We have updated our Insomnia Indication Report and revenue models. Since our original report in December 2004, both Lunesta (Sepracor-SEPR and Sanofi-SNY) and Ambien CR (Sanofi-SNY) have been approved, and the insomnia market has seen substantial growth. Still, while a third of the population have a sleeping problem and half of them consider the problem to be serious, only 36% of patients who see their insomnia as severe enough to try some type of treatment (which could also be OTC medications or psychobehavioral therapy) use prescription medications.

In 2005, Lunesta experienced a strong launch, with $329 million in revenue for the year. The market leader, Ambien (Sanofi-SNY), held steady in terms of total prescriptions, though its share was diluted by market growth. With the launch of Ambien CR, Sanofi stepped up its sales details, and Lunesta’s growth was curtailed.

Ambien had become the market leader due to better safety than the benzodiazepines, the traditional insomnia medications that are more subject to drug abuse. However, Ambien’s label states it should not be prescribed more than a month at a time (though many physicians do) and its short half life makes it ineffective for a number of patients in maintaining sleep through the entire night. Ambien CR has a longer duration of action, but on certain measures appears weaker than Lunesta, and unlike Lunesta its labeling notes that its usage has only been supported by short-term trials. While these features help or hinder the drugs to some extent, marketing factors, such as brand loyalty for Ambien and physician sales visits, will play a major role in this market.

On the horizon are short-acting and longer-acting formulations of Indiplon (Neurocrine-NBIX, Pfizer-PFE, DOV-DOVP), with a PDUFA on May 15, 2006. Indiplon binds to the same site as Ambien, but with a much higher affinity. Given what we deem to be a time of caution at the FDA, and the fact that the two formulations are separate NDAs that will now be decided together, a delay in approval is possible. However, we do not foresee significant issues holding up Indiplon approval. While the recent press regarding Ambien patients and odd parasomnias (such as driving while asleep) heightens uncertainty, Neurocrine has confirmed that these have not been seen in the Indiplon clinical trials.

Differences do exist between these drugs; such as in their half-lives. This makes different drugs more suitable for different insomnia treatment settings. Generally insomniacs can be split into three groups: 1) Those who strictly have problems falling asleep (trouble with sleep onset - estimated at 25% of insomniacs); 2) Those who occasionally awake in the middle of the night (MON - estimated at 10% of insomniacs); 3) Those who have other types of sleep maintenance problems with or without difficulty falling asleep (full night maintenance - estimated at 65% of insomniacs).

A drawback to Lunesta is an unpleasant taste for a high proportion of patients (17-34%). While at first an unpleasant taste does not sound like a significant issue, patients have described the taste as lasting well into the next day and significantly impacting the taste of food and drink. All else being equal, we think it would be more convenient for physicians to prescribe Indiplon first, to prevent dissatisfied patients requiring a switch in medications.

For patients who only have difficulty falling asleep, we think the major battle is between Ambien and Indiplon IR. We project Ambien to retain majority share until generics enter the market, likely in 2007. In patients who occasionally awake in the middle of the night, we believe that Indiplon IR will become the dominant drug, due to its significantly better profile than Ambien (which is more likely to leave patients drowsy in this setting).

Finally, for a large block of patients with other types of sleep maintenance problems, the main battle will be between Ambien, Ambien CR, Lunesta, and Indiplon MR. While we think Indiplon has the most desirable profile (though Lunesta’s longer half-life may make it more effective for some patients), Ambien’s strong historical presence and sales relationships will help it transition patients to Ambien CR and the family will maintain majority share until the advent of generics. For new patients entering the market, we think Indiplon over time will become the most popular drug for this segment, helping it to eventually edge out Lunesta in total market share in this segment, and come close to Ambien CR’s position.

Overall we see the Indiplon franchise becoming the leader in terms of revenue, with the Ambien franchise next, followed by Lunesta. That said, the insomnia market is very large and growing and each drug offers blockbuster potential to their respective companies.

BACKGROUND

Approximately one third of American adults report that they have sleep problems, and half of them consider their sleep disorder significant enough to try different methods of treatment, such as psychobehavioral therapy, OTC agents such as Benadryl, herbal remedies, and conventional pharmacotherapy.

The etiology of insomnia is poorly understood, as is what makes particular individuals predisposed to the condition. In about 60% of insomnia patients, the condition is likely to have been precipitated by underlying diseases. These diseases include anxiety disorders (24%), depression (8%; higher rates such as 30% have also been reported, though in surveys it is not always clear which disorder came first), other mental disorders (10%), general medical conditions (7%), and drug related comormidities (2%).

