The clinical effects of these agents may be through SIPR4 and SIPR5, with the smooth muscle and cardiac side effects likely occurring through SIPR1, SIPR2 and SIPR3.
Early to Mid stage Partnering
Earlier this year, Peptimmune (privately held) and Novartis entered into an agreement giving Novartis an exclusive option to develop PI-2301 for MS. This agent is similar to Copaxone (TEVA, poly[Y,E,A,K]n) in that it is copolymer of 4 different amino acids (in this case, poly[Y, F, A, K]n) and works through the same purported mechanisms via MHC class II binding and resultant stimulation and expansion of myelin-reactive Th2 cells. Two phase 1 studies have been completed and data from the second study in 50 patients with secondary progressive MS will be released at the ECTRIMS meeting later this month. Proof of concept was achieved in the first phase 1 study of 56 volunteers using an escalating dose cohort design. A phase II study in relapsing remitting MS may start by the end of the year. Though still too early to gauge the effectiveness of this product, safety should not be a concern. If as effective as Copaxone or more effective, it would offer an advantage of being a once a week shot over a daily.
Bio-generics
The generic approval process for biologic agents is more complicated than for simpler products and remains a work in progress at the FDA. Novartis obtained the right to its own branded version of interferon-beta-1b, Extavia, when it acquired Chiron. Schering, now part of Bayer (BAYRY.PK), had licensed its version of this compound, Betaseron, from Chiron, and continues to license Betaseron from Novartis. Extavia is identical to Betaseron, which has been available in the US and elsewhere since the early 1990’s. Betaseron offers modest efficacy and tolerability with an excellent safety profile. The FDA granted approval to Extavia in August 2009, allowing Novartis to market its first MS product.
Momenta Pharmaceuticals (MNTA) and its partner, the Sandoz unit of Novartis, submitted an application to the FDA on July 11, 2008 for generic glatiramer acetate (Copaxone, TEVA). No decision has been announced. Unlike Extavia, the approval process for generic glatiramer acetate will be more difficult as the product would be bio-similar, not bio-identical.
Market presence
Currently, Novartis’ only marketable product for MS is Extavia. Betaseron has a market share under 20% for the treatment of relapsing-remitting MS and has an excellent nurse educator program to enhance compliance and education for patients. Extavia may not enter the market at a significant discount to Betaseron. There will be no overwhelming desire to move patients from Betaseron to Extavia unless insurance programs shift more cost to those patients who prefer to stay on Betaseron. I do not think many patients on other interferons, such as Rebif (MKGAY.PK) or Avonex (BIIB) will switch without significant monetary disincentives from their insurers. Additionally, Biogen is developing a PEGylated Avonex that may only need to be taken once or twice a month that may be available by late 2013, further cutting into Extavia’s growth potential.
If Extavia has little potential for blockbuster status, why would Novartis be interested in this product? The main reason is that by marketing Extavia, Novartis will be building up a sales force and relationship with prescribing neurologists to allow a more successful launch for FTY-720 and other pipeline products for MS. I expect the FREEDOMS FTY-720 study to present excellent topline efficacy data against placebo later this month. Efficacy will likely be better than the current self-injectible medications but unlikely to be better than that of Tysabri (ELN, BIIB). The uncertainty about adverse effects will need to be addressed by the data release. If a reasonable balance of risk to efficacy is achieved, submission to the FDA could occur by the end of the year with fast-track approval by summer 2010. FTY-720 has potential to be a blockbuster and may have efficacy in other autoimmune disorders. If skin cancers remain a concern, a risk evaluation and mitigation strategy may need to be developed to protect against this adverse event. Unlike curent FDA approved medications for MS, FTY-720 is an oral agent. Cladribine (MKGAY.PK), another oral agent, is on a similar timeline for FDA approval.
Summary
Using all three strategies, Novartis should be able to step into the MS marketplace with Extavia and pick up the pace as FTY720 and other pipeline products are developed. Because of Novartis’ market cap over 100 billion, these products will only move the share price a few percent as they succeed or fail. However, they will need to have many solid products in the future to replenish the income from other agents facing patent expirations.
