AVANIR Submits NDA for Neurodex(TM)

AVANIR Pharmaceuticals (AMEX:AVN) announced that the
company has submitted the final modules of a "rolling" submission of
the new drug application (NDA) for Neurodex(TM), an investigational
new drug, to the U.S. Food and Drug Administration (FDA) as an
electronic common technical document (eCTD). AVANIR is seeking
approval to market Neurodex as a treatment for pseudobulbar affect
(PBA), also known as pathological laughing and crying, emotional
lability, and emotional incontinence.
   Pseudobulbar affect is a neurological condition characterized by
the disinhibition or loss of control of the motor expression of
emotion. Symptoms of PBA include uncontrollable crying or laughing
that is out of context with the social setting. It can be severe,
unremitting, and persistent. PBA occurs secondary to neurological
disease or injuries including amyotrophic lateral sclerosis (ALS) or
Lou Gehrig's disease, multiple sclerosis (MS), stroke, traumatic brain
injury, and dementias such as Alzheimer's disease. If approved,
Neurodex would be the first drug indicated to treat PBA.
   "Completing this NDA submission marks a key milestone in AVANIR's
commitment to develop treatments for chronic conditions such as PBA,"
said James E. Berg, Vice President of Clinical and Regulatory Affairs
at AVANIR. "The Neurodex NDA submission is supported by a data package
that, as previously reported, shows the consistency with which
Neurodex helps patients address PBA symptoms."
   The NDA for Neurodex contains data from two controlled,
multicenter Phase III clinical trials: one conducted in ALS patients
and the other in MS patients. Data was also submitted from its
open-label clinical study evaluating the safety of long-term exposure
to Neurodex(TM) in patients with PBA associated with a variety of
neurological disorders. Following the NDA submission, AVANIR will
submit additional safety data from the ongoing open-label clinical
study, including a 120-day safety update required by FDA regulations.
   "Neurological disorders and brain injuries can affect patients'
lives in devastating ways, but one consequence, PBA, is frequently
overlooked or misdiagnosed," said Hillel Panitch, MD, Professor of
Neurology at the University of Vermont College of Medicine and
Director of the Multiple Sclerosis Center at Fletcher Allen Health
Care in Burlington, Vermont. "Even though its hallmark characteristics
-- uncontrollable laughing and crying -- were described by Charles
Darwin more than 130 years ago, the mechanisms underlying PBA are
still not clear. However, the debilitating effects of the condition
have been well documented. PBA can be seriously disabling in social or
occupational settings, adversely affecting the quality of life for
patients."
   "With no drug currently approved for the treatment of PBA, this
NDA submission represents an important step forward in the potential
care and treatment of patients suffering from pseudobulbar affect as a
result of devastating underlying neurological disease or injury,"
concluded Dr. Panitch.
   AVANIR Pharmaceuticals is a pharmaceutical company focused on
developing and commercializing novel therapeutic products for the
treatment of chronic diseases. AVANIR's product candidates address
therapeutic markets that include central nervous system and
cardiovascular disorders, inflammation, and infectious disease.
AVANIR's preclinical research and development program for inflammatory
disease is partnered with Novartis. The Company's first commercialized
product, Abreva(R), is marketed in North America by GlaxoSmithKline
Consumer Healthcare and is the leading over-the-counter product for
the treatment of cold sores. Further information about AVANIR can be
found at www.avanir.com.

