Asthma is a disease involving reversible constriction of the airways, leading to difficulty breathing. It affects 14 million adults and 6 million children in the US. Its underlying cause is inflammation of the airways, and in the majority of cases the inflammation is caused by an allergic response to common inhaled allergens. The mainstays of treatment are anti-inflammatory agents, such as inhaled corticosteroids (ICS), and drugs that dilate the airways (bronchodilators).

Currently, the market leader for inhaled asthma maintenance treatment is Advair (GlaxoSmithKline - GSK), the only combination available in the US of an ICS and a long-acting beta agonist bronchodilator (LABA). Its LABA component, Serevent (GlaxoSmithKline - GSK), has come under scrutiny due to an increased number of deaths in the SMART trial, and even though this occurred mostly in patients not taking an ICS, Advair has also received a warning label. Until it is proven that the danger is only for the particular drug Serevent, other LABAs are also under suspicion. Nevertheless, since there are no good substitutes for the combination and since many physicians (in contrast to the FDA) thus far view the LABAs as safe when given with ICSs, we believe it is unlikely that Advair will lose significant share in the near term.

Another similar combination inhaled product, Symbicort (AstraZeneca - AZN), is already approved in Europe and is in NDA status in the US. While approval is likely to be delayed, given the low percentage of inhaled drugs that make it their first time through, we think it will be approved next year. While currently asthma maintenance medications are given according to a fixed dose schedule, Symbicort has applied in Europe for a flexible dosing regimen (called Symbicort Maintenance and Reliever Therapy [MRT]), where the number of inhalations is matched to patient need in order to reduce exacerbations. It is not clear whether MRT dosing was also filed in the US, but even if it were, we think that given the concerns from the SMART trial, MRT will not be approved without a better understanding of the risk with LABAs or more trials on Symbicort. However, the balance of evidence so far is in favor of MRT, and we think it eventually will be approved. (We have modeled this as our most likely scenario, but have included models in case MRT is not approved.) It is possible, though, that if further studies are required, AstraZeneca will wait to see if the fallout from the SMART trial will give fixed dose Symbicort an edge over Advair. Alternatively, if Symbicort MRT is approved in Europe first, AstraZeneca may not want to pursue further trials in the US, since negative results could jeopardize their marketing of MRT in Europe.

MRT would put Symbicort in a leadership position, since certain pharmacokinetic properties of Advair make it unsuitable for flexible dosing. Another combination drug that could work with MRT, Formoterol-Ciclesonide (Altana - AAA and Sanofi - SNY), is likely to do well because of the favorable profile of the ICS Ciclesonide, though we have lingering concerns about its ocular side effects pending release of results for a major safety trial. Other combination drugs that could work with MRT, namely Formoterol-Mometasone (Novartis - NVS, Schering Plough - SGP, Astellas) and Flutiform (SkyePharma - SKYE), both in Phase II, are likely to lag, since they do not have major advantages and will come on the market later.

While all of these combination drugs require dosing twice a day, two LABA drugs with once-a-day dosing are in Phase II, and testing of once-a-day ICS combinations of these drugs are likely to begin this year or next. While data are preliminary, the QAB149-ICS (Novartis - NVS and partner) combination will likely have a faster onset of action, giving it an edge over GSK15979 (GlaxoSmithKline - GSK and Theravance - THRX). We doubt these drugs will be as suitable for MRT as the twice-a-day drugs, but if MRT is approved, we think they will still have a role and take over the fixed dose segment. If MRT is not approved, we project they will take market leadership.

In the short-acting, quick relief segment of the asthma market, the current market leader is generic albuterol, but these formulations will have to be replaced at the end of 2008 with non-CFC aerosol propellants. We think it is likely that non-CFC generics will also enter the market. Xopenex HFA (Sepracor - SEPR), a non-CFC formulation containing only the active form of the albuterol molecule, recently entered the market. The nebulized (a mist from liquid) formulation of the drug has shown some, albeit mixed, evidence of a benefit over regular albuterol, and has done well, though it has a minority market share. Without a more significant demonstration of benefit, we project Xopenex HFA will also gain only a minority share, though significantly higher than other branded non-CFC albuterol inhalers.

Among non-ICS anti-inflammatory drugs, Singulair (Merck - MRK), a leukotriene receptor inhibitor, is the market leader, and there is little on the horizon to challenge it. As an oral agent, it is particularly suited for children. For patients with more severe disease who fail corticosteroids, Xolair (Genentech – DNA, Novartis – NVS, Tanox – TNOX, and PDL BioPharma – PDLI), an anti-IgE antibody (IgE is the type of antibody involved in allergic asthma), has seen strong sales, and we expect it to continue to grow. Enbrel (Amgen – AMGN and Wyeth – WYE), an anti-TNF drug in Phase II development for asthma (and already approved for rheumatic diseases), has shown intriguing preliminary evidence. Though it can have significant side effects, we think it will be used in patients with severe disease. However, a potential competitor, Golimumab (Johnson & Johnson – JNJ, Schering Plough – SGP, and Medarex – MEDX), is being tested with less frequent dosing, which we believe will make it more popular. There are several other drugs in early development that suppress the immune response and have interesting mechanisms of action, though clinical detail is minimal.

BACKGROUND

Asthma affects 20 million people in the US, including 14 million adults and 6 million minors. There are about 1.9 million emergency room visits and 575,000 hospitalizations for asthma each year.

Asthma is typified by reversible constriction of the airways (bronchoconstriction), which causes wheezing, difficulty breathing, and in some cases, death. The wheezing is often caused by an allergic response to a trigger (such as dust-mites, dander, and fungi), but can also be set off by non-allergic triggers (such as exercise, cold air, viruses, and chemicals – chemicals from occupational exposure cause 5% to 25% of adult-onset cases).

Asthma is fundamentally an inflammatory disease. In the allergic form of asthma, the immune system makes certain antibodies (the IgE class) in response to inhaled allergens (substances that cause an allergic reaction) and bind to cells called mast cells and basophils. These cells release chemical mediators, such as histamine and leukotrienes, which cause airway smooth muscle to contract. This is called the early asthmatic response (EAR) and typically occurs within an hour of antigen exposure. Other chemicals (e.g. interleukins, tumor necrosis factor-alpha, and interferon-gamma) are also released which in addition to causing some changes associated with inflammation, recruit inflammatory cells into the area. One such cell thought to play a central role is the eosinophil, which carries proteins that can injure the airway lining and increase bronchial responsiveness to stimulants. The inflammatory response from the recruitment of other cells into the airways is called the late asthmatic response (LAR) and typically occurs 4 to 8 hours after exposure. Chronic inflammation from repeat exposure may lead to permanent changes in the airway called airway remodeling. Of note, a recent study found that a large proportion of certain T cells (these are cells involved in the immune response) found in the lungs of patients with moderate to severe asthma are a type called natural killer T cells. These cells are involved in the production of interleukin-4 and 13. The pathway represents a novel target for asthma therapy.

Nonallergic, or intrinsic asthma, refers to the fact that up to a third of patients have no evidence of an allergic reaction to common allergens. This form of asthma can be associated with nasal polyps and the triggering of asthma by aspirin. The mechanism for intrinsic asthma is not clear, though it shares many of the same pathological features of the allergic form. However, one unanswered question is whether there are groups of patients with different patterns of inflammation requiring distinct treatment.

Asthma that begins in adulthood can either be allergic or intrinsic. About 60% of adults with asthma developed it during childhood. Childhood-onset asthma is often related to atopy, an inherited tendency to produce an allergic response to common environmental allergens. In addition to asthma, atopy manifests with hay fever and skin problems. About a quarter of children with asthma do outgrow it.

One important test for the diagnosis and monitoring of asthma is called spirometry. During the test, a patient exhales rapidly into a tube. If the amount of air exhaled in 1 second (called the FEV1, or Forced Expiratory Volume in 1 second) is low relative to the total amount exhaled (called the FVC, or forced vital capacity), it indicates an obstruction of flow. To see if the obstruction is reversible (reversibility is an indicator of asthma) the test is performed before and after the patient inhales a bronchodilator (a medication that relaxes the muscles around the airways to dilate them). An increase of at least 12 percent or 200 ml in FEV1 due to the bronchodilator indicates significant reversibility (for some patients other methods need to be used).

CURRENT TREATMENT

In addition to minimizing allergic or irritant triggers of asthma, treatment involves drugs to reduce the inflammation and dilate the airways. Many of the drugs used in asthma are given as inhaled formulations, which have the advantage of delivering a higher dose while minimizing systemic side effects. Inhaled drugs also have a faster onset of action.

Inhaler Devices

The most commonly used inhalers are small, portable devices called metered dose inhalers (MDIs) or dry powder inhalers (DPI). MDIs, the traditional pressurized canister and mouthpiece, require coordinating a breath with spraying the aerosol, though breath-activated devices are also available now. In DPIs, the drug is only inhaled when a breath is taken so coordination is less of a problem. However, patients must take more forceful breaths than with an MDI, since there is no aerosol propellant.

If used properly, MDIs are as effective as DPIs. Using a spacer, essentially a chamber into which the aerosol is sprayed, makes use of MDIs easier and also reduces deposition of the drug in the mouth and upper airways. This is particularly important when using inhaled steroids to reduce oral side effects. On the downside, spacers may lead to a reduced dose if not used properly. Another issue with MDIs is that they have traditionally used chlorofluorocarbon (CFC) propellants. The FDA has ruled that single agent albuterol inhalers (see below), the most commonly used inhalers with CFCs, must be phased out by December 31, 2008 due to environmental concerns. There are already alternatives on the market designated HFA for their use of a hydrofluoroalkane propellant (patents on these products extend through 2010-2017). The HFA formulations are more effective than MDIs or DPIs in reaching the lung.

Nebulizers, which create a mist from liquid formulations, are generally more costly and less convenient (delivery of a dose can run from a few minutes to a half hour), though devices are becoming more portable. They are particularly useful for younger children, the elderly, and in treating moderate to severe exacerbations in the emergency room and hospital.

General Approach to Treatment

Treatment revolves around anti-inflammatory drugs and bronchodilators (which dilate the smooth muscle around the airways). While anti-inflammatory medications are very effective in treating asthma, they have not been shown to alter the long-term course of the underlying disease (on the other hand, many children who develop asthma after age 3 do not have worsening of underlying lung function anyway unless they continue with asthma as adults). Except for the non-steroid anti-inflammatory drugs, asthma drugs are also used to treat chronic obstructive pulmonary disease (patients with aspects of both are recommended to be treated as if they have asthma; please see our COPD Indication Report).

Bronchodilators include:

- Beta-agonists: drugs that dilate the airways by stimulating beta 2 adrenergic receptors on airway smooth muscle. They may be short (< 6 hours) or long (>12 hours) acting. Inhaled formulations are much more commonly used than oral ones. These are the primary bronchodilators recommended.

- Anti-cholinergics: inhaled drugs that dilate airways by blocking muscarinic acetylcholine receptors on airway smooth muscle (acetylcholine normally increases muscle contraction). They also diminish mucous secretion. Only short-acting agents are used in asthma, unlike the case with COPD, and they are recommended only if beta-agonists are insufficient or not tolerated.

- Methylxanthines: oral agents used on a daily basis to dilate airway smooth muscle, probably by blocking an enzyme called phosphodiesterase. They may also suppress the response of airways to stimuli. Though among the most inexpensive asthma medications, the levels at which toxicity can occur are close to therapeutic levels, and this class of drugs is not commonly used in developed countries.

Anti-inflammatory agents include:

- Corticosteroids: inhaled corticosteroids (ICSs) are the preferred anti-inflammatory agents for asthma and act at a number of levels to reduce inflammation, such as reducing the number of inflammatory cells and inhibiting the release of chemicals that mediate the inflammatory response. Corticosteroids are also available in oral, injectable, and intravenous formulations.

- Leuokotriene Modifiers: leukotrienes are chemicals released from inflammatory cells (mast cells, eosinophils, and basophils). They contract airway smooth muscle, increase leakage from blood vessels, increase mucus secretions in the airways, and attract and activate inflammatory cells in the airways of asthma patients. Leukotriene modifiers either block leukotrienes from being formed or block receptors for leukotrienes. They are given orally.

- Cromolyn and Nedocromil: thought weaker and not commonly, these agents block the release of mediators from mast cells after exposure to allergens.

Patients with intermittent symptoms are usually treated with the short-acting inhaled bronchodilators, referred to as relievers. Short-acting beta agonists are recommended first, with anticholinergics as an alternate for patients who cannot tolerate beta agonists, whose asthma is triggered by beta-blockers they are taking for heart disease, or who need additional relief in severe exacerbations.

Patients with persistent symptoms require daily maintenance treatment. Inhaled corticosteroids are the mainstay (unlike the situation with COPD, where bronchodilators are recommended initially). In general, if they are not sufficient, addition of a long-acting beta agonist is recommended. Other agents, such as leukotriene modifiers, are alternatives to steroids (in mild patients) or long-acting beta agonists (in moderate patients). However, leukotriene modifiers are used more frequently than would be expected from their designation as alternatives. They represent about a third of prescriptions for asthma maintenance medications, compared to a 40% share for preparations that include inhaled steroids. The following table shows more detail on recommendations from the National Heart, Lung, and Blood Institute (NHLBI) for daily maintenance treatment in patients older than 5 years of age. (Recommendations for patients 5 and younger are similar with only slight alterations.) Since the recommendations were made, an antibody against IgE, Xolair, has been approved for patients with moderate to severe persistent allergic asthma inadequately controlled with inhaled corticosteroids.

Asthma exacerbations, may be treated at home or at the hospital, depending on severity. At worst, exacerbations can be fatal. There are 574,000 hospitalizations and 1.9 million emergency room visits for asthma each year.

For milder cases, home treatment involves increasing the dose and frequency of short-acting inhaled bronchodilators, or using a short-acting nebulized formulation, if the patient has such a device. Dosing of inhaled corticosteroids may be increased, or oral ones added. Emergency room or hospital treatment of exacerbations may also initially involve increasing use of short-acting beta agonists via metered dose inhaler with a spacer or nebulizer. Nebulizers are more effective for patients having difficulty coordinating their inhalation from an MDI, and recommended above MDIs for more severe cases. Anticholinergics are added for more severe exacerbations. Systemic corticosteroids (oral or IV) are also generally used.

Currently Marketed Products

The following are the primary maintenance and quick relief asthma medications along with their formulations available in the US (excluding oral corticosteroids).

