Parkinson disease (PD) is a chronic, progressive neurological disorder caused by the degeneration of dopamine-secreting neurons that control movement and balance. Nonmotor symptoms include dementia and depression. In most cases, the underlying cause is unknown.

There are approximately 1,000,000 patients with Parkinson disease in the US, with 50,000 to 60,000 more diagnosed each year. Worldwide, there are approximately 4 million individuals afflicted (2.7 million in the US, Japan, and the 5 major European markets). Since the incidence of PD increases with age (the average age of onset is 60), the number of patients is likely to climb as the population of older patients grows.

Ultimately, patients all require treatment with levodopa, a precursor to dopamine. However, in advanced disease, as the nerve cell degeneration becomes even more severe, treatment becomes difficult, and patients have periods where the effect of levodopa wears off or too much levodopa results in abnormal movements. There are a variety of current treatments primarily aimed at increasing dopamine stimulation. However all have significant drawbacks, and there still is a great need for medications that prevent nerve cell degeneration or better treat advanced disease.

In early disease, there are no new drugs with both impressive efficacy and side effect profiles, but we believe that Agilect will have modest penetration, especially given the possibility of neuroprotection.

In late disease, there are a number of treatments recently approved, in NDA status, and in development. These include Stalevo, approved last year; Zelapar and Agilect, both in NDA status; and KW-6002 (phase III). While this group will likely gain significant share as a whole, we believe they will split that share (with Stalevo and Zelapar at the top) because of similar efficacy and safety. A drug in early development, E2007 may have an advantage over all of these, but clinical detail is lacking. Two other drugs, SLV-308 (phase III) and NS2330 (phase II) will be very interesting to follow, as they may treat both motor and nonmotor symptoms of PD.

CURRENT TREATMENT STRATEGIES

Treatment issues differ for early and advanced disease. The most effective and commonly prescribed treatment for both is levodopa, a precursor to dopamine introduced in 1967. Levodopa is given instead of dopamine, because the latter cannot cross into the brain. In the US, levodopa is traditionally given in a combination pill with carbidopa, which blocks the enzyme DDC from converting it to dopamine before it reaches the brain.

While all patients eventually require levodopa, complications signifying advanced disease develop after about 5 to 6 years, most likely because the brain loses its capacity to store the drug. These include “off” times (poor motor function from lack of dopamine) and dyskinesias (abnormal movement, probably from too much dopamine). Thus a major need in early treatment is to find drugs that prevent disease progression, and in late disease to avoid motor complications.

Because these motor complications may be related to the short duration of action of levodopa, which may be given multiple times a day, a major strategy to combat them is through prolonging the action of levodopa or related compounds for more continuous stimulation of the dopamine receptors. (As a result, a number of products in development are reformulations of existing products.) Longer-acting levodopa formulations, currently available as generics, are advocated as one means to achieve this, but while useful in some situations, studies have not been able to demonstrate a substantial benefit.

Dopamine Agonists
The dopamine agonists, used since the 1970s, directly stimulate dopamine receptors in the brain and, after levodopa, are the next most effective and commonly prescribed treatment. There is controversy over whether they should be used before levodopa because while they appear to delay the motor complications temporarily, they also have more side effects and do not achieve the same symptomatic benefits as levodopa. Oral dopamine agonists currently approved in the US include: bromocriptine, pergolide, Mirapex (Boehringer Ingelheim and Pfizer - PFE), and Requip (GlaxoSmithKline - GSK). They are also commonly used as an adjunct to levodopa in advanced disease, partially reducing “off” time by up to 1.5 to 2 hours per day relative to placebo. They have not been convincingly shown to differ in effectiveness, though Mirapex has shown the highest reduction in off-time. An injectable dopamine agonist, Apokyn (Mylan - MYL), was approved in 2004 for the acute, intermittent treatment of “off” times. Though effective, because it is an injectable with a high incidence of side effects, it is likely to have limited use.