Update: BioMS
As predicted in last months post here and on seekingalpha.com (http://seekingalpha.com/article/148356-bioms-hype-or-hope), the Phase III, MAESTRO-01 study of 610 patients did not meet its primary endpoint of disability reduction and the stock price dropped precipitously. Other endpoints also did not meet significance. BioMS and Lilly decided to terminate MAESTRO-02 and MAESTRO-03 studies. Terminating the second Phase III likely means there was no suggestion of efficacy. One PR mentioned that the placebo group did not progress much, which would make a treatment effect harder to prove. Blaming the placebo group is easy. However, with 610 patients, the two groups should have been very similar and that sort of bias unlikely. Certainly, the possibility that all patients did not progress as much as expected could decrease the likelihood that a mildly effective medicine could not prove effects on disability progression. But, a moderate or very effective medicine would have still shown effect. One need only look at the Phase II Campath or Phase III cladribine studies aganins active comparator to see how moderate or strongly effective medications can prove efficacy even when comparator groups are less active than predicted. I haven’t seen details from MAESTRO yet but look forward to seeing other details, specifically average EDSS at entry.
I expounded more on the reasons that I felt the Phase III would fail at seekingalpha (see link above). Most importantly, the decision to proceed to expensive Phase III testing (1100 patients in 2 studies) was done prematurely based entirely on a very small study of 32 patients, further split into 20 patients with HLA-DR2 or DR4. These 20 patients were further split into 14 SPMS and 6 PPMS patients, a ridiculously small number of patients to try to analyze. No attempt was made to determine if every 6 month dosing was best by doing a larger and more controlled phase II looking at other doses. Maybe the medicine would have worked with monthly dosing. Second, the mechanism of action — tolerance induction — is unlikely as the MINDSET RRMS study failed to show an effect on inflammation, much more suscetible to tolerance induction than the degeneration of SPMS. I may have felt different about the prospects if a plausible MOA was advanced and supported by evidence. An additional reason I may not have mentioned is that 2 years may have been too short for the study.
As long as I’m rambling, I’ll add my two cents about small companies in Biotech and the decisions to proceed to phase III. BioMS could not have proceeded as far as they did without either a partnership, private equity or a second public offering. They rightly calculated that a product just starting Phase III would have more value than one that just has a 20 patient phase II. Initiating the MAESTRO studies allowed them to gain a reasonably good deal with Lilly, a company that could easily spare 97 million (87 + 10 milestone at 200 patient review) to have a shot at a 1 - 2 billion dollar a year drug. Thus, BioMS had nothing to lose by rushing the drug along. A larger company would have done a larger phase II forst and then decide whether to advance the drug to expensive phase III studies or just let the agent die. BioMS, with one drug, could not afford to let the drug die in Phase II. I’m going to keep this in mind when I look at other small companies that may have rushed a product along.
At this point the company has about 70 million dollars but will need to wind the study down,with payments to the CRO and investigators and likely has other payables. Thus,a price about 40 cents is probably appropriate. With the Phase III stopped, I do not think there will be any further study with this drug.
Update: Cypress BioScience
Decided last week to get out of this position. The stock had a nice run from low 7’s to low 9’s and I was able to get out for a 30% gain.
Reasons I got out:
1. I was having mixed results with Savella, based on initial patients, tolerability appeared worse than Cymbalta. Some patients did very well but many had nausea or other side effects. I will still use Savella but probably not as frequently.
2. Though Savella sales are continuing to grow, they are not exploding (data from Yahoo MB but looks like source is reliable).
3. Managed care will likely put more restrictions on branded products as health care reform proceeds.
4. I feel the market will generaly decline the second half of this year, though we may still have one final rush up. Biotech and Pharma, already up a lot this year, may not participate inthe final rush and I’m happy with 30%. $13-$15 is certainly possible as sales escalate, but less likely than stagnation in 7-10 range if market declines later in year.
BioMS: Upcoming Binary Event
Large price swings frequently occur when Biotech companies release data from a Phase III study.. If primary endpoints do not reach significance, NDA submission to the FDA is unlikely. Strong results with good safety data bring the drug one step closer to approval. Narrow hits and misses or safety concerns result in continued uncertainty until longer term data or a second phase III becomes available.