Short-acting Inhaled Bronchodilators

With their quick onset of action and short duration, inhaled short-acting beta agonists are the drug of choice for as needed use both in patients with intermittent asthma and those on maintenance medications. They are also important in treating exacerbations in the emergency room and hospital setting. Short-acting beta agonists are the most widely sold inhaled drug.

This segment has been dominated by generically available albuterol (the MDI formulation is the most common, with 4 times as many prescriptions as the nebulized version). Most of the albuterol MDIs contain CFC propellants, and will be banned after December 31, 2008. To date, there are two branded formulations of albuterol with the new HFA propellant (Proventil - Schering-Plough and Ventolin - GlaxoSmithKline) and one branded generic (from IVAX). These are generally priced $20 or more over the CFC generics, and so they currently have limited market share. While it is unclear whether 3M, which holds the HFA patents, will license the technology for more generic albuterol HFA competition, this is a likely possibility, which would limit the expansion of the branded HFA formulations after the CFC ban.

Xopenex (Sepracor) is a special albuterol formulation, available as a nebulizer solution since 1999 and an HFA MDI since December 2005 (though approved in March, manufacturing issues caused a delay in bringing it to market). Specifically, it is the R enantiomer of albuterol (albuterol – referred to as racemic albuterol – is a mix of two enantiomers, molecules that have the same chemical makeup but are mirror images of each other). The rationale for Xopenex is that the R enantiomer is the more active form, and there are some concerns that the S enantiomer may increase airway reactivity, though that has not been convincingly documented.

There has been controversy whether nebulized Xopenex provides significant clinical benefits over nebulized racemic albuterol. For example, though some studies showed low doses of Xopenex had comparable efficacy with much higher doses of racemic albuterol, they have been criticized for the possibility that patients were already on the upper, flat portion of the dose response curve, where additional benefits would not be expected with the higher doses. One major study found that Xopenex used in the emergency room for asthma exacerbations led to a lower rate of hospitalizations than racemic albuterol (36% vs 45%, p=0.02), but another study found no difference. While Xopenex appears to be at least as good as racemic albuterol, due to the conflicting evidence one major question has been whether the results are definite enough to justify the greater cost. As a result, while Xopenex has seen solid market uptake in the nebulized medication segment, achieving 27% of prescriptions and 33% of hospital units sold, it has not supplanted generic nebulized albuterol in the majority of patients.

For the recently introduced HFA formulation, pulmonary function tests have been similar to equivalent doses of albuterol HFA. Without more data showing an outcomes benefit, we do not think there will be a compelling clinical argument in favor of Xopenex HFA, though its theoretical advantage and the possible benefit seen with the nebulized form are likely to help it do better than branded albuterol.

Anticholinergics are less effective in asthma than they are in COPD. According to the NHLBI guidelines, short-acting anticholinergics have not been shown to add benefit to beta agonist treatment for long-term maintenance therapy in asthma. As noted above, short-acting anticholinergics are recommended as add-ons to beta agonists for more severe exacerbations, for reliever use in patients who cannot tolerate beta agonists, and for patients whose asthma is triggered by beta-blockers. Atrovent (Boehringer Ingelheim) is the sole single-agent short-acting anticholinergic and Combivent (Boehringer Ingelheim) the sole combination one for use in an MDI inhaler in the US (Combivent combines the active ingredient of Atrovent plus albuterol). Both are available generically as nebulizer solutions. In the overall respiratory market, prescriptions for anticholinergics are about 30% of those for beta-agonists.

Long-acting Inhaled Bronchodilators and Combinations with ICSs

There are two long-acting inhaled bronchodilators approved for asthma; both are twice-a-day beta agonists. Due to their greater effectiveness than short-acting bronchodilators, long-acting beta agonists (LABAs) have been recommended as the preferred add-on treatment to inhaled corticosteroids in patients with moderate asthma (though there has been only limited study of their benefit in pediatrics). However, sales of individual LABAs are a much smaller part of the market (together less than 5% of asthma maintenance prescriptions) because a combination product of a LABA and ICS, Advair (see below), has generally been preferred.

Serevent (GlaxoSmithKline) is the oldest single agent LABA, approved in 1994. Its chemical name is salmeterol, and the active part of the molecule is the same as for albuterol. The patent on the active ingredient of Serevent is due to expire in 2008 in the USA and expired in 2005 in most of Europe. Serevent is one of the components of Advair.

Foradil (Schering-Plough, Novartis, Astellas), or formoterol, was approved in 2001 in the US, and has a faster onset of action than Serevent. However, Foradil sales have been hampered because the device is less convenient to use: unlike Serevent, blisters containing the medication must be put into the device one at a time. The patents and exclusivity for Foradil’s active ingredient have expired in the US and other major countries. Another competitor with the same active ingredient, Oxis (AstraZeneca), is available in Europe, where a generic version has also appeared.

Advair (GlaxoSmithKline) is a combination of the LABA Serevent and the ICS Flovent (see below). A DPI formulation was first approved in the US in 2000 for patients 12 years and older, with expanded use to children age 4 to 11 approved in May 2004. (An Advair non-CFC MDI, already approved in Europe, received approvable letters from the FDA in 2001 and 2002. GSK expects it to be approved and introduced in the US in the third quarter of 2006.)

Advair has shown significant effects on pulmonary function and symptoms beyond those compared to either component alone, and at least as good as that of its individual components used together. In addition, the symptomatic relief patients feel with the LABA may increase compliance above an ICS alone. The clinical benefits and convenience of Advair has propelled it to the best selling (in terms of revenue) pulmonary drug, with US and worldwide sales of $2.4 billion and $4.5 billion respectively in 2004. The majority of the sales are due to asthma. GSK’s patent on the specific combination of active ingredients in Advair expires in 2010 in the US and 2013 in Europe. While there may be more difficulties in developing generic inhaled combination products compared to oral ones, we believe the market is so large that a generic substitute for Advair will arise.

The SMART Trial
LABAs have come under increased scrutiny since Serevent was associated with an increase in mortality in the SMART trial (Salmeterol [Serevent] Multi-center Asthma Research Trial). The trial compared Serevent to placebo when added to usual asthma treatment and was stopped early after an interim analysis. The study found a small but significant increase in both asthma- and respiratory-related mortality (13 versus 3 deaths and 24 versus 11 deaths, respectively, out of more than 13,000 patients in each group). The negative results were seen primarily in the African-American subgroup and those not using ICSs. It is still unclear whether the excess mortality seen in the SMART trial was due to Serevent itself, a class toxicity of the LABAs, a genetic predisposition to react to these drugs in a certain way, or Serevent masking worsening inflammation such that patients delayed seeking necessary additional treatment. Studies on the genetic issue and in African-Americans are ongoing.

Interim data from the trial led to an initial black-box warning for Serevent and Advair in 2003. After the FDA re-analyzed the data, an FDA Advisory Committee Meeting was convened in 2005, which recommended that the long-acting beta agonists remain on the market, but with increased warnings on their labels. On November 18, 2005, the FDA requested changes in the label for Serevent, as well as Advair (which contains the active ingredient of Serevent) and Foradil (since it could well be a class effect), recommending that these agents only be used for asthma patients failing corticosteroids. Though initially criticizing them, GSK agreed to the new labels for Serevent and Advair.

The manufacturers of Foradil did not agree to use the label, and the FDA is apparently not pursuing the issue further (they had sent out a warning to physicians on Foradil along with the one they sent for the other two). While Foradil itself has not been associated with an increase in mortality, several studies have raised the concern of serious asthma exacerbations when a high dose of the agent is administered. A 2,000 patient study was designed to answer this and did not find a significant deleterious effect for Foradil. However, the patients had less severe asthma than in the SMART study and a higher proportion were taking ICSs.

Despite the concerns raised by the SMART trial, we feel the medical community views combination treatment with ICSs and LABAs (including Advair) as safe for non-African-Americans. Many patients need the extra treatment with LABAs to control their symptoms. Hence while we think that growth in Advair sales will slow as doctors are more cautious in some patients (particularly African-Americans); without further negative data, we do not think there will be a dramatic reduction in Advair sales.

On the other hand, without further evidence, it is unlikely that the FDA will accept the view that combined use of LABAs with ICSs is safe, particularly since patients in the ICS subgroup of the SMART trial were not separately randomized. Hence the issue could crop up in future approvals.

Inhaled Corticosteroids (ICS)

Though glucocorticoids are effective at reducing inflammation (such as by decreasing the number and availability of inflammatory cells and inhibiting the production and release of chemicals that mediate the inflammatory response), they also have numerous side effects in the body. As a result, oral corticosteroids are recommended only for short periods to treat exacerbations or when inhaled corticosteroids are not adequate. ICSs, on the other hand, are generally thought to be safe for long-term use, since they are inhaled and less drug reaches the circulation to cause side effects. While some systemic effects of ICSs have been seen, the clinical implications are unclear. ICSs used in asthma have been associated with reduced bone mineral density in adults, but it is not known whether this increases the risks for fractures. High cumulative lifetime doses may slightly increase the risk for cataracts. Low to medium doses may have the potential to decrease growth in children, but the effects are small, do not appear to progress with further treatment, and may be reversible. Finally, there may be slight suppression of the body’s normal mechanism for producing natural hormones similar to corticosteroids (hypothalamic-pituitary-adrenal axis function). This is not thought to be clinically significant.

Pharmacological properties of different ICSs influence how effective they are in the respiratory system and the degree of systemic side effects. For example, particle size and formulation influences how much gets to the lung after an inhaled dose and how much is deposited in the mouth, where it can be swallowed and potentially have a systemic effect. Even if a drug is swallowed, if it is metabolized to a high degree in the liver (resulting in what is called low bioavailability), it will also have less systemic absorption. Products that dissolve better in lipid solutions (lipophilicity) than in water are thought to have a more prolonged action in the lung.

Due to their advantage in factors such as these, Flovent (GlaxoSmithKline) and Pulmicort (AstraZeneca) are the most commonly used single agents, with 45% and 30% of single-agent ICS prescriptions respectively. Most of Pulmicort US sales are for the nebulized formulation, which is particularly useful for smaller children. Since these two are the most potent ICSs, fewer doses per day can be used for the same clinical benefit as less potent ICSs. In addition, the active ingredient of Flovent is used in Advair, and the active ingredient of Pulmicort is being developed as a combination product with the long-acting beta-agonist formoterol (see Symbicort below). To put single agent ICSs use in context, the combined use of all single agents is half that of Advair.

Asmanex (Schering-Plough) is a long-acting corticosteroid delivered by DPI, which unlike current formulations is usually dosed once-a-day (twice-a-day for the highest dose). It was approved for asthma in March 2005, and was launched late in the third quarter. Schering-Plough started with a limited launch targeting specialists and only recently began to market to primary care physicians. Of approved ICSs, Asmanex has the highest affinity for the corticosteroid receptor. It has a number of pharmacologic properties thought to reduce systemic effects, though it shares these with other competitors. Its major point of differentiation is convenience. There have not been a great deal of comparative trials, and the high once-daily dose of Asmanex is listed in international guidelines as a low medium dose. In addition, that dose (400 micrograms/day) led to numerically worse (but not statistically significant) FEV1 than a moderate dose of Flovent (250 micrograms twice a day). Hence without further information, we think it is unclear whether patients who need moderate doses at the higher end can be effectively treated with the once-a-day dosing.

Nevertheless, the convenience of the once-a-day dosing will be a plus for Asmanex in patients with milder disease who only need an ICS (without a LABA). We think it will steadily gain popularity as doctors try the product and become comfortable with it, though it will fact competition from Alvesco (see below). Asmanex may benefit slightly from the fallout from the Serevent SMART trial (discussed in the section on LABAs) in these patients, though we think there will only be a modest effects on Advair sales, as many patients benefit from the addition of the LABA to the ICS. A combination of the active ingredients of Asmanex and the LABA Foradil, Formoterol-Mometasone, (Novartis, Schering-Plough, Astellas), is in Phase I development for COPD and Phase II for asthma. However, formoterol requires twice-a-day dosing, so this product would not have added convenience. The onset of action of formoterol is quicker than that of the LABA in Advair, but the same as that in the combination product Symbicort, which should have the advantage of reaching the market sooner.

Leukotriene Modifiers (LTRAs)

These agents affect leukotrienes, which are released from inflammatory cells and have a number of actions, including airway smooth muscle contraction and the attraction of more inflammatory cells. Singulair (Merck) and Accolate (AstraZeneca) block leukotriene receptors, and Zyflo (Critical Therapeutics and Abbott) inhibits the formation of leukotrienes. They are all given orally.

While no definitive difference in efficacy has been demonstrated for the three, Singulair only needs to be given once a day, has a better side effect/drug interaction profile, and is available as a chewable tablet for children. As a result, it dominates the sector, with 97% of prescriptions.

While ICSs generally outperform LTRAs in clinical trials, there may be more compliance with LTRAs in children, since as oral drugs they are easier to take. This has helped make Singulair quite popular: it is the most widely prescribed asthma maintenance medication in the US (it was also approved for seasonal allergic rhinitis in 2002 and perennial allergic rhinitis in July 2005). We do not see a near term challenger.

The US patent protection on Singulair ends in February 2012, with pediatric exclusivity until August of the same year.

Anti-Immunoglobulin E (Anti-IgE)

Xolair (Genentech, Novartis, Tanox, PDL) is a monoclonal antibody that binds IgE, and is the sole such agent approved for asthma. By binding IgE, Xolair prevents it from interacting with IgE receptors on mast cells and basophils, thus reducing the release of allergy mediators. Xolair was approved in the US in 2003 and in Europe in October 2005. It is given as a subcutaneous injection every 2 or 4 weeks, and due to a small risk of a severe allergic reaction, it is recommended that doses be given in a physician’s office. Dosing is based on a patient’s baseline IgE levels. It is approved for patients 12 and above, who have moderate to severe persistent asthma inadequately controlled with ICSs (only approved for severe in Europe), as well as a positive test for sensitivity to a year-round airborne allergen.

Xolair decreases the incidence of asthma exacerbations and emergency room visits in patients who are not controlled by ICSs. (Exacerbations were not reduced in a study of patients on oral corticosteroids.) There is question about whether it increases the risk of malignancies: though showing numerically higher rates, the difference from placebo has been small, and the significance is unclear.

Due to such considerations, as well as high cost, Xolair is primarily used in more severe cases of asthma, which represents less than 5% of the asthma population. Nevertheless, it has shown strong revenue growth, and we expect this to continue, though it will only be used in a small percentage of patients.