   Editor's Note

   Under normal conditions, the frontal, temporal, and motor cortices
of the brain send carefully modulated signals to the brainstem and
cerebellum to coordinate appropriate, voluntary displays of emotion.
It is hypothesized that neurological diseases and injuries, such as
ALS, MS, Alzheimer's disease, stroke, and traumatic brain injury,
impact the excitatory action of glutamate.(2) In these disease and
injury states, excessive glutamate signaling occurs, leading to
neurological damage.(3, 4) When this type of damage occurs along the
corticobulbar pathways of the brain, the result may lead to
involuntary displays of affect such as crying and/or laughing.(5-9)
   Originally discovered by Richard A. Smith, MD, of the Center for
Neurologic Study, Neurodex is believed to be an anti-excitatory
therapy that is comprised of dextromethorphan and an enzyme inhibitor
that slows the otherwise rapid metabolism of dextromethorphan.(10)
Dextromethorphan has activity both as a sigma-1 receptor agonist and
as an NMDA receptor antagonist, and may work to control PBA by
reducing excessive glutamate excitatory neurotransmission.(3, 11) Both
of these receptor types are located in regions of the brain that are
believed to be involved in controlling emotional expression. Neurodex
is thought to act by binding to sigma-1 receptors inhibiting glutamate
release, and by binding to post-synaptic NMDA receptors, inhibiting
glutamate mediated responses.
   Because epidemiological data are not available, the precise number
of patients with PBA in the United States is currently not known.
Based on estimates of the number of patients with each primary
disorder, the available clinical literature and primary market
research, it is surmised that approximately one million individuals
may suffer from PBA.
   The most recent double blind Phase III clinical study in MS
patients with PBA was completed in June 2004, in which 150 patients at
22 clinical sites were randomized to receive either placebo or
Neurodex on a 12-hour dosing schedule for 85 days. The Center for
Neurological Study Lability Scale (CNS-LS), a validated instrument
that assesses frequency and severity of PBA episodes,(1) was utilized
as the primary efficacy endpoint. A minimum CNS-LS score was required
for inclusion in the study. For the primary endpoint, patients
receiving Neurodex benefited from a significantly greater reduction in
CNS-LS score than those receiving placebo (p less than 0.0001). The
four secondary endpoints evaluated in the trial also significantly
favored Neurodex, namely reduction in number of PBA episodes
(p=0.0002), improvement in assessed quality of life (p less than
0.0001), improvement in assessed quality of relationships (p=0.0001),
and reduction in pain intensity (p=0.0271).
   Overall, Neurodex was well-tolerated in the MS patient population.
The majority of reported side effects (adverse events) were mild or
moderate. Of the adverse events reported in 5% or more of the
patients, a statistically significant difference between Neurodex and
placebo was observed only for dizziness. In the Neurodex group, 14% of
patients withdrew from therapy due to adverse events, compared to 11%
withdrawing in the placebo group.
   In 2002, AVANIR completed a Phase III clinical trial of Neurodex
in the treatment of PBA in patients with ALS. One hundred forty
patients with ALS who also suffered from PBA were enrolled at 17
academic study sites throughout the U.S. in the double blind,
controlled Phase III study. The patients received capsules containing
either Neurodex, dextromethorphan (DM) alone, or quinidine (Q) alone,
twice daily for four weeks. The Neurodex arm showed statistically and
clinically greater improvements than both the DM-only arm (p=0.0013)
and the Q-only arm (p=0.0002) in the primary efficacy endpoint: change
in the CNS-LS score. In the intent-to-treat population, the adjusted
mean improvement in CNS-LS for patients treated with Neurodex was
approximately twice the improvement in patients treated with either DM
only or Q only. Neurodex patients also exhibited lower overall episode
rates, (p less than 0.0001 for both arms) improved quality of life
scores, (p=0.0009 for the DM-only arm and p=0.0041 for the Q-only
arm), and improved quality of relationship scores (p less than 0.0001
for the DM-only arm and p=0.0007 for the Q-only arm).
   Neurodex's side effects reported in the Phase III clinical trial
in ALS patients were anticipated, based on the pharmacology of DM and
the patient population. Early withdrawal due to adverse events
occurred in 24% of Neurodex patients, 6% of DM-only patients, and 8%
of Q-only patients. Adverse events reported for Neurodex included
nausea, dizziness, somnolence, and diarrhea. All of these adverse
events resolved when therapy was discontinued.
   An open label study is being conducted to assess the safety of
chronic exposure to Neurodex in patients with PBA associated with
various neurological disorders including ALS, MS, stroke, Alzheimer's
disease, Parkinson's disease, and traumatic brain injury. Results from
this study reported in the NDA show a safety profile similar to that
observed in the controlled efficacy studies of Neurodex. As in the
ordinary course, additional results from this study will be reported
to FDA during the review period.

   References:

   1. Moore SR, Gresham LS, Bromberg MB, Kasarkis E. Smith RA. A self
report measure of affective lability. Journal of Neurology,
Neurosurgery, and Psychiatry. 1997; 63:89-93.
   2. Greenamyre JT. The role of glutamate in neurotransmission and
in neurologic disease. Arch Neurol. 1986; 43:1058-1063.
   3. Mattson MP. Excitotoxic and excitoprotective mechanisms:
abundant targets for the prevention and treatment of neurodegenerative
disorders. Neuromolecular Med. 2003; 3:65-94.
   4. Bittigau P, Ikonomidou C. Glutamate in neurologic diseases. J
Child Neurol. 1997; 12:471-485.
   5. Robinson RG. Neuropsychiatric consequences of stroke. Annu Rev
Med. 1997; 48:217-229.
   6. Poeck K. Pathophysiology of emotional disorders associated with
brain damage. In: Vinken PJ, Bruyn GW, eds. Handbook of Clinical
Neurology. Amsterdam: North-Holland Publishing Co; 1969:343-367.
   7. Feinstein A, Feinstein K, Gray T, et al. Prevalence and
neurobehavioral correlates of pathological laughing and crying in
multiple sclerosis. Arch Neurol. 1997; 54:1116-1121.
   8. Starkstein SE, Migliorelli R, Teson A, et al. Prevalence and
clinical correlates of pathological affective display in Alzheimer's
disease. J Neurol Neurosurg Psychiatry. 1995; 59:55-60.
   9. Arciniegas DB, Topkoff J. The neuropsychiatry of pathologic
affect: an approach to evaluation and treatment. Semin Clin
Neuropsychiatry. 2000; 5:290-306.
   10. Pope LE, Khalil MH, Berg JE, Stiles M, Yakatan GJ, Sellers EM.
Pharmacokinetics of dextromethorphan after single or multiple dosing
in combination with quinidine in extensive and poor metabolizers. J
Clin Pharmacol. 2004; 10:1132-1142.
   11. Maurice T, Lockhart BP. Neuroprotective and anti-amnesic
potentials of sigma-receptor ligands. Prog Neuropsychopharmacol Biol
Psychiatry. 1997; 21:69-102.

   Statements in this press release that are not historical facts,
including statements that are preceded by, followed by, or that
include such words as "estimate," "anticipate," "believe," "plan," or
"expect," or similar statements, are forward-looking statements that
are subject to certain risks and uncertainties that could cause actual
results to differ materially from the future results expressed or
implied by such statements. There can be no assurance that AVANIR's
new drug application for Neurodex(TM) will be accepted for filing by
the FDA, that Neurodex will receive regulatory approval, or that even
if such regulatory approval is received, AVANIR will be able to market
Neurodex successfully. Final review decisions made by the FDA and
other regulatory agencies concerning clinical trial results are often
unpredictable and outside the influence and/or control of the company.
Risks and uncertainties also include the risks set forth in AVANIR's
most recent Annual Report on Form 10-K and subsequent Quarterly
Reports on Form 10-Q and from time-to-time in other publicly available
information regarding the Company. Copies of this information are
available from AVANIR upon request. AVANIR disclaims any intent or
obligation to update these forward-looking statements.