These can be binary events for smaller companies that have only one lead candidate, often resulting in price drops greater than 50% or increases greater than 200%. Larger companies, not dependent on one or two candidates, are unlikely to see price changes greater than 10-20 % after release. A lag period of 1 to 3 months is common between study completion and topline data release. More complete data is typically presented at conferences 3 to 6 months later. Before phase III data is released, bets on a company’s future are placed based on the more limited results from earlier phase II studies.
BioMS
BioMS (MS on Toronto Exchange and BOMSF.PK on OTC) completed their study of MBP8298 (also known as dirucotide) in late May 2009 and topline data from the MAESTRO-01 study will be released shortly. This study compared dirucotide to placebo in 553 patients with secondary progressive multiple sclerosis. The primary endpoint compared the time to sustained worsening of disability using a standard scale. From Phase II studies, dirucotide appears to be a very safe medicine and needs to be infused only twice a year. Secondary progressive multiple sclerosis (without relapses) has no current treatment and FDA approval for this indication could lead to sales over 1 billion/year.
BioMS and Lilly (LLY) entered into a global licensing and development agreement on December 17, 2007 regarding dirucotide. BioMS received upfront payments of $87 million and will have potential development and sales milestones of $410 million and possible further escalating royalties.
Multiple Sclerosis
With about 400,000 patients, MS affects more than 1 out of every 1000 Americans. About 85% present with relapsing remitting MS (RRMS) and half of these eventually develop secondary progressive MS (SPMS). The other 10-15% present with primary progressive or progressive relapsing multiple sclerosis. Thus, there are about 160,000 patients in the US and Canada with SPMS. There are currently 6 FDA approved medicines for MS: Betaseron (Bayer;BAYRY.PK), Avonex (Biogen Idec, BIIB), Copaxone (Teva), Rebif (Merck KGaA, MKGAY.PK), Novantrone (Merck KGaA and generic) and Tysabri (Elan and Biogen, ELN, BIIB). These treatments have efficacy for RRMS and some also for the relapses of early SPMS. However, there are no current medications that are FDA approved for non-relapsing SPMS.
Clinical Studies
We have data from two phase II studies, one small study involving 32 patients with progressive MS and another involving 218 patients with relapsing remitting MS. In the first, smaller, study, there was no significant difference in disability progression between placebo-treated and dirucotide-treated patients. However, when the 20 patients who had at least one copy of either MHC HLA DR2 or DR4 were evaluated separately, there was a significant difference in disability progression between treated and untreated patients. The MINDSET-01 phase II study in RRMS enrolled 70% patients with either HLA DR2 or DR4 and the study planned to examine all enrolled patients and this subset of patients separately. They were followed for 15 months. The primary endpoints of relapse rate reduction and MRI changes were not met in either the whole population or the HLA DR2/DR4 subgroup. However, the secondary endpoint of reduced disability progression was met in the treated HLA DR2/DR4 subgroup. Patients with one or both of these genes comprise 50-70% of MS patients. Thus, there are about 100,000 potential secondary progressive MS patients in the US and Canada and about 200,000 worldwide.
MAESTRO-01 and MAESTRO-03 are similar studies in the EU and US, respectively, with > 500 patients each. Secondary progressive MS patients, who have at least one copy of either DR2 or DR4, were enrolled and treated with either 500 mg IV dirucotide every 6 months or placebo. MAESTRO-01 ended in late May and MAESTRO-03 will end in late 2010. As BioMS is currently trading at $2.35 and has a market cap of about 215 million dollars (252 million CAD), the topline data release shortly will lead to a huge swing —- but in which direction?
Hints from prior studies
The purported mechanism of action for dirucotide is immune tolerance induction. This implies that repeated infusions of this molecule should induce a potent anti-inflammatory response by down-regulating the immune system’s attack on myelin. However, inflammatory MRI changes were not reduced in the MINDSET study. Also, relapses, that should have been reduced by tolerance induction, were unaffected by dirucotide. Could there be another non-inflammatory mechanism of action that is more specific to the secondary progressive degenerative process? That is possible, but neuroprotection is hard to prove and difficult to establish in a one or two year study. It is just as likely, or more likely, that the effect on disability noted in the first phase II study was random chance in a small subgroup of only 20 patients. Thus, I believe there is a high likelihood that dirucotide will miss its primary endpoint in the MAESTRO-01 study and that its stock price will drop by half or more. Absence of a trend would doom the product and BioMS. Lilly would just take a small hit, write off the program and move on.