DRUGS IN DEVELOPMENT

DRUGS IN NDA

Symbicort - AstraZeneca (AZN)

Symbicort, a competitor for Advair, is a combination of the long-acting beta agonist formoterol (the active ingredient in Foradil and Oxis - the latter is AZN’s drug) and the ICS budesonide (the active ingredient in Pulmicort). It is a twice daily formulation, being developed for an MDI in the US (according to AstraZeneca, the Turbohaler DPI formulation is not being developed in the US due to the regulatory burden regarding manufacturing issues). Currently, an NDA was submitted in September 2005 for asthma (it is in Phase III development for COPD). Patents are due to expire 2012 in most major European countries and in approximately 2014 in the US.

The beta-agonist component of Symbicort has a more rapid onset of action than that in Advair. In Europe, where Symbicort is already approved, this has not given it enough of an advantage to supplant Advair. However, AstraZeneca has used this difference -- along with the fact that unlike the case with the LABA in Advair, increasing the dose of formoterol beyond standard doses provides an additional benefit -- to develop what it calls maintenance and reliever therapy (MRT). Instead of a fixed dosage, as is currently the standard with maintenance medications, maintenance and reliever therapy involves a lower fixed maintenance dose, with patients increasing the dosage of the same medication as needed for relief of symptoms (according to a prespecified plan).

In Europe, Symbicort has submitted maintenance and reliever dosing for approval in September 2005 (a prior submission in 2003 where this was called Single Inhaler Therapy was not successful, and more studies were required by the regulatory authorities). Although AstraZeneca IR representative has stated that they are not revealing the type of dosing they submitted in the US NDA (or details on the trials they are using to support the submission), at their 2005 fourth quarter earnings call, the director for development mentioned they had submitted an NDA for fixed dosing. Still, it is unclear if this means that a flexible dosing regimen was not submitted as well.

There have been a number of studies showing MRT resulting in fewer exacerbations than standard fixed dosing. A couple of trials have compared Symbicort MRT to fixed dose Advair. These studies have showed a large benefit in endpoints such as reducing exacerbations. However, the direct comparison trials for which data are available have also been open-label. (A blinded comparison was done for the EMEA resubmission, but data have not been released: we assume it was positive.) On the other side, a blinded study comparing Symbicort MRT and Advair, sponsored by GSK, showed the opposite, with Advair coming out ahead with a large benefit. However, in this study, patients could reduce their Symbicort dosage to only 1 puff per day (rather than 2 as in other studies). This most likely put them at risk for exacerbations.

The increased use of LABAs with worsening symptoms in Symbicort MRT goes counter to the concerns raised by the SMART trial over the potential deleterious effects of LABAs, and raises several questions. (Ironically, the full acronym for Symbicort Maintenance and Reliever Therapy is SMART, the same as that for the fateful Serevent trial.) It is unclear which component of Symbicort accounts for the purported benefit in MRT. One trial done for the European resubmission looked at whether it was the LABA component, but data have not been released. If that trial showed that a LABA as a reliever on top of an ICS reduced exacerbations to the same extent as Symbicort MRT, it would conflict with concerns raised by the SMART trial. (It would imply that it was the LABA component that provided the benefit of MRT. One hypothesis regarding the SMART trial is that though LABAs added to ICSs reduce exacerbations, in other situations where asthma is poorly controlled, LABAs might give temporary symptomatic benefit but also delay needed treatment, resulting in increased mortality. The Symbicort trial may not have been large enough to detect this.) As a result, in this situation, we think regulators would be hard pressed to approve the drug without further study. If the resubmission study did not find that, it is still unclear whether adding an ICS alone on top of fixed combination treatment is safer than using the combination treatment as a reliever. (Control groups for the MRT studies generally did not involve increasing ICS with worsening symptoms. Also, while other trials where the single-agent ICS has been doubled in exacerbations -- which is commonly recommended -- have found no benefit, most patients in those trials did not receive any LABA treatment, which conceivably could affect the results). The GSK study, where lower mean doses of Symbicort were ultimately used and where Symbicort MRT did worse than fixed dose treatment, raises the question of whether it is too risky to leave dose adjustments to patients. Finally, there is a general question of whether using the lower doses of ICSs in MRT could lead to more problems down the road due to less suppression of inflammation.

Despite these concerns, we feel overall that Symbicort MRT is an attractive therapy and the weight of evidence is in its favor. However, in contrast to a number of opinions expressed in the medical literature, the FDA has not appeared to accept the notion that LABAs in conjunction with ICSs are safe without further evidence, so even if MRT has been submitted as part of Symbicort’s NDA, we think it is unlikely it will be approved without further study of this issue. In Europe, the additional trials required after the first submission were most likely already ongoing when more information regarding the SMART trial came out due to the FDA reanalysis of data. Hence the additional studies done for resubmission will not necessarily resolve the issue for the EMEA, either, though we think they are more likely to approve MRT than the FDA. If Symbicort MRT were to be approved, we feel it is likely to become the dominant therapy in patients with less severe disease, and would likely propel Symbicort into a leadership position. As a result, we think AstraZeneca will continue to pursue approval of MRT, though it is possible that they will opt to try and gain a marketing advantage over Advair with their fixed dose, hoping physicians will prefer a different option due to concerns raised by the SMART trial. Also, if the European authorities approve the drug first and the FDA requires major new studies, AstraZeneca may delay pursuing the US studies because a negative result could jeopardize the marketing of MRT in Europe.

The PDUFA for Symbicort is July 23, 2006. Even if AstraZeneca has not applied for Maintenance and Reliever Therapy, it is very likely approval will be delayed. When asked about approval at the recent earnings call, AstraZeneca’s CFO noted that the rate of approval for respiratory devices is only 10% the first time through, and most drugs can expect at least 2 cycles of regulatory submissions.

Foradil Certihaler - Schering-Plough (SGP), Novartis (NVS), SkyePharma (SKYE)

Foradil Certihaler is a multidose dry powder formulation of Foradil Aerolizer, which is a single dose DPI. Hence the advantage of the new formulation is convenience of the multi-dose feature. Though approved in a number of countries in Europe, it has received two approvable letters in the US and is currently in NDA status for both asthma and COPD. Details on what is holding up approval have not been released. In any case, if approved, while it is likely to do better than and take share from Foradil (and Serevent), single agent LABAs have a relatively limited market potential in asthma compared to combination medications.

DRUGS IN PHASE III DEVELOPMENT

Alvesco – Altana (AAA) and Sanofi-Aventis (SNY)

Alvesco is a once-a-day corticosteroid (twice-a-day for high doses) in Phase III development. It is delivered by an HFA MDI. Alvesco is a prodrug that is activated in the lungs. It scores high on the list of pharmacologic properties of a desired ICS. In addition to convenience, a significant advantage is a lower incidence of oral fungal infections, a side effect of corticosteroid deposition in the mouth. However, of concern was a study of high doses showing an increased rate of nuclear cataracts (cataracts in the center of the lens) with high dose Alvesco (2.4% for Alvesco 160 microgram twice-a-day; 6.2% for Alvesco 320 micrograms twice-a-day, 0.7% for Flovent 440 micrograms twice-a-day, and 0.7% for placebo). While nuclear cataracts are not the type most typically associated with steroid treatment (ie, subcapsular), they too have been implicated. On the other hand, it is puzzling why Alvesco should cause cataracts, since on other measures it appear to have lower systemic effects.

Alvesco received an FDA approvable letter in 2004 as further clinical information was needed. One trial evaluating the eye effects of higher dose Alvesco has already been completed, but the data have not yet been released. However, representatives from both Altana and Sanofi have stated that results were positive for safety of the drug. Two additional trials, 12 week and 16 week studies looking at once-a-day versus twice-a-day dosing, were instituted this year and are slated to be finished by August 2006. Alvesco was approved in its first European country in April 2004, but only for the lower dosages of the drug (160 micrograms once a day was the top dose approved).

The main negative issue for Alvesco has been the higher rates of cataracts seen in the one trial. We are encouraged by the companies’ statements, but remain somewhat cautious prior to seeing the actual results. If the recently completed trial indeed confirms that patients are not at risk, we feel Alvesco would do well against other ICSs, due to its lower systemic effects and once-a-day dosing. For it to be successful, the higher doses need to be approved as well, which we feel will happen if safety is demonstrated.

Altana and Sanofi are also developing a combination DPI product using Alvesco and a LABA, Formoterol-Ciclesonide, currently in Phase IIb for asthma (results for the IIb trial are expected before the end of the year, and regulatory submission is planned for 2007/2008). Again, we feel approval of the higher doses hinges on the safety data. The use of formoterol will require twice-a-day dosing, negating a convenience advantage. However, with the rapid onset of action of formoterol and the low systemic effects of ciclesonide, we feel this will still be a strong competitor, which will also be suitable for Maintenance and Reliever Therapy. On the other hand, if it turns out there are elevated risks for cataracts at higher doses of Alvesco, we do not think those doses will be approved, and so Alvesco's uptake will be limited.

Daxas – Altana (AAA)

Daxas is a once-a-day oral phosphodiesterase-4 (PDE4) inhibitor in Phase III development for asthma and COPD. It is thought to work by reducing the activity of a variety of inflammatory cells. Of note, the methylxanthine Theophylline is also a phosphodiesterase inhibitor, though it is not specific to PDE4.

Daxas has shown similar effects on FEV1 compared to a low dose of an inhaled steroid (over 12 weeks) and compared to Singulair (over 6 months). However, it only showed a trend towards improvement when added on to low-dose Flovent.

In July 2005, Pfizer terminated its agreement with Altana after a COPD trial failed to show a benefit in overall exacerbations. And in November 2005, Altana withdrew its application for Daxas with the EMEA, pending further clinical data, despite the fact that it apparently had thousands of patients in its Phase II and III studies. These are obviously perturbing developments. In addition, though Daxas has shown some benefit in some measures, we are concerned that we have not seen mention of its ability to reduce exacerbations in comparison with agents such as inhaled corticosteroids or Singulair, or its ability to reduce ICS use. Hence we are cautious in our outlook on Daxas for asthma.

DRUGS IN EARLY DEVELOPMENT

There are a variety of drugs in early development. In addition to improved LABAs and ICSs, there are a number of immunosuppressant drugs. While some of these have interesting clinical information, we have also included others based solely on the novelty of their mechanism of action to give a better idea of what is in the pipeline.

Beyond Advair Collaboration – GlaxoSmithKline (GSK) and Theravance (THRX)

The Beyond Advair collaboration between Theravance and GlaxoSmithKline began in 2002 with the ultimate purpose of developing a once-daily LABA to use in a replacement product for GSK's Advair, which is given twice-a-day. Each company contributed four LABA product candidates.

GSK will provide the ICS for the combination product, most likely 685698, which is in Phase II development for asthma and also Phase III for allergic rhinitis under the name Allermist. GSK also has another ICS in Phase II development, 799943. Little information about these compounds has been released. As single agents, these ICSs will likely compete against Asmanex and Alvesco.

The lead once-daily LABA of the Beyond Advair Collaboration is GSK159797, in Phase II development. It is to be delivered via DPI. Though there were some formulation issues causing a delay in 2005, these were apparently resolved, and the company announced plans to initiate a Phase IIb trial in the first half of 2006.

Some clinical information has been released: in a trial of 20 patients, GSK15979 showed increases in FEV1 at 24 hours of 460 ml to 540 ml, compared to an increase of 130 ml for placebo. These are in the range of what has been seen at 12 hours with existing LABA treatment in asthma, though on the low end. Importantly, the heart rate was only minimally increased by the drug (a potential side effect of beta-agonists).

It is too soon to tell whether the drug will be successful in achieving 24-hour control for asthma. Given the concerns from the SMART trial, the product will come under increased scrutiny, though we think the main factor will be whether the product can provide 24-hour coverage comparable to what the twice-a-day products now give. The largest market opportunity is in combination with a once-a-day ICS. The convenient dosing will, of course, provide a large advantage over twice-a-day products. Of concern, GSK159797’s onset of action is likely more similar to that of Serevent, and though little data has been released, we think likely slower than that of another competitor in development, QAB149 (see below). Furthermore, QAB149 is now ahead of GSK15979 in development (though Novartis is not specifying whether the same delivery device will be used for the combination QAB149 product). It is not yet clear how these two drugs stack up against each other on other measures, and also with which 24-hour ICS QAB149 will be combined. A likely partner candidate for QAB149 is Asmanex, though as previously noted, it is not clear how patients with more severe disease fare on Asmanex’s once-a-day dosing compared to twice-a-day drugs. There is little information on the GSK 24-hour ICSs. Though there are pros and cons for each and the competition could be close, based on QAB149’s likely faster onset of action (and potential earlier time to market), we currently think it will come out ahead.

As a 24-hour drug, we believe GSK15979 is likely to be less suited for maintenance and reliever therapy (if MRT is approved), and as a result, we believe it would be restricted to a smaller market share.

There are two other LABAs from the collaboration now in Phase IIa development, as shown in the table below. The table also shows which company contributed the compound (though Theravance receives the same royalties on all the drugs, the milestone payments vary).

QAB149 – Novartis (NVS)

Like GSK159797, QAB149 is a once-daily inhaled beta-agonist. It has completed several Phase II trials in asthma and COPD, and Phase III trials are expected to begin in the coming months. The drug has been tested in the past using SkyePharma’s powder formulation, but Novartis is not specifying what type of delivery device will be used in phase III. QAB149 appears to have a relatively rapid onset of action. In 42 patients with intermittent or persistent asthma, the increase in FEV1 at 24 hours was 10.6% higher than for placebo. While on the order of some data on twice-a-day LABAs, the effect appears on the low end. Like GSK159797, there is not enough information to tell if the drug will be successful for 24 hour control of asthma. If it is, it would have to be combined with an ICS to gain a large market share, which Novartis has stated it is exploring. As noted above, Novartis has already partnered with Schering to develop a Formoterol (LABA)-Mometasone (ICS) combination. Since mometasone (unlike formoterol) can be given once-a-day, it would be a logical choice to combine with QAB149 for a once-a-day combination product. While it is unclear how well once-a-day mometasone will fare in patients with more severe disease, it is likely to be fine in patients with milder and more moderate disease. Another option is Alvesco. Given that QAB149 likely has a more rapid onset of action than GSK15979, we currently think the QAB149-ICS combo will come out ahead in the market. In addition, single-agent QAB149 is currently ahead in development, thought it is unclear whether the combination product will use the same delivery device. Like GSK15979, as a 24-hour drug, we believe the QAB149-ICS combo is not likely to be suited for maintenance and reliever therapy if that is approved, and hence, it would be restricted to a smaller segment of the market. If MRT is not approved, we believe the combo could become a market leader.