MAO-B Inhibitors
MAO-B is the enzyme that breaks down dopamine in the brain, so blocking this enzyme increases the amount of dopamine available and prolongs its duration of action. The only MAO-B inhibitor currently on the market is selegiline, which is a generically available. Its symptomatic effect in early PD is considered to be modest, and its benefit in late disease appears to wane after 7 to 8 months. One pivotal study seemed to suggest a neuroprotective benefit; however, the current consensus is this was more likely a result of a direct effect on symptoms. Two other promising MAO-B inhibitors, Agilect and Zelapar, currently have NDAs submitted (see below).

COMT Inhibitors
COMT is an enzyme that breaks down levodopa and dopamine. Inhibition of the enzyme results in increased levodopa and smoothed out blood levels. Comtan (Novartis -NVS), approved in 1999, is an inhibitor that is known to work outside of the brain (its effects in the brain are unknown in humans). A more convenient formulation, Stalevo (Novartis -NVS), combines Comtan with levodopa/carbidopa. It was approved in 2003 and, as expected, is taking significant share from Comtan, as well as growing the market share for this class of drugs. Comtan reduces “off” time by up to 1.2 hours per day relative to placebo with fewer side effects than the dopamine agonists. Stalevo is also being investigated in the STRIDE-PD trial to see if its use early in the disease process delays motor complications seen with levodopa therapy. Results are expected in 2007 and, if positive, could greatly expand Stalevo’s use.

Other Agents
Amantadine and the anti-cholinergics are generically available medications that have mild symptomatic benefits in early PD. Amantadine also reduces dyskinesias in advanced PD, though its effectiveness wears off after about 7 months. The side effect of confusion limits use of these agents in the elderly.

A large British trial comparing the various classes of PD drugs should have interim results available in November 2006. This trial may have a significant impact on treatment patterns, though it is unlikely to be powered to distinguish particular drugs within a class.

DRUGS WITH NDA SUBMITTED

Zelapar (Zydis selegiline) - Valeant (VRX), Elan (ELN), and Cardinal Health (CAH)

Zelapar is a once-daily orally disintegrating and absorbed form of selegiline. After a 2003 approvable letter, Valeant submitted additional safety information last January. While it expected an FDA decision this month, they are still in discussion over reportedly minor issues. A PDUFA date has been set for this year, but specifics have not been revealed. Hence we estimate the approval of Zelapar in the US will be later than that for Agilect, which is expected in August.

Zelapar gives similar levels of selegiline as the traditional formulation, but with fewer amphetamine metabolites, which are of theoretical concern. On the other hand, one of these metabolites appeared to be responsible for the neuroprotective effect of selegiline seen in animal studies. Zelapar has not been studied as monotherapy in early PD, though like selegiline it is likely to have a modest effect.

In advanced disease, Zelapar showed a 1.6 hour reduction in “off” time compared to placebo. This was relatively impressive, but we were disappointed at the lack of detail given in the published study, making it difficult to fully assess.

Agilect (Rasagiline) – Teva (TEVA), Eisai, and Lundbeck

Agilect is another once-daily MAO-B inhibitor. In early PD, Agilect has a modest effect as monotherapy, on the same scale as selegiline but smaller than the dopamine agonists. While patients do better on test scores, the same proportion as placebo (about 30%) require additional therapy over a year. Interestingly, in one study, patients treated for a year did better than those started on placebo for 6 months and then treated for the remaining 6 months. This could represent either a neuroprotective effect or some type of cumulative symptomatic benefit. Agilect may have effects on cognitive function and mood, and is also being developed for Alzheimer’s disease.

Studies in advanced stage PD have shown a reduction in off-time up to 1 hour (compared to placebo) depending on dosage. A study of the higher dose of Agilect and Comtan showed comparable efficacy and side effects. Importantly, Agilect’s benefit in advanced PD was shown to be similar whether or not patients were taking dopamine agonists.

Neupro (Rotigotine TDS) – Schwarz and Aderis

Neupro is a transdermal, continuous delivery system containing rotigotine, a dopamine agonist. An NDA has been submitted for early PD, with an estimated approval date of November 2005. All in all for early PD, Neupro appears similar to the dopamine agonists. Because it is a patch and has a high rate of skin reactions, we do not think it is a break-through treatment for this segment of the market.