Hopefully, I’ll be proven wrong as SPMS patients who no longer have relapses need a safe medication that can retard disability progression. If MAESTRO-01 hits with great numbers, then the similar MAESTRO-03 study should also meet endpoints and submission to the FDA could occur in early 2011. If that occurs and the drug is approved, dirucotide could be used by up to 50,000 SPMS patients in the US and Canada within two to three years of approval and a similar number in the ROW. That would lead to sales well over 1 billion a year. Even after LLY takes the lion’s share, milestone and royalty payments would drive BioMS share prices to impressive gains. If one believes that the trial has a better chance of succeeding than I do, then the ‘pot odds’ would support an investment decision. I’ll sit this one out.
Cypress Bioscience: Savella for Fibromyalgia
Cypress Bioscience, Inc (CYPB) is a small company with a 287 million dollar market cap that will soon begin to receive royalties for its first FDA approved product. Savella (milnacipran) was approved on January 14, 2009 by the FDA for the management of FMS (Fibromyalgia Syndrome) and became available at pharmacies on April 28. This drug will be marketed by Forest Laboratories (FRX), a company that already has a large sales force marketing Namenda for Alzheimer’s and Lexapro for depression and generalized anxiety.
Fibromyalgia
Patients with fibromyalgia syndrome experience pain, fatigue, poor sleep and other symptoms. Most experience mood disorders as well. Diagnosis remains difficult due to the absence of laboratory and imaging abnormalities. Therefore, identification of the syndrome is by history and physical exam. Up to 2% of the population, with a 9:1 ratio of female to male, meets official criteria for FMS, with most people presenting in young adulthood. Thus, there are up to 6 million people in North America with this disorder.
Because of the difficulty in proper diagnosis and the interplay between psychological and physical symptoms, FMS is often difficult to treat. Though many small studies in the past showed the benefit of some older drugs like the tricyclic antidepressant and anti-epileptic agents, larger studies were not performed before this decade. Over the past 2 years, Lyrica (Pfizer, PFE) and Cymbalta (Lilly, LLY) received FDA approval for FMS, making Savella the third drug to receive this indication.
Many patients with FMS receive some benefit from drugs that work through central pain pathways. The ‘gate control’ theory of pain asserts that pain may be modulated in the spinal cord, before being appreciated in the brain. Older tricyclic antidepressants, such as amitriptyline, imipramine and nortriptyline, work through multiple pathways including serotonin and norepinephrine reuptake inhibition and have been used off-label for years for FMS and other pain disorders. Unfortunately, side effects are common with these drugs, especially when used at higher dose. Cymbalta (duloxetine) was the first true dual serotonin and norepinephrine reuptake inhibitor marketed and was first approved for depression, then for anxiety, diabetic peripheral neuropathic pain and most recently for FMS. Norepinephrine, transmitted to the gray matter of the spinal cord from neurons in the brainstem, lingers longer at the synapse when reuptake is inhibited by medications such as Cymbalta or Savella. The role, of serotonin in pain is less defined. Lyrica, a medicine first approved for epilepsy, and also approved for post-herpetic neuropathy and diabetic neuropathy, was the first medication approved for FMS in 2007. Among other actions, Lyrica works through a calcium channel that can modulate pain.
Savella
Where does Savella fit in? Savella inhibits norepinephrine and serotonin reuptake in a 3:1 ratio, where Cymbalta has this action in a 1:9 ratio — thus, Savella is more balanced in this regard. Theoretically, Savella should be more potent for pain modulation than Cymbalta. Reviewing the studies, side effects appear to be fairly equal, though I have not yet had the opportunity to get feedback from patients who have been prescribed Savella over the last two weeks. Cymbalta has become a blockbuster with over 2 billion in sales last year. There is huge potential for Savella to get a very large slice of this market. Because of the favorable norepinephrine and serotonin reuptake ratio, Savella may be a better agent for FMS, though no comparison study has yet been done. Additionally, it should be effective for neuropathic pain and depression. It has been used as an antidepressant in Europe and Japan.