Flutiform – SkyePharma (SKYE)

Flutiform is a combination of the inhaled corticosteroid fluticasone with the long-acting bronchodilator formoterol, being developed as an HFA metered-dose inhaler for asthma using a fixed dosing schedule. It has finished its Phase II trial and a Phase III trial is to begin this year (originally planned for January). SkyePharma plans to submit their NDA by mid-2007, and is hoping for approval in the first half of 2009. Of note, the company has undergone some internal turmoil recently after receiving an unsolicited bid and demands from a group of institutional investors to install a director of their choosing. The company underwent a strategic review and replaced its top management. An agreement on directors was recently reached, but the company is selling off a division, apparently for needed cash to keep Flutiform their top focus in development.

Since the components of Flutiform are well-established drugs, it is likely the combination will work and be as effective as others on the market. However, while this is a strong combination of what are some of the best agents currently in their classes, it is unclear whether Flutiform offers a significant enough clinical benefit to supplant well-established competitors. Also, we think the Formoterol-Ciclesonide combination will be preferred. The company is not currently planning to develop Flutiform for maintenance and reliever therapy, though it is likely to be suited for this, and if that should that become popular, as we believe, additional trials would be needed. In the fixed-dose segment, Flutiform will likely need to compete after a few years with the more convenient once-a-day preparations now in Phase II development. Finally, the internal turmoil at the company could delay development. Thus overall, we are currently pessimistic about Flutiform’s prospects.

Formoterol-Mometasone – Novartis (NVS), Schering-Plough (SGP), and Astellas

This combination of the LABA formoterol and the active ingredient in the ICS Asmanex is in Phase II development for asthma and Phase I for COPD. Mometasone can be given once a day, but formoterol requires twice-a-day dosing. As a result this product would not have added convenience over the current leader Advair. The onset of action of formoterol is quicker than that of the LABA in Advair, but the same as that in the combination product Symbicort, which should reach the market sooner. Hence Formoterol-Mometasone will suffer the same latecomer competitive pressures as Flutiform. We also think the Formoterol-Ciclesonide combination will be preferred.

Enbrel – Amgen (AMGN) and Wyeth (WYE)

Enbrel is a recombinant fusion protein consisting of the constant region of a human antibody grafted to the binding portion of the human tumor necrosis factor (TNF) receptor. The TNF receptor part of Enbrel functionally acts like an anti-TNF antibody, binding to TNF (both alpha and beta forms). TNF is an important cytokine (an immune regulatory protein) that is involved in a number of the body’s inflammatory and immune responses. There are high levels of TNF-alpha in fluid from the respiratory tract of patients with severe asthma.

Enbrel is approved for rheumatoid arthritis and other rheumatic diseases. It is given as a subcutaneous injection once or twice a week. Though development in asthma had previously been suspended in 2003, it is now again in Phase II. Recently, an investigator-initiated study was published providing evidence of a role for TNF-alpha in patients with asthma refractory to other treatment (including corticosteroids). Enbrel also improved FEV1 by 320 ml compared to placebo over 10 weeks. Changes in markers of TNF-alpha were correlated with a measure of bronchial hyperreactivity. Though small, the study provided intriguing data on the potential benefit of Enbrel in refractory asthma patients. There was some preliminary evidence that markers of TNF-alpha were not elevated in patients with more moderate disease, so Enbrel’s effects may be limited to the subset of patients with severe asthma. Since Enbrel affects the body’s immune response, concerns over potential side-effects such as infection is likewise likely to limit its use to the more severe patients. Significant safety studies will need to be done, as these patients are also on corticosteroids, which likewise suppress the immune system.

We are encouraged by the preliminary data for Enbrel in asthma. A 12-week Phase II study began in May 2005, but results have not been released. One source of market risk for Enbrel is that there are already other anti-TNF antibodies on the market and in development for rheumatic conditions which could also be developed for asthma. For example, Golimumab (see below) has already seen clinical testing in asthma -- and has a more convenient once-a-month dosing. While we think that Golimumab will be favored, its relative efficacy remains to be seen. We also think there could be a number of patients who respond best to Enbrel or do better on its weekly dosing.

Golimumab – Johnson & Johnson (JNJ), Schering-Plough (SGP), and Medarex (MEDX)

Golimumab is the fully-human anti-TNF-alpha monoclonal antibody version of Remicade in Phase II for asthma (and Phase III for rheumatoid arthritis). Remicade has sequences of mouse protein in it, which can lead to the formation of antibodies that reduce Remicade’s effectiveness and cause a hypersensitivity reaction -- hence the desire to create a fully-human version. A one-year Phase II trial began in August of 2004, but we have not seen any clinical data yet on Golimumab for asthma. Since a drug with the same mechanism of action, Enbrel (see above), has shown some preliminary evidence of efficacy, we are encouraged about the prospects for Golimumab. One factor in its favor is dosing: it is being developed as a once-monthly subcutaneous injection versus Enbrel’s weekly regimen.

Mepolizumab – GlaxoSmithKline (GSK) and PDL BioPharma (PDLI)

In Phase II development, Mepolizumab is a monoclonal antibody targeting interleukin-5 (IL-5), which regulates the production of eosinophils and their release from the bone marrow. Eosinophils are one of the prime cells implicated in asthma. Mepolizumab has been shown to reduce eosinophils in the airway and also the deposition of some proteins that may lead to more permanent airway changes due to inflammation. However, it did not reduce a measure of an allergic response to an inhaled antigen (the late phase response which occurs 4 to 8 hour after exposure to an allergen due to the infiltration of inflammatory cells). This has raised questions about the precise role of IL-5 and eosinophils in asthma. Thus, though quite preliminary, the data thus far are mixed, and suggest the drug would only have an effect on part of the asthma process. Also, as a monoclonal antibody, Mepolizumab has the inconvenience of needing to be given by injection or infusion. We are currently doubtful about the prospects of this drug.

Inhaled Bimosiamose – Revotar and Encysive (ENCY)

Bimosiamose is a selectin inhibitor in Phase II development. During inflammation, selectins on cells lining blood vessels (E- and P-selectin) and inflammatory cells (L-selectin) help the inflammatory cells leave the circulation and enter the site of inflammation. Bimosiamose blocks all three of the known selectin receptors.

In a small study of patients with mild asthma, Inhaled Bimosiamose reduced the late asthmatic response (drop in FEV1 3-8 hours after exposure to an inhaled allergen) by 50%. There was no effect on the early asthmatic response or on airway hyperresponsiveness. These results are difficult to compare with others due to differences in the patient population. However, in a study of Singulair and Flovent with a similar placebo response to the current study, the magnitude of the effect with Singulair was smaller and that with Flovent larger than that seen here. On the other hand, the percent improvement seen here with Inhaled Bimosiamose was similar to that seen with ICSs in other studies of patients with more severe disease. There is little information on adverse events.

Overall, the results for Inhaled Bimosiamose are encouraging, albeit preliminary. We like the idea of an inhaled immunosuppressant, since side effects are likely to be lower than the injectable ones, and we think it will be able to be used in a broader population. Of some concern, though, is the similar mechanism of action as the integrin inhibitors, which have been put on hold due to the risks seen with Tysabri (see R411 below).

R411 – Roche (RHHBY)

R411 is an oral dual alpha4beta1-alpha4beta7 integrin antagonist in Phase II development. These integrins are receptors on inflammatory cells (particular Th2 lymphocytes, eosinophils and mast cells). Like the selectins, they are involved in migration of inflammatory cells from the circulation to the site of inflammation, and may also be involved in cell activation.

In a twelve-week study, R411 was first added on to ICSs and then the ICSs were withdrawn. While patients showed a better response than placebo, they still experienced a decline in FEV1 with the withdrawal of ICS, suggesting that at the doses tested, the drug is not as effective as ICSs.

One major issue facing the drug is that the FDA had placed this class of drugs on hold pending its evaluation of another integrin inhibitor, Tysabri for multiple sclerosis, which was associated with several cases of a fatal neurologic disease. Since an FDA Advisory Committee recently recommended Tysabri be reintroduced, we think the research hold on the class will be lifted. At the same time, further development of the integrin inhibitors will come under a great deal of scrutiny. Given the potential risks seen with Tysabri, we think R411 will have to demonstrate impressive efficacy. Overall, we do not hold high hopes for this drug.

Kinase gene inhibitor - deCODE genetics (DCGN)

This drug inhibits MAP3K9 kinase, the product of a gene with variants that deCODE has linked to the risk of asthma. The company states that the variants are present in 40% of patients in the UK and Iceland. Protein kinases are enzymes involved in transferring signals within cells to perform or inhibit a function. A number of kinases are thought to play a role in the activation of chemical mediators and immune cells involved in airway inflammation. deCode developed a computerized DNA databank of asthma patients to identify a kinase gene variant linked to asthma. The compound is currently undergoing an 8-week Phase II trial as add-on therapy, which is expected to conclude this quarter. This is a very novel approach, and it will be quite interesting to see if there is anything to it.

Zenapax – PDL BioPharma (PDLI) and Roche (RHHBY)

Zenapax is a monoclonal antibody that binds to the interleukin-2 (IL-2) receptor on immune system cells known as T cells. This keeps IL-2 from stimulating the T cells to divide and participate in an immune response. It is interesting that one type of cell that IL-2 stimulates is the natural killer cell, which has recently been implicated in asthma. PDL has partnered it with Roche for transplant rejection (for which it is already approved) and respiratory diseases and with Biogen IDEC for other indications. In Phase II clinical testing of patients poorly controlled on corticosteroids, intravenous Zenapax has shown a reduction in exacerbations and a modest improvement in FEV1 of 5.9% compared to placebo. Reductions in blood eosinophil levels were correlated with the absence of exacerbations.

Single and multiple dosing trials of a more convenient, subcutaneous formulation were recently initiated with healthy volunteers. Dosing is complete and PDL expects to start a Phase IIb dose-ranging study for chronic persistent asthma in the second half of 2006. One concern we have are the hypersensitivity reactions that can develop to the drug seen in its use in transplantation. Also as a general immunosuppressant, the drug may render patients more susceptible to infection with long-term use.

FK506 - Astellas

FK506 is in inhaled formulation of the approved immunosuppressant Tacrolimus, which inhibits the production of interleukin IL-2 by helper T-cells, blocking T cell activation and proliferation. FK506 may also inhibit the production of other cytokines, including IL-3, IL-4, IL-5, and TNF-a. Of interest, it has been shown in some situations to inhibit natural killer cell activity. Other formulations of the drug, which is an immunosuppressant, are approved for use in organ transplants and for atopic dermatitis, which is related to asthma. Of concern is the possible risk for kidney toxicity and the development of lymphoma with the agent. The company hopes that the inhaled formulation will lead to fewer adverse events. Detailed clinical information on the inhaled formulation is not yet available.

REVENUE MODELS

We have added eleven new revenue models for asthma and incorporated two others with revisions (Xolair and Inhaled Bimosiamose). In addition, we have added four new revenue models for COPD (Formoterol-Ciclesonide, Formoterol-Mometasone, QAB149-ICS, and Xopenex Inhaled Solution) and revised our COPD projections for the ICS-LABA combination drugs and Xopenex HFA based on an updated evaluation. Finally, while our base case projects that Maintenance and Reliever Therapy (MRT) will eventually be approved for asthma (it is not applicable in COPD), we have also added a fixed dose scenario in asthma, showing projections for the LABA-ICS combination products should MRT not be approved.

Advair - GlaxoSmithKline (GSK)

While Advair is currently the lead inhaled respiratory product, we expect sales growth to slow, particularly with the labeling changes due to concerns over the SMART trial. However, we expect it to maintain a leadership position for a few years until Symbicort MRT is approved. Also, once Advair’s US patent expires in 2010, we think generics are likely to enter and take away much of its share. Finally, the once-a-day combination products will take over leadership of the fixed dose market.

Our 5- and 10-year projected US revenues for asthma are $681.2 million (after the entrance of generics) and $56.2 million, with corresponding worldwide revenues of $2.2 billion (as its share declines in Europe but without generics) and $150.9 million. Please see our spreadsheet for revenues in the fixed dose scenario.

In COPD, our 5- and 10-year projected US revenues are $305.6 million and $86.2 million, with corresponding worldwide revenues of $1.4 billion and 272.1 million (10-year revenues are higher than in asthma due to Maintenance and Reliever Therapy in the latter).

Symbicort - AstraZeneca (AZN)

We expect approval to be delayed until the end of 2007 for the fixed dose, and are projecting approval for Maintenance and Reliever Therapy (MRT) a couple years later. Our Europe model presumes an earlier MRT approval in 2007. With MRT, we expect Symbicort to take a leadership position, until patent expirations in the US (2014) and Europe (2012). Since Advair cannot be used for MRT, its patent expirations will not affect Symbicort in this model. While we expect MRT to take a majority share of patients, we also believe a sizeable number will remain on fixed dose, especially those who have more severe asthma. In the fixed dose scenario, we project Symbicort to be behind Advair, with shares dropping once Advair goes off patent. Also, in that scenario, the 24-hour drugs take majority share.

Our 5- and 10-year projected US revenues for asthma in our base case (MRT) are $679.2 billion and $412.6 million, with corresponding worldwide revenues of $2.8 billion and $1.1 billion.

In COPD, our 5- and 10-year projected US revenues are $172.6 million and $43.4 million, with corresponding worldwide revenues of $644.6 million and $136.1 million.

Formoterol-Ciclesonide – Altana (AAA) and Sanofi-Aventis (SNY)

Given Alvesco’s benefits over other ICSs, we think this combination will be preferred over Formoterol-Mometasone and Flutiform. We also think it will take some share from Advair and Symbicort, though these will have the advantage of being well-established in the market. At the same time, we are tempering our share projections for Formoterol-Ciclesonide, pending actual release of data from the safety trial evaluating the eye effects of higher doses of the drug. If it appears likely MRT will be approved, we believe Formoterol-Ciclesonide will be developed for the flexible dosing regimen and do well.

Our 5- and 10-year projected US revenues for asthma are $214.1 million and $415.6 million, with corresponding worldwide revenues of $458.9 million and $668.4 million.