Insomnia may be transient (e.g. a person facing acute stress for several days), short-term (usually related to situational stress), or chronic (definitions regarding the length of symptoms vary between 1 and 6 months). It may involve difficulty initiating sleep, difficulty in maintaining sleep, waking up too early, or any combinations of these.

Insomnia studies may use either subjective patient measures of sleep or a sleep laboratory test that assesses physiologic variables (polysomnography, PSG). Some terms used in insomnia studies include sleep latency (time it takes to fall asleep, a related PSG measure is latency to persistent sleep), wake after sleep onset (total time awake from onset of sleep to final awakening), and total sleep time (the total minutes of sleep). Data will differ depending on whether measures are made by the patient or in the sleep lab.

For modeling purposes we have separated the market into three categories: 1) Those who strictly have problems falling asleep (trouble with sleep onset - estimated at 25% of insomniacs); 2) Those who occasionally awake in the middle of the night (MON - estimated at 10% of insomniacs); 3) Those who have other types of sleep maintenance problems with or without difficulty falling asleep (full night maintenance - estimated at 65% of insomniacs).

CURRENT TREATMENT

Since our original 2004 indication report, both Lunesta (Sepracor-SEPR and Sanofi-SNY) and Ambien CR (controlled release, Sanofi-SNY) have been approved and have joined Ambien and Sonata on the market.

Since being approved in 1992, the regular formulation of Ambien (Sanofi-SNY) has become the most widely prescribed treatment for insomnia, taking share away from the benzodiazepines, largely due to their risks of abuse, dependence, and next day residual effects. Ambien’s better safety profile is because it specifically targets one site of the GABA-A receptor, whereas the benzodiazepines can bind at several sites. However, since the benzodiazepines can be better than Ambien for treating sleep maintenance and are also generally effective against anxiety, which affects a large segment of insomniacs, they have retained a significant share (15+%). Ambien’s patent will expire in October 2006, though a pediatric extension is expected to block generic competition until April 2007.

Ambien’s weaknesses include the fact that, due to its short half-life, it is insufficient for sleep maintenance in a number of patients (30+%). It can also have some next day residual effects, and its label states it should not be prescribed for over a month at a time. While physicians have been treating patients for longer periods, this still remains a concern for some (partly for liability reasons). It is also labeled a controlled substance by the DEA (Schedule IV), meaning there is potentially some, albeit low, risk for dependence.

Finally, Ambien has recently seen a lot of negative press around side effects, including driving and eating while still asleep (parasomnias). The most widely publicized reports came out in March and not only impacted Ambien’s market share, but resulted in a flattening of the fast growing insomnia market as a whole. Sanofi has since combated these reports to the FDA, by claiming that people with insomnia who are untreated also have been reported to show these types of strange behaviors. To date the FDA has been very open to Sanofi’s argument, but it is still too early to tell if there will be a significant long-term impact to Ambien’s or other drugs’ shares.

Another non-benzodiazepine, Sonata (King-KG), has a shorter half-life than Ambien. While this makes it more suitable for some situations (such as dosing in the middle of the night), it is not effective enough for most patients and hence has not been popular.

Lunesta was approved in December 2004, and fueled by its marketing efforts (1500 sales reps and an expensive consumer ad campaign), saw strong market uptake, growing revenues to $329 million in 2005. Lunesta’s share growth was curtailed in September 2005, when Ambien CR was approved and Sanofi stepped up its own marketing, using 3500 sales reps to detail Ambien CR. Lunesta’s uptake has primarily been due to an increase in the insomnia market size, rather than from significantly reducing the usage of other drugs. Specifically, Lunesta did curtail Ambien’s growth, but Ambien prescriptions stayed at a constant level (and since Lunesta grew the market by adding patients, Ambien’s overall market share declined).

Ambien CR’s growth has similarly been largely due to overall market growth, though we estimate that approximately a third of its growth came from a switch from regular Ambien prescriptions. Ambien CR has almost caught up with Lunesta, with both having around 11% of the total prescription share.

Another major competitor is generically available trazodone, which is used off-label and has about a 19% share (usage has been fairly stable, with only a small reduction over the last year). Trazodone is actually an anti-depressant, and despite multiple side effects (dizziness, dry mouth, nausea, blurred vision, and potential increased risk for arrhythmias) is used in insomnia because it does not have significant abuse or withdrawal issues (also, a sizeable number of insomnia patients have depression).

Finally, Rozerem (Takeda), which stimulates melatonin receptors thought to be involved in the sleep-wake cycle, was approved in July 2005, but has not been as popular as the other novel entrants: though it has the advantage of not being a controlled drug, it is primarily useful for sleep initiation and not maintenance, and it has not demonstrated efficacy as strong as that seen for other options.