Savella will have several advantages and disadvantage, compared to Cymbalta. Its higher ratio of norepinephrine and possibility of higher efficacy will be the main advantage. Its disadvantages include being a twice a day drug (vs. Cymbalta at once a day). Compared to the generic tricyclics, Savella will be much more expensive but have a better side effect profile. Lyrica has a different mechanism of action and will be less affected by the introduction of Savella than Cymbalta. Further competition with other new agents will be avoided as Pfizer recently halted development of esreboxetine, an enantiomer of the anti-depressant reboxetine (sold in Europe) that has an even higher ratio of norepinephrine to serotonin reuptake inhibition. I expect both Lilly and Pfizer to ramp up their marketing for their products for FMS.
Cypress Bioscience, Inc.
Cypress initially licensed milnacipran from Pierre Fabre Medicament of France for FMS and related chronic pain syndromes in North America in 2001 for payment, landmark payments and royalties. The arrangement was modified in 2003 to allow development and sales of milnacipran for any indication in North America in exchange for shares of stock and warrants. In 2004, they entered into collaboration with Forest Laboratories to further develop milnacipran and to market it. This was in exchange for upfront and milestone payments. Forest assumed future development costs for milnacipran and Cypress will receive royalties. Outside North America, Pierre Fabre controls the drug.
Cypress has a very strong balance sheet with 142 million in cash and negligible debt. Many details of the Forest and Cypress agreements have not been made public and it is unclear how much of each dollar in sales will go to Cypress. Besides Savella, Cypress has a personalized medicine division with a product, Avise PG, to help determine response to methotrexate in rheumatoid arthritis patients.
The Bull Case
Pulling the trigger to buy CYPB was difficult for several reasons. First, the absence of details regarding royalty payments to Pierre Fabre and receipts from Forest Laboratories is unsettling. However, analyst reports have estimated a 15% royalty on sales. Second, I feel Cypress (and now Forest) has done a poor job developing a potentially very useful drug. Why have they not initiated neuropathy and depression studies? When I first followed Cypress in 2005, I decided not to buy because I disagreed with their plan to go for a FMS indication from the FDA before going for the easier depression and neuropathic pain indications. Clinical studies with depression and neuropathy would likely have yielded cleaner studies than FMS and might have led to approval of Savella one or two years earlier. For years, FMS has been treated with off-label drugs and approval for another indication would still have led to its use in this indication. Also, I’m not a big fan of the brand name — sounds too much like saliva. So why buy now? Sales for Savella could easily reach 500 million. If the 15% figure is correct, royalties could reach 75 million/year by 2012. Additionally, if sales ramp up quickly, Cypress will be a takeover target, most likely by their partner, Forest. Though most of Savella’s use will be for FMS, some will used be off-label for other types of pain and depression. Furthermore, from a technical standpoint, CYPB has consistently bounced off of its 200 day moving average this year and appears to be forming a reverse head and shoulder pattern on long-term (weekly) charts. Downside risk seems limited to $5.00 as this value has help up twice in the past 4 years and share price is supported by almost $4 per share of cash and a low burn rate.
Cladribine Update
Further details of the CLARITY study comparing cladribine to placebo and further safety from the extension data was presented yesterday at AAN. This adds to the review of the topline data contained in the 4/19/09 “Will the new oral MS therapies be a hard pill to swallow” post. This study of 1326 patients over two year had 3 groups: placebo (PBO), low dose 3.5 mg/kg (LD) and high dose 5.25 mg/kg (HD). The primary endpoint was annualized relapse rate (ARR) reduction. The placebo arm had an ARR of 0.33 compared to 0.14 and 0.15 in the LD and HD groups, respectively, representing a 55-58% reduction. Relapse free proportion of patients were 61%, 80% and 79% in the PBO, LD and HD groups, respectively. New data relewased at AAN includes a 33% and 31% reduction in disability progression, compared to placebo, in the LD and HD groups, respectively. MRI T1 Gadolinium positive lesions were reduced 86-88% in the treated groups compared to the placebo group.