In COPD, our 5- and 10-year projected US revenues are $116.9 million and $87.1 million, with corresponding worldwide revenues of $255.9 million and $215 million.

Formoterol-Mometasone – Novartis (NVS) and Schering Plough (SGP)

As a late entrant without major advantages, we expect Formoterol-Mometasone to lag behind Symbicort and Advair, as well as Formoterol-Ciclesonide. If it looks likely that MRT will be approved, we believe Formoterol-Mometasone will also be developed for MRT.

Our 5- and 10-year projected US revenues for asthma are $71.7 million and $92 million, with corresponding worldwide revenues of $153.4 million and $225.3 million.

In COPD, our 5- and 10-year projected US revenues are $29 million and $14.3 million, with corresponding worldwide revenues of $63.8 million and $38.6 million.

Flutiform – SkyePharma (SKYE)

Like Formoterol-Mometasone, we think Flutiform will suffer from its late entrance and lack of significant advantages. We are somewhat concerned about its development path, given the recent turmoil at the company.

Our 5- and 10-year projected US revenues for asthma are $71.7 million and $92 million, with corresponding worldwide revenues of $153.4 million and $225.3 million.

In COPD, our 5- and 10-year projected US revenues are $29 million and $14.3 million, with corresponding worldwide revenues of $63.8 million and $38.6 million.

QAB149-ICS – Novartis (NVS)

We expect QAB149-ICS to take a strong share of the fixed dose segment, and currently project it to come out slightly ahead of the GSK159797-ICS combo. As noted above, QAB149 is ahead of GSK159797 in development and likely has a faster onset of action, though it is unclear what type of delivery device will be used and which ICS QAB149 will be combined with. Asmanex is a good possibility. And while we are concerned about the latter’s 24-hour ability in more severe patients, it should do fine in patients with mild and more moderate disease.

Our 5- and 10-year projected US revenues for asthma in the MRT scenario are $275.1 million and $1.3 billion, with corresponding worldwide revenues of $593.3 million and $2.9 billion. In the fixed dose scenario, the 5- and 10-year projected US revenues are $403.9 million and $1.9 billion, with corresponding worldwide revenues of $870.8 million and $4.2 billion.

In COPD, our 5- and 10-year projected US revenues are $111.4 million and $655.6 million, with corresponding worldwide revenues of $230.2 million and $1.5 billion.

GSK159797-ICS – GlaxoSmithKline (GSK) and Theravance (THRX)

As noted above, we expect GSK159797 to take a second position in the fixed dose combination market behind a QAB149-ICS combination. Since current information on the drug and its competitors is preliminary, we have modeled it only slightly behind the QAB149-ICS combo.

Our 5- and 10-year projected US revenues for asthma in the MRT scenario are $159.5 million and $1.0 billion, with corresponding worldwide revenues of $337.8 million and $2.3 billion. In the fixed dose scenario, the 5- and 10-year projected US revenues are $239.3 million and $1.6 billion, with corresponding worldwide revenues of $507.7 million and $3.5 billion.

In COPD, our 5- and 10-year projected US revenues are $50.2 million and $536.4 million, with corresponding worldwide revenues of $103.9 million and $1.2 billion.

Xopenex HFA - Sepracor (SEPR) and 3M (MMM)

Though there is some evidence of a benefit for the nebulized formulation over racemic albuterol, Xopenex HFA has not shown a benefit in clinical testing. The HFA formulation is also likely to be used in different settings from the nebulized version (the nebulized formulation is mainly used in children and older patients, as well as in exacerbations, where side effects are particularly important).

In addition, through the near future, Xopenex HFA will compete with low cost generic albuterol inhalers. Once the CFC ban occurs at the end of 2008, only HFA inhalers will be allowed, of which there are currently four. However, we are assuming that more generic HFA inhalers will emerge, such that Xopenex will continue to have to compete against lower cost albuterol generics, though it is not as clear what the price differential will be. We are also assuming that a generic version of Xopenex will emerge after expiration of its 2013 patent (the nebulized formulation already has two patent challenges).

As a result of these factors, we do not think the HFA formulation will achieve as high a share as the nebulized, and project a peak share of 15% under the fixed dose ICS-LABA scenario that is currently approved. Under Maintenance and Reliever Therapy, less short-term rescue medication would be needed, so in that scenario (our base case) we have modeled slightly reduced Xopenex sales in asthma. Also, before the CFC ban, we are projecting a somewhat smaller peak share for Xopenex.

We have updated the COPD model to reflect actual pricing of Xopenex since its recent introduction. We have also reduced the revenue from COPD based on changed assumptions of the split of usage of the short-acting beta agonists between asthma and COPD.

Our 5- and 10-year projected US revenues for asthma are $213.8 million and $112.8 million. We are not currently projecting overseas sales.

In COPD, our 5- and 10-year projected US revenues are $64.9 million and $32.5 million.

Xopenex Inhalation Solution – Sepracor (SEPR)

While the nebulized formulation of Xopenex has seen good growth, particularly in the hospital and institutional setting, we believe it is now beginning to level off. Though Xopenex has shown some advantages, trial results are mixed and there are questions about whether it is worth the additional cost. In addition, patents on the compound expire between 2010 and 2013 (there is a formulation patent until 2021). Our model presumes generic competitors entering towards the end of 2013.

Our 5- and 10-year projected US revenues for asthma are $522.2 million and $166.4 million.

In COPD, our 5- and 10-year projected US revenues are $174.8 million and $51.2 million.

Xolair – Genentech (DNA), Novartis (NVS), Tanox (TNOX), and PDL BioPharma (PDLI)

We expect Xolair to continue its good growth, since it is currently without competitors in patients refractory to steroids. We think due to its mechanism of action involving IgE, which is important in atopic patients, it will retain its share once the anti-TNF antibodies enter.

Our 5- and 10-year projected US revenues for asthma are $1 billion and $1.4 billion, with corresponding worldwide revenues of $2 billion and $2.8 billion.

Enbrel – Amgen (AMGN) and Wyeth (WYE)

Enbrel is most likely to be used in the small segment of patients refractory to corticosteroids. Given the limited information on its efficacy, we are currently cautious about projections. We think it will lag behind the more conveniently dosed Golimumab, though still have a relatively significant share of this segment of the market. There is still the risk of other anti-TNF drugs entering this market.

Our 5- and 10-year projected US revenues for asthma are $180.1 million and $496.2 million, with corresponding worldwide revenues of $281.8 million and $1 billion.

Golimumab – J & J (JNJ), Schering Plough (SPG), and Medarex (MEDX)

We project Golimumab to gain share above Enbrel, due to its convenient dosing. However, it will be important to see clinical data ensuring that it is as effective.

Our 5- and 10-year projected US revenues for asthma are $311.7 million and $856 million, with corresponding worldwide revenues of $489.6 million and $1.7 billion.

Inhaled Bimosiamose – Revotar and Encysive (ENCY)

We are projecting that this agent could be used in a broader population than the anti-TNF drugs, but we think it will also be priced closer to the regular asthma market. Given that the clinical information is very preliminary, we are currently relatively conservative on our market share estimates.

Our 5- and 10-year projected US revenues for asthma are $174.9 million and $608.6 million, with corresponding worldwide revenues of $280 million and $1.4 billion.

Short-acting Inhaled Bronchodilators

With their quick onset of action and short duration, inhaled short-acting beta agonists are the drug of choice for as needed use both in patients with intermittent asthma and those on maintenance medications. They are also important in treating exacerbations in the emergency room and hospital setting. Short-acting beta agonists are the most widely sold inhaled drug.

This segment has been dominated by generically available albuterol (the MDI formulation is the most common, with 4 times as many prescriptions as the nebulized version). Most of the albuterol MDIs contain CFC propellants, and will be banned after December 31, 2008. To date, there are two branded formulations of albuterol with the new HFA propellant (Proventil - Schering-Plough and Ventolin - GlaxoSmithKline) and one branded generic (from IVAX). These are generally priced $20 or more over the CFC generics, and so they currently have limited market share. While it is unclear whether 3M, which holds the HFA patents, will license the technology for more generic albuterol HFA competition, this is a likely possibility, which would limit the expansion of the branded HFA formulations after the CFC ban.

Xopenex (Sepracor) is a special albuterol formulation, available as a nebulizer solution since 1999 and an HFA MDI since December 2005 (though approved in March, manufacturing issues caused a delay in bringing it to market). Specifically, it is the R enantiomer of albuterol (albuterol – referred to as racemic albuterol – is a mix of two enantiomers, molecules that have the same chemical makeup but are mirror images of each other). The rationale for Xopenex is that the R enantiomer is the more active form, and there are some concerns that the S enantiomer may increase airway reactivity, though that has not been convincingly documented.

There has been controversy whether nebulized Xopenex provides significant clinical benefits over nebulized racemic albuterol. For example, though some studies showed low doses of Xopenex had comparable efficacy with much higher doses of racemic albuterol, they have been criticized for the possibility that patients were already on the upper, flat portion of the dose response curve, where additional benefits would not be expected with the higher doses. One major study found that Xopenex used in the emergency room for asthma exacerbations led to a lower rate of hospitalizations than racemic albuterol (36% vs 45%, p=0.02), but another study found no difference. While Xopenex appears to be at least as good as racemic albuterol, due to the conflicting evidence one major question has been whether the results are definite enough to justify the greater cost. As a result, while Xopenex has seen solid market uptake in the nebulized medication segment, achieving 27% of prescriptions and 33% of hospital units sold, it has not supplanted generic nebulized albuterol in the majority of patients.

For the recently introduced HFA formulation, pulmonary function tests have been similar to equivalent doses of albuterol HFA. Without more data showing an outcomes benefit, we do not think there will be a compelling clinical argument in favor of Xopenex HFA, though its theoretical advantage and the possible benefit seen with the nebulized form are likely to help it do better than branded albuterol.

Anticholinergics are less effective in asthma than they are in COPD. According to the NHLBI guidelines, short-acting anticholinergics have not been shown to add benefit to beta agonist treatment for long-term maintenance therapy in asthma. As noted above, short-acting anticholinergics are recommended as add-ons to beta agonists for more severe exacerbations, for reliever use in patients who cannot tolerate beta agonists, and for patients whose asthma is triggered by beta-blockers. Atrovent (Boehringer Ingelheim) is the sole single-agent short-acting anticholinergic and Combivent (Boehringer Ingelheim) the sole combination one for use in an MDI inhaler in the US (Combivent combines the active ingredient of Atrovent plus albuterol). Both are available generically as nebulizer solutions. In the overall respiratory market, prescriptions for anticholinergics are about 30% of those for beta-agonists.

Long-acting Inhaled Bronchodilators and Combinations with ICSs

There are two long-acting inhaled bronchodilators approved for asthma; both are twice-a-day beta agonists. Due to their greater effectiveness than short-acting bronchodilators, long-acting beta agonists (LABAs) have been recommended as the preferred add-on treatment to inhaled corticosteroids in patients with moderate asthma (though there has been only limited study of their benefit in pediatrics). However, sales of individual LABAs are a much smaller part of the market (together less than 5% of asthma maintenance prescriptions) because a combination product of a LABA and ICS, Advair (see below), has generally been preferred.

Serevent (GlaxoSmithKline) is the oldest single agent LABA, approved in 1994. Its chemical name is salmeterol, and the active part of the molecule is the same as for albuterol. The patent on the active ingredient of Serevent is due to expire in 2008 in the USA and expired in 2005 in most of Europe. Serevent is one of the components of Advair.

Foradil (Schering-Plough, Novartis, Astellas), or formoterol, was approved in 2001 in the US, and has a faster onset of action than Serevent. However, Foradil sales have been hampered because the device is less convenient to use: unlike Serevent, blisters containing the medication must be put into the device one at a time. The patents and exclusivity for Foradil’s active ingredient have expired in the US and other major countries. Another competitor with the same active ingredient, Oxis (AstraZeneca), is available in Europe, where a generic version has also appeared.

Advair (GlaxoSmithKline) is a combination of the LABA Serevent and the ICS Flovent (see below). A DPI formulation was first approved in the US in 2000 for patients 12 years and older, with expanded use to children age 4 to 11 approved in May 2004. (An Advair non-CFC MDI, already approved in Europe, received approvable letters from the FDA in 2001 and 2002. GSK expects it to be approved and introduced in the US in the third quarter of 2006.)

Advair has shown significant effects on pulmonary function and symptoms beyond those compared to either component alone, and at least as good as that of its individual components used together. In addition, the symptomatic relief patients feel with the LABA may increase compliance above an ICS alone. The clinical benefits and convenience of Advair has propelled it to the best selling (in terms of revenue) pulmonary drug, with US and worldwide sales of $2.4 billion and $4.5 billion respectively in 2004. The majority of the sales are due to asthma. GSK’s patent on the specific combination of active ingredients in Advair expires in 2010 in the US and 2013 in Europe. While there may be more difficulties in developing generic inhaled combination products compared to oral ones, we believe the market is so large that a generic substitute for Advair will arise.

The SMART Trial
LABAs have come under increased scrutiny since Serevent was associated with an increase in mortality in the SMART trial (Salmeterol [Serevent] Multi-center Asthma Research Trial). The trial compared Serevent to placebo when added to usual asthma treatment and was stopped early after an interim analysis. The study found a small but significant increase in both asthma- and respiratory-related mortality (13 versus 3 deaths and 24 versus 11 deaths, respectively, out of more than 13,000 patients in each group). The negative results were seen primarily in the African-American subgroup and those not using ICSs. It is still unclear whether the excess mortality seen in the SMART trial was due to Serevent itself, a class toxicity of the LABAs, a genetic predisposition to react to these drugs in a certain way, or Serevent masking worsening inflammation such that patients delayed seeking necessary additional treatment. Studies on the genetic issue and in African-Americans are ongoing.

Interim data from the trial led to an initial black-box warning for Serevent and Advair in 2003. After the FDA re-analyzed the data, an FDA Advisory Committee Meeting was convened in 2005, which recommended that the long-acting beta agonists remain on the market, but with increased warnings on their labels. On November 18, 2005, the FDA requested changes in the label for Serevent, as well as Advair (which contains the active ingredient of Serevent) and Foradil (since it could well be a class effect), recommending that these agents only be used for asthma patients failing corticosteroids. Though initially criticizing them, GSK agreed to the new labels for Serevent and Advair.