DRUGS IN DEVELOPMENT

Indiplon Capsules (IR) and Tablets (MR) - Neurocrine (NBIX), Pfizer (PFE), and DOV (DOVP)

Indiplon comes in two forms: an immediate release (IR) short acting capsule, for help with sleep onset and night wakenings, and a modified release (MR) longer acting tablet, for both sleep onset and maintenance. Indiplon binds to the same GABA-A site as Ambien and Sonata (the specific site for Lunesta is not known), but with a much higher affinity. Indiplon has demonstrated improvement in sleep onset similar to other drugs on the market. In a middle of the night dosing study, Indiplon IR did not lead to sleepiness four hours after dosing, whereas Ambien did. While the data released are limited, the MR formulation has shown good efficacy for both sleep onset and sleep maintenance. More details are listed in the table below, comparing Indiplon to current drugs.

Indiplon has an upcoming PDUFA on May 15, 2006. Given what we deem to be a time of caution at the FDA, and the fact that the two formulations are separate NDAs that will now be decided together, a delay in approval is possible. However, we do not foresee significant issues holding up Indiplon approval. While the recent press regarding Ambien patients and odd parasomnias heightens uncertainty, Neurocrine has confirmed that these have not been seen in the Indiplon clinical trials. In addition, Sanofi has asserted to the FDA that parasomnias do occur in patients with insomnia and may not be drug-related.

Silenor - Somaxon (SOMX)

Silenor, a low dose formulation of the anti-depressant doxepin, recently reported results from a phase III study. While the results for sleep maintenance measures were as good as other agents on the market, since the study focused on patients with sleep maintenance problems, the results are not necessarily directly comparable. Also, latency to persistent sleep, a measure of sleep onset, was not statistically significant compared to placebo after the initial period of the study, due to improvements in the placebo group. If further results confirm this, the drug might still be valuable for the 30% of patients who only have sleep maintenance and not sleep onset difficulties. This is particularly true because it is not a controlled drug with an addiction potential.

Gaboxadol – Merck (MRK) and Lundbeck

Gaboxadol is in Phase III development. Though it is a GABA-A agonist like Ambien, it demonstrated fairly modest efficacy in a Phase II trial. Though further testing may show better results, given the other options on the market or in development, we are pessimistic about Gaboxadol’s prospects.

The table below lists characteristics of selected drugs, which will be the primary near term competitors in the insomnia market.

MARKET OUTLOOK

Based on a national survey, approximately 6% of the US population uses prescription medications for insomnia. However, this is only a small portion of the 17% of the population who believe their sleep disorder is serious enough to try some type of treatment (which could also be OTC medications or psychobehavioral therapy). With Lunesta’s and Ambien CR’s large sales forces and improved features over other options, a number of new patients entered the US prescription market over the past year, expanding it by approximately 25%.

While it appears market growth has flattened since the negative reports around Ambien have surfaced, we believe it will re-accelerate after Indiplon’s approval. Our belief on the market growth is based on the ever-increasing sales efforts going on between the major players. The marketing battle will be very interesting, as it does appear that sales visits play a large role in new prescriptions. Sepracor seemed to recognize this, as it has recently added 450 new sales reps and re-concentrated its sales focus on doctor visits rather than consumer marketing. Pfizer will soon bring its massive resources and marketing expertise to the party, should Indiplon be approved by May 15. Given Pfizer’s strength, it should be able to make inroads on the well-established Ambien and fairly-new-to-the-market Lunesta. An intriguing issue will be the brand loyalty Ambien has built up over the years, as many people may elect to try Ambien CR over Lunesta or Indiplon. Throw into the mix generic Ambien coming to the market in 2007 (it is unclear, but the cheaper price of trazodone may play a role in the share it maintains), and the market becomes a very difficult one to predict in the short-term.

The different drugs have features which make them more or less suitable for particular types of insomnia. As such, we have chosen to separate the market in three categories to better understand each drug’s longer-term potential in the market.

1) Those who strictly have problems falling asleep (trouble with sleep onset).
We estimate 25% of insomnia patients treated with pharmacologic drugs exclusively have sleep onset problems. For sleep onset, most of the drugs have similar efficacy, and we believe the shorter acting ones are likely to be the primary contenders. Indiplon IR has a better profile than Ambien, but Ambien appears to have strong brand loyalty. Hence we think in this segment, Indiplon IR will gain a minority share, with growth from new users as well as some switching from Ambien. We believe that when Ambien is available as a generic, it will moderate Indiplon IR’s growth in this segment.