The drug was generally tolerated well. However, lymphopenia and leukopenia were seen about 10 times as commonly in the two treatment groups (2% placebo and 21% LD and 31% HD, compared to controls. Additionally, there were cases of zoster in the 2 treated groups (2% of each group) and none in placebo. There were 6 deaths during the study, 2 in each groups. One placebo group death was due to hemorrhage and one to suicide. One LD cladribine group death was due to myocardial infarction and one to pancreatic cancer. One HD cladribine group death was due to cardiac arrest but also to pancytopenia and tuberculosis, while the other death was due to drowning. There were 4 malignancies in the cladribine groups during the study and none in the placebo group. These cancers were cervical, ovarian, malignant melanoma and pancreatic death (this one leading to death). In the extension study a case of choriocarcinoma occurred and was treated.
Analysis of the new data: The excellent ARR efficacy data was confirmed with further significant differences in reducing disability progression and reducing MRI lesions. Efficacy is at least as potent as the interferons, and probably a little better. Tolerability is good. The verdict is still out on safety with 5 malignancies being reported. They are all in different organ systems so no clear trend has arisen. But, the lack of an ‘organ at risk’ will make cancer screening difficult. Lymphopenia/leukopenia was very common but may be a manageable event. It is possible that dosing intervals may be extended in patients with low counts. Thus, the drug could be taken at 6 month intervals when blood counts have returned to normal and held for several months longer, until counts improve, for other patients. This could impact revenue models, assuming the drug is approved and adjusting for lymphopenia or leukopenia is suggested. Given that the HD was no more effective than LD and the HD had more lymphopenia, I would expect the lower dose to be the recommended dose for FDA review. The bottom line is that efficacy and convenience will need to be balanced with possible increased risk of cancer and infection (zoster) and possible risk of prolonged lymphopenia. The new data has not changed my opionions from 4/19/09
FTY-720 update
Further details from the 1280 patient TRANSFORMS data comparing 2 doses of FTY-720 to Avonex were presented this afternoon. Details about FTY-720’s MOA and earlier results are in the 4/19/09 “Will the new oral MS therapies be a hard pill to swallow” post. A little more flavor on the patient characteristics was provided. At study entry, 40% had at least one Gd positive lesion and 55% had been on at least one other MS treatment, though it needed to be washed out for 90 days. As previously posted, the Annualized relapse rate (ARR) was 0.33 with AAvonex, 0.16 with low dose FTY-720 (0.5 mg a day;LD) and 0.20 with high dose FTY-720 (1.25 mg a day; HD). Relapse free rates over the one year study were 69% for Avonex, 83% for LD and 80% for HD. COmbined unique new lesions or expanding lesions on MRI were 2.1 for Avonex, 1.5 for LD FTY-720 and 1.4 for HD FTY-720. Disability trends favored FTY doses over Avonex, though significance was not met. There were two deaths in the FTY-720 group, one with disseminated herpes zoster and one with herpes encephalitris. The Herpes Zoster case was also on prednisone and I overheard a poster presenter stating that the patient had not had Chicken Pox as a child. The herpes encephalitis case apparently had delayed anti-viral treatment. Malignancies were 2 with Avonex (1 Basal cell and 1 squamous cell) while the LD FTY cohort had 3 Basal cell, 3 melanoma and 2 breast cancers and the HD group had 2 Basal cell and 2 breast cancers. Of note, 8/10 of the skin cancers were detected at the first dermatologic visit. In a separtate poster, four year follow up of the Phase 2 patients show continued improved ARR and the inital placebo patients who were randomized also had imporved ARR.
Analysis: FTY-720 showed good efficacy against an active comparator with significant ARR and MRI data. The absence of significant disability data from this one year comparison study is fine, because it was against Avonex and not against placebo. However, efficacy in this measure will need to be demonstrated when FREEDOMS is reported later this year or early next year. In this study, no new deaths occurred during the additonal several months of follow-up. The additional data helps to predict that the FREEDOMS study will show highly significant efficacy data against placebo for ARR, disability and MRI. To my knowledge, no deaths have occurred in FREEDOMS. Malignancy data would not be expected to be released until the study is complete.