The manufacturers of Foradil did not agree to use the label, and the FDA is apparently not pursuing the issue further (they had sent out a warning to physicians on Foradil along with the one they sent for the other two). While Foradil itself has not been associated with an increase in mortality, several studies have raised the concern of serious asthma exacerbations when a high dose of the agent is administered. A 2,000 patient study was designed to answer this and did not find a significant deleterious effect for Foradil. However, the patients had less severe asthma than in the SMART study and a higher proportion were taking ICSs.

Despite the concerns raised by the SMART trial, we feel the medical community views combination treatment with ICSs and LABAs (including Advair) as safe for non-African-Americans. Many patients need the extra treatment with LABAs to control their symptoms. Hence while we think that growth in Advair sales will slow as doctors are more cautious in some patients (particularly African-Americans); without further negative data, we do not think there will be a dramatic reduction in Advair sales.

On the other hand, without further evidence, it is unlikely that the FDA will accept the view that combined use of LABAs with ICSs is safe, particularly since patients in the ICS subgroup of the SMART trial were not separately randomized. Hence the issue could crop up in future approvals.

Inhaled Corticosteroids (ICS)

Though glucocorticoids are effective at reducing inflammation (such as by decreasing the number and availability of inflammatory cells and inhibiting the production and release of chemicals that mediate the inflammatory response), they also have numerous side effects in the body. As a result, oral corticosteroids are recommended only for short periods to treat exacerbations or when inhaled corticosteroids are not adequate. ICSs, on the other hand, are generally thought to be safe for long-term use, since they are inhaled and less drug reaches the circulation to cause side effects. While some systemic effects of ICSs have been seen, the clinical implications are unclear. ICSs used in asthma have been associated with reduced bone mineral density in adults, but it is not known whether this increases the risks for fractures. High cumulative lifetime doses may slightly increase the risk for cataracts. Low to medium doses may have the potential to decrease growth in children, but the effects are small, do not appear to progress with further treatment, and may be reversible. Finally, there may be slight suppression of the body’s normal mechanism for producing natural hormones similar to corticosteroids (hypothalamic-pituitary-adrenal axis function). This is not thought to be clinically significant.

Pharmacological properties of different ICSs influence how effective they are in the respiratory system and the degree of systemic side effects. For example, particle size and formulation influences how much gets to the lung after an inhaled dose and how much is deposited in the mouth, where it can be swallowed and potentially have a systemic effect. Even if a drug is swallowed, if it is metabolized to a high degree in the liver (resulting in what is called low bioavailability), it will also have less systemic absorption. Products that dissolve better in lipid solutions (lipophilicity) than in water are thought to have a more prolonged action in the lung.

Due to their advantage in factors such as these, Flovent (GlaxoSmithKline) and Pulmicort (AstraZeneca) are the most commonly used single agents, with 45% and 30% of single-agent ICS prescriptions respectively. Most of Pulmicort US sales are for the nebulized formulation, which is particularly useful for smaller children. Since these two are the most potent ICSs, fewer doses per day can be used for the same clinical benefit as less potent ICSs. In addition, the active ingredient of Flovent is used in Advair, and the active ingredient of Pulmicort is being developed as a combination product with the long-acting beta-agonist formoterol (see Symbicort below). To put single agent ICSs use in context, the combined use of all single agents is half that of Advair.

Asmanex (Schering-Plough) is a long-acting corticosteroid delivered by DPI, which unlike current formulations is usually dosed once-a-day (twice-a-day for the highest dose). It was approved for asthma in March 2005, and was launched late in the third quarter. Schering-Plough started with a limited launch targeting specialists and only recently began to market to primary care physicians. Of approved ICSs, Asmanex has the highest affinity for the corticosteroid receptor. It has a number of pharmacologic properties thought to reduce systemic effects, though it shares these with other competitors. Its major point of differentiation is convenience. There have not been a great deal of comparative trials, and the high once-daily dose of Asmanex is listed in international guidelines as a low medium dose. In addition, that dose (400 micrograms/day) led to numerically worse (but not statistically significant) FEV1 than a moderate dose of Flovent (250 micrograms twice a day). Hence without further information, we think it is unclear whether patients who need moderate doses at the higher end can be effectively treated with the once-a-day dosing.

Nevertheless, the convenience of the once-a-day dosing will be a plus for Asmanex in patients with milder disease who only need an ICS (without a LABA). We think it will steadily gain popularity as doctors try the product and become comfortable with it, though it will fact competition from Alvesco (see below). Asmanex may benefit slightly from the fallout from the Serevent SMART trial (discussed in the section on LABAs) in these patients, though we think there will only be a modest effects on Advair sales, as many patients benefit from the addition of the LABA to the ICS. A combination of the active ingredients of Asmanex and the LABA Foradil, Formoterol-Mometasone, (Novartis, Schering-Plough, Astellas), is in Phase I development for COPD and Phase II for asthma. However, formoterol requires twice-a-day dosing, so this product would not have added convenience. The onset of action of formoterol is quicker than that of the LABA in Advair, but the same as that in the combination product Symbicort, which should have the advantage of reaching the market sooner.

Leukotriene Modifiers (LTRAs)

These agents affect leukotrienes, which are released from inflammatory cells and have a number of actions, including airway smooth muscle contraction and the attraction of more inflammatory cells. Singulair (Merck) and Accolate (AstraZeneca) block leukotriene receptors, and Zyflo (Critical Therapeutics and Abbott) inhibits the formation of leukotrienes. They are all given orally.

While no definitive difference in efficacy has been demonstrated for the three, Singulair only needs to be given once a day, has a better side effect/drug interaction profile, and is available as a chewable tablet for children. As a result, it dominates the sector, with 97% of prescriptions.

While ICSs generally outperform LTRAs in clinical trials, there may be more compliance with LTRAs in children, since as oral drugs they are easier to take. This has helped make Singulair quite popular: it is the most widely prescribed asthma maintenance medication in the US (it was also approved for seasonal allergic rhinitis in 2002 and perennial allergic rhinitis in July 2005). We do not see a near term challenger.

The US patent protection on Singulair ends in February 2012, with pediatric exclusivity until August of the same year.

Anti-Immunoglobulin E (Anti-IgE)

Xolair (Genentech, Novartis, Tanox, PDL) is a monoclonal antibody that binds IgE, and is the sole such agent approved for asthma. By binding IgE, Xolair prevents it from interacting with IgE receptors on mast cells and basophils, thus reducing the release of allergy mediators. Xolair was approved in the US in 2003 and in Europe in October 2005. It is given as a subcutaneous injection every 2 or 4 weeks, and due to a small risk of a severe allergic reaction, it is recommended that doses be given in a physician’s office. Dosing is based on a patient’s baseline IgE levels. It is approved for patients 12 and above, who have moderate to severe persistent asthma inadequately controlled with ICSs (only approved for severe in Europe), as well as a positive test for sensitivity to a year-round airborne allergen.

Xolair decreases the incidence of asthma exacerbations and emergency room visits in patients who are not controlled by ICSs. (Exacerbations were not reduced in a study of patients on oral corticosteroids.) There is question about whether it increases the risk of malignancies: though showing numerically higher rates, the difference from placebo has been small, and the significance is unclear.

Due to such considerations, as well as high cost, Xolair is primarily used in more severe cases of asthma, which represents less than 5% of the asthma population. Nevertheless, it has shown strong revenue growth, and we expect this to continue, though it will only be used in a small percentage of patients.

DRUGS IN DEVELOPMENT

DRUGS IN NDA

Symbicort - AstraZeneca (AZN)

Symbicort, a competitor for Advair, is a combination of the long-acting beta agonist formoterol (the active ingredient in Foradil and Oxis - the latter is AZN’s drug) and the ICS budesonide (the active ingredient in Pulmicort). It is a twice daily formulation, being developed for an MDI in the US (according to AstraZeneca, the Turbohaler DPI formulation is not being developed in the US due to the regulatory burden regarding manufacturing issues). Currently, an NDA was submitted in September 2005 for asthma (it is in Phase III development for COPD). Patents are due to expire 2012 in most major European countries and in approximately 2014 in the US.

The beta-agonist component of Symbicort has a more rapid onset of action than that in Advair. In Europe, where Symbicort is already approved, this has not given it enough of an advantage to supplant Advair. However, AstraZeneca has used this difference -- along with the fact that unlike the case with the LABA in Advair, increasing the dose of formoterol beyond standard doses provides an additional benefit -- to develop what it calls maintenance and reliever therapy (MRT). Instead of a fixed dosage, as is currently the standard with maintenance medications, maintenance and reliever therapy involves a lower fixed maintenance dose, with patients increasing the dosage of the same medication as needed for relief of symptoms (according to a prespecified plan).

In Europe, Symbicort has submitted maintenance and reliever dosing for approval in September 2005 (a prior submission in 2003 where this was called Single Inhaler Therapy was not successful, and more studies were required by the regulatory authorities). Although AstraZeneca IR representative has stated that they are not revealing the type of dosing they submitted in the US NDA (or details on the trials they are using to support the submission), at their 2005 fourth quarter earnings call, the director for development mentioned they had submitted an NDA for fixed dosing. Still, it is unclear if this means that a flexible dosing regimen was not submitted as well.

There have been a number of studies showing MRT resulting in fewer exacerbations than standard fixed dosing. A couple of trials have compared Symbicort MRT to fixed dose Advair. These studies have showed a large benefit in endpoints such as reducing exacerbations. However, the direct comparison trials for which data are available have also been open-label. (A blinded comparison was done for the EMEA resubmission, but data have not been released: we assume it was positive.) On the other side, a blinded study comparing Symbicort MRT and Advair, sponsored by GSK, showed the opposite, with Advair coming out ahead with a large benefit. However, in this study, patients could reduce their Symbicort dosage to only 1 puff per day (rather than 2 as in other studies). This most likely put them at risk for exacerbations.

The increased use of LABAs with worsening symptoms in Symbicort MRT goes counter to the concerns raised by the SMART trial over the potential deleterious effects of LABAs, and raises several questions. (Ironically, the full acronym for Symbicort Maintenance and Reliever Therapy is SMART, the same as that for the fateful Serevent trial.) It is unclear which component of Symbicort accounts for the purported benefit in MRT. One trial done for the European resubmission looked at whether it was the LABA component, but data have not been released. If that trial showed that a LABA as a reliever on top of an ICS reduced exacerbations to the same extent as Symbicort MRT, it would conflict with concerns raised by the SMART trial. (It would imply that it was the LABA component that provided the benefit of MRT. One hypothesis regarding the SMART trial is that though LABAs added to ICSs reduce exacerbations, in other situations where asthma is poorly controlled, LABAs might give temporary symptomatic benefit but also delay needed treatment, resulting in increased mortality. The Symbicort trial may not have been large enough to detect this.) As a result, in this situation, we think regulators would be hard pressed to approve the drug without further study. If the resubmission study did not find that, it is still unclear whether adding an ICS alone on top of fixed combination treatment is safer than using the combination treatment as a reliever. (Control groups for the MRT studies generally did not involve increasing ICS with worsening symptoms. Also, while other trials where the single-agent ICS has been doubled in exacerbations -- which is commonly recommended -- have found no benefit, most patients in those trials did not receive any LABA treatment, which conceivably could affect the results). The GSK study, where lower mean doses of Symbicort were ultimately used and where Symbicort MRT did worse than fixed dose treatment, raises the question of whether it is too risky to leave dose adjustments to patients. Finally, there is a general question of whether using the lower doses of ICSs in MRT could lead to more problems down the road due to less suppression of inflammation.

Despite these concerns, we feel overall that Symbicort MRT is an attractive therapy and the weight of evidence is in its favor. However, in contrast to a number of opinions expressed in the medical literature, the FDA has not appeared to accept the notion that LABAs in conjunction with ICSs are safe without further evidence, so even if MRT has been submitted as part of Symbicort’s NDA, we think it is unlikely it will be approved without further study of this issue. In Europe, the additional trials required after the first submission were most likely already ongoing when more information regarding the SMART trial came out due to the FDA reanalysis of data. Hence the additional studies done for resubmission will not necessarily resolve the issue for the EMEA, either, though we think they are more likely to approve MRT than the FDA. If Symbicort MRT were to be approved, we feel it is likely to become the dominant therapy in patients with less severe disease, and would likely propel Symbicort into a leadership position. As a result, we think AstraZeneca will continue to pursue approval of MRT, though it is possible that they will opt to try and gain a marketing advantage over Advair with their fixed dose, hoping physicians will prefer a different option due to concerns raised by the SMART trial. Also, if the European authorities approve the drug first and the FDA requires major new studies, AstraZeneca may delay pursuing the US studies because a negative result could jeopardize the marketing of MRT in Europe.

The PDUFA for Symbicort is July 23, 2006. Even if AstraZeneca has not applied for Maintenance and Reliever Therapy, it is very likely approval will be delayed. When asked about approval at the recent earnings call, AstraZeneca’s CFO noted that the rate of approval for respiratory devices is only 10% the first time through, and most drugs can expect at least 2 cycles of regulatory submissions.

Foradil Certihaler - Schering-Plough (SGP), Novartis (NVS), SkyePharma (SKYE)

Foradil Certihaler is a multidose dry powder formulation of Foradil Aerolizer, which is a single dose DPI. Hence the advantage of the new formulation is convenience of the multi-dose feature. Though approved in a number of countries in Europe, it has received two approvable letters in the US and is currently in NDA status for both asthma and COPD. Details on what is holding up approval have not been released. In any case, if approved, while it is likely to do better than and take share from Foradil (and Serevent), single agent LABAs have a relatively limited market potential in asthma compared to combination medications.

DRUGS IN PHASE III DEVELOPMENT

Alvesco – Altana (AAA) and Sanofi-Aventis (SNY)

Alvesco is a once-a-day corticosteroid (twice-a-day for high doses) in Phase III development. It is delivered by an HFA MDI. Alvesco is a prodrug that is activated in the lungs. It scores high on the list of pharmacologic properties of a desired ICS. In addition to convenience, a significant advantage is a lower incidence of oral fungal infections, a side effect of corticosteroid deposition in the mouth. However, of concern was a study of high doses showing an increased rate of nuclear cataracts (cataracts in the center of the lens) with high dose Alvesco (2.4% for Alvesco 160 microgram twice-a-day; 6.2% for Alvesco 320 micrograms twice-a-day, 0.7% for Flovent 440 micrograms twice-a-day, and 0.7% for placebo). While nuclear cataracts are not the type most typically associated with steroid treatment (ie, subcapsular), they too have been implicated. On the other hand, it is puzzling why Alvesco should cause cataracts, since on other measures it appear to have lower systemic effects.