The development of Neupro for advanced disease is ongoing. There is a lack of detailed information, but in October, the company announced that Neupro had met its endpoints in a US phase III study, where it was compared to placebo. Results from a large European phase III study are expected in early 2006. Continuous delivery is theoretically a good idea for advanced disease, so we are cautiously optimistic. The main issue is whether it is more effective than oral dopamine agonists. If so, Neupro is likely to take significant share from them. If not, it may still be used in patients who have significant difficulties swallowing. Like the oral dopamine agonists, it is likely that Neupro can be used in combination with the MAO-B and COMT inhibitors, though this should be documented in a trial.

DRUGS IN PHASE III DEVELOPMENT

KW-6002 (Istradefylline) - Kyowa Hakko Kogyo

KW-6002 blocks the adenosine A2a receptor. These receptors are thought to modulate the effect of dopamine on dopamine receptors. In patients with advanced PD, KW-6002 has shown reductions in “off” time of 1.2 to 1.9 hours compared to placebo, though results have been inconsistent. Of concern were relatively high rates of nausea, though this lessens after the first 4 weeks, and a possibly drug-related heart attack in a patient, which needs to be watched for in future studies. In addition, we are concerned about the increase in dyskinesias with higher doses of KW-6002. If KW-6002 does show more consistent efficacy in phase III trials, it is likely to be another significant competitor for the MAO-B and COMT inhibitors, though combination therapy will probably be used to some extent (limited by adverse effects of polypharmacy).

Of note, there are other adenosine 2a receptor antagonists being developed by Schering-Plough (unnamed, phase II) and Biogen-Idec/Vernalis (v2006, just beginning phase II).

Requip OCR (Ropinirole OCR) - SkyePharma (SKYE) and GlaxoSmithKline (GSK)

Requip OCR is a once-daily formulation of the dopamine agonist with the same name. In addition to convenience, there is theoretical advantage of less fluctuation in drug levels with a more consistent symptom response throughout the day. Trial data has not been released to assess this drug, but in theory this should be of benefit in advanced disease. If excessive adverse effects can be avoided, this drug is likely to be a strong competitor for Neupro and the dopamine agonists.

SLV-308 - Solvay

SLV-308 has agonist effects on various receptors for dopamine, norepinephrine, and serotonin. It is being developed for Parkinson disease, panic disorders, and depression. Since depressive symptoms are seen in about 20% of PD patients and there may be interactions between PD drugs and antidepressants, this could be quite a useful treatment. No clinical data is available, however.

DRUGS IN PHASE II DEVELOPMENT

NS2330 - NeuroSearch and Boehringer Ingelheim

NS2330 is being studied in both early and advanced PD. This compound affects the same general classes of neurotransmitter receptors as SLV-308, but by inhibiting re-uptake of dopamine, norepinephrine, and serotonin. It also increases the release of acetylcholine and has shown ability to improve cognitive function in Alzheimer’s disease. It will be interesting to see whether along with its effect on dopamine, it helps cognitive function and depression in PD. The drug appears to have a strong development program, with over 200 patients in each of two phase II trials. The company expects to receive results in the very near future. On the cautionary side, NS2330 has some similarities to stimulant drugs such as cocaine and amphetamines and so is being studied for its risk of abuse.

E2007 - Eisai

E2007 is an antagonist of the AMPA-receptor for glutamate. This is an interesting target because glutamate has been implicated as a causal factor in the nerve degeneration of PD, certain motor and non-motor symptoms, and dyskinesias related to levodopa treatment. E2007 has been found to reduce “off” time in advanced PD on the same order as Comtan and Agilect. While available details are sketchy, company literature suggests E2007 does not increase dyskinesias, which could give it a significant advantage over these other treatments. It will be interesting to see whether E2007 affects other symptoms as well, though the safety evaluation will be important, since AMPA-receptors are apparently one of the most common in the brain. After meeting with regulatory officials, Eisai hopes to conduct phase III trials, with a target NDA submission in 2006-2007 (E2007 is also being studied in epilepsy and multiple sclerosis). Another AMPA-receptor antagonist, Talampanel (Ivax, IVX) , is also in phase II development, but is being studied primarily for treatment of dyskinesias (as well as for epilepsy).