2) Those who occasionally awake in the middle of the night (MON).
For patients who occasionally wake up in the middle of the night (we estimate 10%), we believe Indiplon IR will be the leader with its marketing focus on this segment and better performance over Ambien. Sonata is likely to retain its small degree of share as well. Patients who wake up in the middle of the night on a frequent and regular basis are likely to use the drugs better suited for sleep maintenance to help prevent waking up in the first place.

3) Those who have other types of sleep maintenance problems with or without difficulty falling asleep (full night maintenance).
For the large remaining segment (65%) of patients with sleep maintenance problems, who may or may not also have sleep onset difficulties, we believe Indiplon has the best product profile, though Lunesta’s longer half-life is likely to make it more helpful for certain patients. The main drawback to Lunesta is an unpleasant taste for a high proportion of patients (17-34%). While at first an unpleasant taste does not sound like a significant issue, patients have described the taste as lasting well into the next day and significantly impacting the taste of food and drink. All else being equal, we think it would be more convenient for physicians to prescribe Indiplon first, to prevent dissatisfied patients requiring a switch in medications.

On the other hand, Lunesta does have data showing an improvement in depressed patients that may help it penetrate into trazodone’s market share, which has remained stable. However, we think much of the near term share in this segment will be decided by sales efforts, with Sanofi’s Ambien having the advantage of strong brand loyalty and established relationships with physicians. Pfizer will have an advantage, particularly over Sepracor, due to its strong and sizeable sales force.

As marketing efforts intensify and physicians become more familiar with the advantages of the longer acting drugs, we see significant switching from Ambien, particularly to Ambien CR and Indiplon. However, we think Ambien will still retain much of its share, until the availability of generics, which will also moderate the switching from Ambien. We also project that Indiplon will erode Lunesta’s share of new patients and become the preferred option, though we do not see a great deal of switching from current users of Lunesta.

REVENUE MODELS

We have updated seven revenue models within the insomnia market, added a Silenor model and discontinued the Gaboxadol model, as we feel Gaboxadol's potential is limited should it come to market. In addition, we have chosen not to model the markets outside the US, as the economics there are poor for insomnia drugs and even if Pfizer or Sepracor decide to promote their products outside the US, the revenue contribution would not be particularly significant.

Ambien - Sanofi-Aventis (SNY)

We have modeled branded Ambien to lose share and have a dramatic price reduction once generics come to market in 2007. We do however expect Ambien to retain a modest share, particularly among patients with sleep onset problems.

Ambien CR - Sanofi-Aventis (SNY)

Patients should continue to switch from Ambien to Ambien CR as Sanofi works out reimbursement issues. We expect the switch to primarily take place in the sleep maintenance segment. Ambien CR will likely benefit from Ambien’s brand familiarity and maintain a solid revenue flow to Sanofi.

Ambien Family Projection

We are currently projecting Ambien and Ambien CR to achieve 2006 revenues of $1.92 billion, of which we feel CR will contribute $568 million. After Ambien goes off-patent combined revenues will never reach the same level. We are currently projecting peak Ambien CR U.S. revenue of $1.37 billion.

Lunesta - Sepracor (SEPR) and Sanofi-Aventis (SNY)

We believe Lunesta will primarily be used in the sleep maintenance portion of the market, though we have modeled for a small degree of share in the sleep onset segment. Our model assumes Lunesta will start losing its share of new patients once Indiplon is approved. We also feel that some patients who no longer want to endure Lunesta’s unpleasant taste will switch to Indiplon.

While we project that Lunesta’s growth will slow after 2006, the overall growth in the insomnia market should help expand the drug’s annual revenue to over $1 billion. We are currently projecting 2006 revenues of $654 million and peak U.S. revenue of $1.21 billion.

Indiplon MR - Neurocrine (NBIX) and Pfizer (PFE)

Once approved, we expect Indiplon to start taking new patient share from Lunesta in the sleep maintenance setting, as well as some from Ambien, and become the preferred choice. However, because there are no real clear efficacy advantages for Indiplon and Lunesta’s longer half life may be attractive, we feel that the difference in share will not be dramatic.

Indiplon IR - Neurocrine (NBIX) and Pfizer (PFE)

We project Indiplon IR will dominate the MON segment of the market -- but this represents only an estimated 10% of patients. We also foresee some uptake in the sleep onset segment.

Indiplon Family Projections

We expect the combination of Indiplon MR and IR to achieve $94.3 million in 2006 and peak US revenue of $2.44 billion, becoming the largest revenue-generating franchise in the insomnia market.