Alvesco received an FDA approvable letter in 2004 as further clinical information was needed. One trial evaluating the eye effects of higher dose Alvesco has already been completed, but the data have not yet been released. However, representatives from both Altana and Sanofi have stated that results were positive for safety of the drug. Two additional trials, 12 week and 16 week studies looking at once-a-day versus twice-a-day dosing, were instituted this year and are slated to be finished by August 2006. Alvesco was approved in its first European country in April 2004, but only for the lower dosages of the drug (160 micrograms once a day was the top dose approved).

The main negative issue for Alvesco has been the higher rates of cataracts seen in the one trial. We are encouraged by the companies’ statements, but remain somewhat cautious prior to seeing the actual results. If the recently completed trial indeed confirms that patients are not at risk, we feel Alvesco would do well against other ICSs, due to its lower systemic effects and once-a-day dosing. For it to be successful, the higher doses need to be approved as well, which we feel will happen if safety is demonstrated.

Altana and Sanofi are also developing a combination DPI product using Alvesco and a LABA, Formoterol-Ciclesonide, currently in Phase IIb for asthma (results for the IIb trial are expected before the end of the year, and regulatory submission is planned for 2007/2008). Again, we feel approval of the higher doses hinges on the safety data. The use of formoterol will require twice-a-day dosing, negating a convenience advantage. However, with the rapid onset of action of formoterol and the low systemic effects of ciclesonide, we feel this will still be a strong competitor, which will also be suitable for Maintenance and Reliever Therapy. On the other hand, if it turns out there are elevated risks for cataracts at higher doses of Alvesco, we do not think those doses will be approved, and so Alvesco's uptake will be limited.

Daxas – Altana (AAA)

Daxas is a once-a-day oral phosphodiesterase-4 (PDE4) inhibitor in Phase III development for asthma and COPD. It is thought to work by reducing the activity of a variety of inflammatory cells. Of note, the methylxanthine Theophylline is also a phosphodiesterase inhibitor, though it is not specific to PDE4.

Daxas has shown similar effects on FEV1 compared to a low dose of an inhaled steroid (over 12 weeks) and compared to Singulair (over 6 months). However, it only showed a trend towards improvement when added on to low-dose Flovent.

In July 2005, Pfizer terminated its agreement with Altana after a COPD trial failed to show a benefit in overall exacerbations. And in November 2005, Altana withdrew its application for Daxas with the EMEA, pending further clinical data, despite the fact that it apparently had thousands of patients in its Phase II and III studies. These are obviously perturbing developments. In addition, though Daxas has shown some benefit in some measures, we are concerned that we have not seen mention of its ability to reduce exacerbations in comparison with agents such as inhaled corticosteroids or Singulair, or its ability to reduce ICS use. Hence we are cautious in our outlook on Daxas for asthma.

DRUGS IN EARLY DEVELOPMENT

There are a variety of drugs in early development. In addition to improved LABAs and ICSs, there are a number of immunosuppressant drugs. While some of these have interesting clinical information, we have also included others based solely on the novelty of their mechanism of action to give a better idea of what is in the pipeline.

Beyond Advair Collaboration – GlaxoSmithKline (GSK) and Theravance (THRX)

The Beyond Advair collaboration between Theravance and GlaxoSmithKline began in 2002 with the ultimate purpose of developing a once-daily LABA to use in a replacement product for GSK's Advair, which is given twice-a-day. Each company contributed four LABA product candidates.

GSK will provide the ICS for the combination product, most likely 685698, which is in Phase II development for asthma and also Phase III for allergic rhinitis under the name Allermist. GSK also has another ICS in Phase II development, 799943. Little information about these compounds has been released. As single agents, these ICSs will likely compete against Asmanex and Alvesco.

The lead once-daily LABA of the Beyond Advair Collaboration is GSK159797, in Phase II development. It is to be delivered via DPI. Though there were some formulation issues causing a delay in 2005, these were apparently resolved, and the company announced plans to initiate a Phase IIb trial in the first half of 2006.

Some clinical information has been released: in a trial of 20 patients, GSK15979 showed increases in FEV1 at 24 hours of 460 ml to 540 ml, compared to an increase of 130 ml for placebo. These are in the range of what has been seen at 12 hours with existing LABA treatment in asthma, though on the low end. Importantly, the heart rate was only minimally increased by the drug (a potential side effect of beta-agonists).

It is too soon to tell whether the drug will be successful in achieving 24-hour control for asthma. Given the concerns from the SMART trial, the product will come under increased scrutiny, though we think the main factor will be whether the product can provide 24-hour coverage comparable to what the twice-a-day products now give. The largest market opportunity is in combination with a once-a-day ICS. The convenient dosing will, of course, provide a large advantage over twice-a-day products. Of concern, GSK159797’s onset of action is likely more similar to that of Serevent, and though little data has been released, we think likely slower than that of another competitor in development, QAB149 (see below). Furthermore, QAB149 is now ahead of GSK15979 in development (though Novartis is not specifying whether the same delivery device will be used for the combination QAB149 product). It is not yet clear how these two drugs stack up against each other on other measures, and also with which 24-hour ICS QAB149 will be combined. A likely partner candidate for QAB149 is Asmanex, though as previously noted, it is not clear how patients with more severe disease fare on Asmanex’s once-a-day dosing compared to twice-a-day drugs. There is little information on the GSK 24-hour ICSs. Though there are pros and cons for each and the competition could be close, based on QAB149’s likely faster onset of action (and potential earlier time to market), we currently think it will come out ahead.

As a 24-hour drug, we believe GSK15979 is likely to be less suited for maintenance and reliever therapy (if MRT is approved), and as a result, we believe it would be restricted to a smaller market share.

There are two other LABAs from the collaboration now in Phase IIa development, as shown in the table below. The table also shows which company contributed the compound (though Theravance receives the same royalties on all the drugs, the milestone payments vary).

QAB149 – Novartis (NVS)

Like GSK159797, QAB149 is a once-daily inhaled beta-agonist. It has completed several Phase II trials in asthma and COPD, and Phase III trials are expected to begin in the coming months. The drug has been tested in the past using SkyePharma’s powder formulation, but Novartis is not specifying what type of delivery device will be used in phase III. QAB149 appears to have a relatively rapid onset of action. In 42 patients with intermittent or persistent asthma, the increase in FEV1 at 24 hours was 10.6% higher than for placebo. While on the order of some data on twice-a-day LABAs, the effect appears on the low end. Like GSK159797, there is not enough information to tell if the drug will be successful for 24 hour control of asthma. If it is, it would have to be combined with an ICS to gain a large market share, which Novartis has stated it is exploring. As noted above, Novartis has already partnered with Schering to develop a Formoterol (LABA)-Mometasone (ICS) combination. Since mometasone (unlike formoterol) can be given once-a-day, it would be a logical choice to combine with QAB149 for a once-a-day combination product. While it is unclear how well once-a-day mometasone will fare in patients with more severe disease, it is likely to be fine in patients with milder and more moderate disease. Another option is Alvesco. Given that QAB149 likely has a more rapid onset of action than GSK15979, we currently think the QAB149-ICS combo will come out ahead in the market. In addition, single-agent QAB149 is currently ahead in development, thought it is unclear whether the combination product will use the same delivery device. Like GSK15979, as a 24-hour drug, we believe the QAB149-ICS combo is not likely to be suited for maintenance and reliever therapy if that is approved, and hence, it would be restricted to a smaller segment of the market. If MRT is not approved, we believe the combo could become a market leader.

Flutiform – SkyePharma (SKYE)

Flutiform is a combination of the inhaled corticosteroid fluticasone with the long-acting bronchodilator formoterol, being developed as an HFA metered-dose inhaler for asthma using a fixed dosing schedule. It has finished its Phase II trial and a Phase III trial is to begin this year (originally planned for January). SkyePharma plans to submit their NDA by mid-2007, and is hoping for approval in the first half of 2009. Of note, the company has undergone some internal turmoil recently after receiving an unsolicited bid and demands from a group of institutional investors to install a director of their choosing. The company underwent a strategic review and replaced its top management. An agreement on directors was recently reached, but the company is selling off a division, apparently for needed cash to keep Flutiform their top focus in development.

Since the components of Flutiform are well-established drugs, it is likely the combination will work and be as effective as others on the market. However, while this is a strong combination of what are some of the best agents currently in their classes, it is unclear whether Flutiform offers a significant enough clinical benefit to supplant well-established competitors. Also, we think the Formoterol-Ciclesonide combination will be preferred. The company is not currently planning to develop Flutiform for maintenance and reliever therapy, though it is likely to be suited for this, and if that should that become popular, as we believe, additional trials would be needed. In the fixed-dose segment, Flutiform will likely need to compete after a few years with the more convenient once-a-day preparations now in Phase II development. Finally, the internal turmoil at the company could delay development. Thus overall, we are currently pessimistic about Flutiform’s prospects.

Formoterol-Mometasone – Novartis (NVS), Schering-Plough (SGP), and Astellas

This combination of the LABA formoterol and the active ingredient in the ICS Asmanex is in Phase II development for asthma and Phase I for COPD. Mometasone can be given once a day, but formoterol requires twice-a-day dosing. As a result this product would not have added convenience over the current leader Advair. The onset of action of formoterol is quicker than that of the LABA in Advair, but the same as that in the combination product Symbicort, which should reach the market sooner. Hence Formoterol-Mometasone will suffer the same latecomer competitive pressures as Flutiform. We also think the Formoterol-Ciclesonide combination will be preferred.

Enbrel – Amgen (AMGN) and Wyeth (WYE)

Enbrel is a recombinant fusion protein consisting of the constant region of a human antibody grafted to the binding portion of the human tumor necrosis factor (TNF) receptor. The TNF receptor part of Enbrel functionally acts like an anti-TNF antibody, binding to TNF (both alpha and beta forms). TNF is an important cytokine (an immune regulatory protein) that is involved in a number of the body’s inflammatory and immune responses. There are high levels of TNF-alpha in fluid from the respiratory tract of patients with severe asthma.

Enbrel is approved for rheumatoid arthritis and other rheumatic diseases. It is given as a subcutaneous injection once or twice a week. Though development in asthma had previously been suspended in 2003, it is now again in Phase II. Recently, an investigator-initiated study was published providing evidence of a role for TNF-alpha in patients with asthma refractory to other treatment (including corticosteroids). Enbrel also improved FEV1 by 320 ml compared to placebo over 10 weeks. Changes in markers of TNF-alpha were correlated with a measure of bronchial hyperreactivity. Though small, the study provided intriguing data on the potential benefit of Enbrel in refractory asthma patients. There was some preliminary evidence that markers of TNF-alpha were not elevated in patients with more moderate disease, so Enbrel’s effects may be limited to the subset of patients with severe asthma. Since Enbrel affects the body’s immune response, concerns over potential side-effects such as infection is likewise likely to limit its use to the more severe patients. Significant safety studies will need to be done, as these patients are also on corticosteroids, which likewise suppress the immune system.

We are encouraged by the preliminary data for Enbrel in asthma. A 12-week Phase II study began in May 2005, but results have not been released. One source of market risk for Enbrel is that there are already other anti-TNF antibodies on the market and in development for rheumatic conditions which could also be developed for asthma. For example, Golimumab (see below) has already seen clinical testing in asthma -- and has a more convenient once-a-month dosing. While we think that Golimumab will be favored, its relative efficacy remains to be seen. We also think there could be a number of patients who respond best to Enbrel or do better on its weekly dosing.

Golimumab – Johnson & Johnson (JNJ), Schering-Plough (SGP), and Medarex (MEDX)

Golimumab is the fully-human anti-TNF-alpha monoclonal antibody version of Remicade in Phase II for asthma (and Phase III for rheumatoid arthritis). Remicade has sequences of mouse protein in it, which can lead to the formation of antibodies that reduce Remicade’s effectiveness and cause a hypersensitivity reaction -- hence the desire to create a fully-human version. A one-year Phase II trial began in August of 2004, but we have not seen any clinical data yet on Golimumab for asthma. Since a drug with the same mechanism of action, Enbrel (see above), has shown some preliminary evidence of efficacy, we are encouraged about the prospects for Golimumab. One factor in its favor is dosing: it is being developed as a once-monthly subcutaneous injection versus Enbrel’s weekly regimen.

Mepolizumab – GlaxoSmithKline (GSK) and PDL BioPharma (PDLI)

In Phase II development, Mepolizumab is a monoclonal antibody targeting interleukin-5 (IL-5), which regulates the production of eosinophils and their release from the bone marrow. Eosinophils are one of the prime cells implicated in asthma. Mepolizumab has been shown to reduce eosinophils in the airway and also the deposition of some proteins that may lead to more permanent airway changes due to inflammation. However, it did not reduce a measure of an allergic response to an inhaled antigen (the late phase response which occurs 4 to 8 hour after exposure to an allergen due to the infiltration of inflammatory cells). This has raised questions about the precise role of IL-5 and eosinophils in asthma. Thus, though quite preliminary, the data thus far are mixed, and suggest the drug would only have an effect on part of the asthma process. Also, as a monoclonal antibody, Mepolizumab has the inconvenience of needing to be given by injection or infusion. We are currently doubtful about the prospects of this drug.

Inhaled Bimosiamose – Revotar and Encysive (ENCY)

Bimosiamose is a selectin inhibitor in Phase II development. During inflammation, selectins on cells lining blood vessels (E- and P-selectin) and inflammatory cells (L-selectin) help the inflammatory cells leave the circulation and enter the site of inflammation. Bimosiamose blocks all three of the known selectin receptors.

In a small study of patients with mild asthma, Inhaled Bimosiamose reduced the late asthmatic response (drop in FEV1 3-8 hours after exposure to an inhaled allergen) by 50%. There was no effect on the early asthmatic response or on airway hyperresponsiveness. These results are difficult to compare with others due to differences in the patient population. However, in a study of Singulair and Flovent with a similar placebo response to the current study, the magnitude of the effect with Singulair was smaller and that with Flovent larger than that seen here. On the other hand, the percent improvement seen here with Inhaled Bimosiamose was similar to that seen with ICSs in other studies of patients with more severe disease. There is little information on adverse events.