Spheramine - Titan Pharmaceuticals (TTP) and Schering (SHR)

Spheramine involves an intriguing technology: dopamine-producing human retinal pigment epithelial (RPE) cells adhered to spherical microscopic carriers, which are injected into the region of the brain lacking dopamine. Though details are limited, initial results in a small group of patients appears promising, with substantial improvements. However, there may be some indication in the 2-year data that the effect may be wearing off. Larger and longer safety studies are needed, and this treatment is likely to be reserved for more severe cases.

CNP1512 (melevodopa/carbidopa) - Cita Neuropharmaceuticals

CNP1512 is an effervescing tablet of a methyl-ester variant of levodopa/carbidopa (taken after dissolving in water). It is approved in Italy, and licensed to Cita for further development in the US and Europe. The formulation reportedly has a faster onset and longer duration of action (12-18 minutes), which is anticipated to translate into a 1 hr reduction in off-time (patients usually take multiple doses of levodopa). While the data the company has presented is interesting, it is curious why this formulation has taken so long to develop (there are studies of levodopa methyl-ester in the literature going back to at least 1991). Hence, we are skeptical about the commercial potential for this drug.

GPI 1485 (NIL-A) - Symphony Neuro Development and Guilford Pharmaceuticals (GLFD)

GPI 1485 is a neuroimmunophilin that is neuroprotective and neuroregenerative in animal models. While an initial study showed no improvement in symptoms (leading Amgen to terminate its rights), there was evidence of reduced deterioration seen on imaging, which is being studied further. Because of the lack of clinical results thus far and the rocky development, we are pessimistic about this drug’s prospects.

REVENUE MODELS

We have added revenue models for the following drugs in the Parkinson disease market: Stalevo, Comtan, Zelapar, Agilect, and KW-6002.

Stalevo/Comtan
Stalevo is much more convenient than its predecessor, Comtan, and is likely to take the bulk of Comtan’s share. Together, we project the two to reach peak share of 20% of advanced patients. While Stalevo will grow the franchise, this will be limited by heavy competition from the new MAO-B inhibitors entering the market. However, Stalevo’s share will not drop precipitously, as patients who have been successfully treated are unlikely to switch. We expect 2005 worldwide revenue for Stalevo to be $205 million. Peak US revenue for Stalevo is projected at $181 million in 2009 and peak worldwide revenue at $434 million. Stalevo could achieve significantly greater revenue if results of the STRIDE-PD study, expected in 2007, show a benefit for its use earlier in the disease process.

Zelapar
Because of somewhat better efficacy than Agilect, as well as ease of use in patients with difficulty swallowing, we expect Zelapar (orally dissolving selegiline) to be the next largest competitor in the advanced PD market. While Zelapar’s effect as monotherapy has not been documented in early PD, there is likely to be some use based on previous experience with selegiline. Due to uncertainty in the PDUFA date timing, we modeled approval for the fourth quarter of this year - shortly after Agilect’s expected approval. Peak US revenues are projected at $163 million in 2012 and peak worldwide revenue at $412 million.

Agilect
We expect Agilect to be a strong competitor, but in advanced PD we project its share to be below Zelapar due to its smaller efficacy. In early PD, with its limited benefit, we expect modest penetration, though there will be interest due to the possible neuroprotection and its good side effect profile. This use in early disease will make up for its lower share in advanced PD, and we expect peak US revenue of $201 million in 2011, and peak worldwide revenue of $500 million.

KW-6002
Despite results showing relatively good efficacy, we have concerns about how consistent these are, as well as safety issues. Also, KW-6002 will be a late entrant. We estimate peak US revenue of $74 million in 2013, with peak worldwide revenue of $189 million.