Overall, the results for Inhaled Bimosiamose are encouraging, albeit preliminary. We like the idea of an inhaled immunosuppressant, since side effects are likely to be lower than the injectable ones, and we think it will be able to be used in a broader population. Of some concern, though, is the similar mechanism of action as the integrin inhibitors, which have been put on hold due to the risks seen with Tysabri (see R411 below).

R411 – Roche (RHHBY)

R411 is an oral dual alpha4beta1-alpha4beta7 integrin antagonist in Phase II development. These integrins are receptors on inflammatory cells (particular Th2 lymphocytes, eosinophils and mast cells). Like the selectins, they are involved in migration of inflammatory cells from the circulation to the site of inflammation, and may also be involved in cell activation.

In a twelve-week study, R411 was first added on to ICSs and then the ICSs were withdrawn. While patients showed a better response than placebo, they still experienced a decline in FEV1 with the withdrawal of ICS, suggesting that at the doses tested, the drug is not as effective as ICSs.

One major issue facing the drug is that the FDA had placed this class of drugs on hold pending its evaluation of another integrin inhibitor, Tysabri for multiple sclerosis, which was associated with several cases of a fatal neurologic disease. Since an FDA Advisory Committee recently recommended Tysabri be reintroduced, we think the research hold on the class will be lifted. At the same time, further development of the integrin inhibitors will come under a great deal of scrutiny. Given the potential risks seen with Tysabri, we think R411 will have to demonstrate impressive efficacy. Overall, we do not hold high hopes for this drug.

Kinase gene inhibitor - deCODE genetics (DCGN)

This drug inhibits MAP3K9 kinase, the product of a gene with variants that deCODE has linked to the risk of asthma. The company states that the variants are present in 40% of patients in the UK and Iceland. Protein kinases are enzymes involved in transferring signals within cells to perform or inhibit a function. A number of kinases are thought to play a role in the activation of chemical mediators and immune cells involved in airway inflammation. deCode developed a computerized DNA databank of asthma patients to identify a kinase gene variant linked to asthma. The compound is currently undergoing an 8-week Phase II trial as add-on therapy, which is expected to conclude this quarter. This is a very novel approach, and it will be quite interesting to see if there is anything to it.

Zenapax – PDL BioPharma (PDLI) and Roche (RHHBY)

Zenapax is a monoclonal antibody that binds to the interleukin-2 (IL-2) receptor on immune system cells known as T cells. This keeps IL-2 from stimulating the T cells to divide and participate in an immune response. It is interesting that one type of cell that IL-2 stimulates is the natural killer cell, which has recently been implicated in asthma. PDL has partnered it with Roche for transplant rejection (for which it is already approved) and respiratory diseases and with Biogen IDEC for other indications. In Phase II clinical testing of patients poorly controlled on corticosteroids, intravenous Zenapax has shown a reduction in exacerbations and a modest improvement in FEV1 of 5.9% compared to placebo. Reductions in blood eosinophil levels were correlated with the absence of exacerbations.

Single and multiple dosing trials of a more convenient, subcutaneous formulation were recently initiated with healthy volunteers. Dosing is complete and PDL expects to start a Phase IIb dose-ranging study for chronic persistent asthma in the second half of 2006. One concern we have are the hypersensitivity reactions that can develop to the drug seen in its use in transplantation. Also as a general immunosuppressant, the drug may render patients more susceptible to infection with long-term use.

FK506 - Astellas

FK506 is in inhaled formulation of the approved immunosuppressant Tacrolimus, which inhibits the production of interleukin IL-2 by helper T-cells, blocking T cell activation and proliferation. FK506 may also inhibit the production of other cytokines, including IL-3, IL-4, IL-5, and TNF-a. Of interest, it has been shown in some situations to inhibit natural killer cell activity. Other formulations of the drug, which is an immunosuppressant, are approved for use in organ transplants and for atopic dermatitis, which is related to asthma. Of concern is the possible risk for kidney toxicity and the development of lymphoma with the agent. The company hopes that the inhaled formulation will lead to fewer adverse events. Detailed clinical information on the inhaled formulation is not yet available.

REVENUE MODELS

We have added eleven new revenue models for asthma and incorporated two others with revisions (Xolair and Inhaled Bimosiamose). In addition, we have added four new revenue models for COPD (Formoterol-Ciclesonide, Formoterol-Mometasone, QAB149-ICS, and Xopenex Inhaled Solution) and revised our COPD projections for the ICS-LABA combination drugs and Xopenex HFA based on an updated evaluation. Finally, while our base case projects that Maintenance and Reliever Therapy (MRT) will eventually be approved for asthma (it is not applicable in COPD), we have also added a fixed dose scenario in asthma, showing projections for the LABA-ICS combination products should MRT not be approved.

Advair - GlaxoSmithKline (GSK)

While Advair is currently the lead inhaled respiratory product, we expect sales growth to slow, particularly with the labeling changes due to concerns over the SMART trial. However, we expect it to maintain a leadership position for a few years until Symbicort MRT is approved. Also, once Advair’s US patent expires in 2010, we think generics are likely to enter and take away much of its share. Finally, the once-a-day combination products will take over leadership of the fixed dose market.

Our 5- and 10-year projected US revenues for asthma are $681.2 million (after the entrance of generics) and $56.2 million, with corresponding worldwide revenues of $2.2 billion (as its share declines in Europe but without generics) and $150.9 million. Please see our spreadsheet for revenues in the fixed dose scenario.

In COPD, our 5- and 10-year projected US revenues are $305.6 million and $86.2 million, with corresponding worldwide revenues of $1.4 billion and 272.1 million (10-year revenues are higher than in asthma due to Maintenance and Reliever Therapy in the latter).

Symbicort - AstraZeneca (AZN)

We expect approval to be delayed until the end of 2007 for the fixed dose, and are projecting approval for Maintenance and Reliever Therapy (MRT) a couple years later. Our Europe model presumes an earlier MRT approval in 2007. With MRT, we expect Symbicort to take a leadership position, until patent expirations in the US (2014) and Europe (2012). Since Advair cannot be used for MRT, its patent expirations will not affect Symbicort in this model. While we expect MRT to take a majority share of patients, we also believe a sizeable number will remain on fixed dose, especially those who have more severe asthma. In the fixed dose scenario, we project Symbicort to be behind Advair, with shares dropping once Advair goes off patent. Also, in that scenario, the 24-hour drugs take majority share.

Our 5- and 10-year projected US revenues for asthma in our base case (MRT) are $679.2 billion and $412.6 million, with corresponding worldwide revenues of $2.8 billion and $1.1 billion.

In COPD, our 5- and 10-year projected US revenues are $172.6 million and $43.4 million, with corresponding worldwide revenues of $644.6 million and $136.1 million.

Formoterol-Ciclesonide – Altana (AAA) and Sanofi-Aventis (SNY)

Given Alvesco’s benefits over other ICSs, we think this combination will be preferred over Formoterol-Mometasone and Flutiform. We also think it will take some share from Advair and Symbicort, though these will have the advantage of being well-established in the market. At the same time, we are tempering our share projections for Formoterol-Ciclesonide, pending actual release of data from the safety trial evaluating the eye effects of higher doses of the drug. If it appears likely MRT will be approved, we believe Formoterol-Ciclesonide will be developed for the flexible dosing regimen and do well.

Our 5- and 10-year projected US revenues for asthma are $214.1 million and $415.6 million, with corresponding worldwide revenues of $458.9 million and $668.4 million.

In COPD, our 5- and 10-year projected US revenues are $116.9 million and $87.1 million, with corresponding worldwide revenues of $255.9 million and $215 million.

Formoterol-Mometasone – Novartis (NVS) and Schering Plough (SGP)

As a late entrant without major advantages, we expect Formoterol-Mometasone to lag behind Symbicort and Advair, as well as Formoterol-Ciclesonide. If it looks likely that MRT will be approved, we believe Formoterol-Mometasone will also be developed for MRT.

Our 5- and 10-year projected US revenues for asthma are $71.7 million and $92 million, with corresponding worldwide revenues of $153.4 million and $225.3 million.

In COPD, our 5- and 10-year projected US revenues are $29 million and $14.3 million, with corresponding worldwide revenues of $63.8 million and $38.6 million.

Flutiform – SkyePharma (SKYE)

Like Formoterol-Mometasone, we think Flutiform will suffer from its late entrance and lack of significant advantages. We are somewhat concerned about its development path, given the recent turmoil at the company.

Our 5- and 10-year projected US revenues for asthma are $71.7 million and $92 million, with corresponding worldwide revenues of $153.4 million and $225.3 million.

In COPD, our 5- and 10-year projected US revenues are $29 million and $14.3 million, with corresponding worldwide revenues of $63.8 million and $38.6 million.

QAB149-ICS – Novartis (NVS)

We expect QAB149-ICS to take a strong share of the fixed dose segment, and currently project it to come out slightly ahead of the GSK159797-ICS combo. As noted above, QAB149 is ahead of GSK159797 in development and likely has a faster onset of action, though it is unclear what type of delivery device will be used and which ICS QAB149 will be combined with. Asmanex is a good possibility. And while we are concerned about the latter’s 24-hour ability in more severe patients, it should do fine in patients with mild and more moderate disease.

Our 5- and 10-year projected US revenues for asthma in the MRT scenario are $275.1 million and $1.3 billion, with corresponding worldwide revenues of $593.3 million and $2.9 billion. In the fixed dose scenario, the 5- and 10-year projected US revenues are $403.9 million and $1.9 billion, with corresponding worldwide revenues of $870.8 million and $4.2 billion.

In COPD, our 5- and 10-year projected US revenues are $111.4 million and $655.6 million, with corresponding worldwide revenues of $230.2 million and $1.5 billion.

GSK159797-ICS – GlaxoSmithKline (GSK) and Theravance (THRX)

As noted above, we expect GSK159797 to take a second position in the fixed dose combination market behind a QAB149-ICS combination. Since current information on the drug and its competitors is preliminary, we have modeled it only slightly behind the QAB149-ICS combo.

Our 5- and 10-year projected US revenues for asthma in the MRT scenario are $159.5 million and $1.0 billion, with corresponding worldwide revenues of $337.8 million and $2.3 billion. In the fixed dose scenario, the 5- and 10-year projected US revenues are $239.3 million and $1.6 billion, with corresponding worldwide revenues of $507.7 million and $3.5 billion.

In COPD, our 5- and 10-year projected US revenues are $50.2 million and $536.4 million, with corresponding worldwide revenues of $103.9 million and $1.2 billion.

Xopenex HFA - Sepracor (SEPR) and 3M (MMM)

Though there is some evidence of a benefit for the nebulized formulation over racemic albuterol, Xopenex HFA has not shown a benefit in clinical testing. The HFA formulation is also likely to be used in different settings from the nebulized version (the nebulized formulation is mainly used in children and older patients, as well as in exacerbations, where side effects are particularly important).

In addition, through the near future, Xopenex HFA will compete with low cost generic albuterol inhalers. Once the CFC ban occurs at the end of 2008, only HFA inhalers will be allowed, of which there are currently four. However, we are assuming that more generic HFA inhalers will emerge, such that Xopenex will continue to have to compete against lower cost albuterol generics, though it is not as clear what the price differential will be. We are also assuming that a generic version of Xopenex will emerge after expiration of its 2013 patent (the nebulized formulation already has two patent challenges).

As a result of these factors, we do not think the HFA formulation will achieve as high a share as the nebulized, and project a peak share of 15% under the fixed dose ICS-LABA scenario that is currently approved. Under Maintenance and Reliever Therapy, less short-term rescue medication would be needed, so in that scenario (our base case) we have modeled slightly reduced Xopenex sales in asthma. Also, before the CFC ban, we are projecting a somewhat smaller peak share for Xopenex.

We have updated the COPD model to reflect actual pricing of Xopenex since its recent introduction. We have also reduced the revenue from COPD based on changed assumptions of the split of usage of the short-acting beta agonists between asthma and COPD.

Our 5- and 10-year projected US revenues for asthma are $213.8 million and $112.8 million. We are not currently projecting overseas sales.

In COPD, our 5- and 10-year projected US revenues are $64.9 million and $32.5 million.

Xopenex Inhalation Solution – Sepracor (SEPR)

While the nebulized formulation of Xopenex has seen good growth, particularly in the hospital and institutional setting, we believe it is now beginning to level off. Though Xopenex has shown some advantages, trial results are mixed and there are questions about whether it is worth the additional cost. In addition, patents on the compound expire between 2010 and 2013 (there is a formulation patent until 2021). Our model presumes generic competitors entering towards the end of 2013.

Our 5- and 10-year projected US revenues for asthma are $522.2 million and $166.4 million.

In COPD, our 5- and 10-year projected US revenues are $174.8 million and $51.2 million.

Xolair – Genentech (DNA), Novartis (NVS), Tanox (TNOX), and PDL BioPharma (PDLI)

We expect Xolair to continue its good growth, since it is currently without competitors in patients refractory to steroids. We think due to its mechanism of action involving IgE, which is important in atopic patients, it will retain its share once the anti-TNF antibodies enter.

Our 5- and 10-year projected US revenues for asthma are $1 billion and $1.4 billion, with corresponding worldwide revenues of $2 billion and $2.8 billion.

Enbrel – Amgen (AMGN) and Wyeth (WYE)

Enbrel is most likely to be used in the small segment of patients refractory to corticosteroids. Given the limited information on its efficacy, we are currently cautious about projections. We think it will lag behind the more conveniently dosed Golimumab, though still have a relatively significant share of this segment of the market. There is still the risk of other anti-TNF drugs entering this market.

Our 5- and 10-year projected US revenues for asthma are $180.1 million and $496.2 million, with corresponding worldwide revenues of $281.8 million and $1 billion.

Golimumab – J & J (JNJ), Schering Plough (SPG), and Medarex (MEDX)

We project Golimumab to gain share above Enbrel, due to its convenient dosing. However, it will be important to see clinical data ensuring that it is as effective.

Our 5- and 10-year projected US revenues for asthma are $311.7 million and $856 million, with corresponding worldwide revenues of $489.6 million and $1.7 billion.

Inhaled Bimosiamose – Revotar and Encysive (ENCY)

We are projecting that this agent could be used in a broader population than the anti-TNF drugs, but we think it will also be priced closer to the regular asthma market. Given that the clinical information is very preliminary, we are currently relatively conservative on our market share estimates.

Our 5- and 10-year projected US revenues for asthma are $174.9 million and $608.6 million, with corresponding worldwide revenues of $280 million and $1.4 billion.