Eli Lilly and Company
David Shaffer (US), 317-651-3710
Jennifer Yoder (Global), 317-652-0912

Cymbalta(TM) (duloxetine HCl)(TM) was safe and well tolerated in a 52-week
open-label study of 1,279 patients with major depressive disorder, according to
data published in the current issue of the Journal of Clinical Psychiatry.

There were no significant tolerability issues attributable to chronic vs. acute
use of the investigational agent Cymbalta; patients who tolerated Cymbalta
during the early period of the trial were likely to tolerate long-term dosing.
Of the patients in the study, 520 remained on Cymbalta for at least 360 days,
yielding approximately 808 patient-years of total exposure.

In addition, patients in the study who responded to treatment with Cymbalta
had a high probability of achieving remission. Remission rates at 52 weeks in
this study were close to response rates (81.8 percent and 89.1 percent,
respectively), implying that patients who responded had a high probability of
achieving complete resolution of their depressive symptoms.

"Long-term tolerability is important for antidepressant effectiveness in the
complete treatment of depression. The higher-than-expected completion rate in
this open-label study implies that Cymbalta was well-tolerated in these patients
and was effective in helping them relieve their depressive illness," said Joel
Raskin, MD, Lilly senior clinical research physician and lead author of the
study.

Study Highlights

    --  Cymbalta was safely administered and well-tolerated in long-term chronic
        dosing, despite higher dosages than those used in most other Cymbalta
        studies.

    --  Most treatment-emergent adverse events were either mild or moderate in
        severity and occurred early in the study.

    --  Efficacy was demonstrated on all assessed measures, both clinician- and
        patient-rated. Furthermore, discontinuation due to adverse events over
        the entire 52-week study was 17 percent - a favorable rate given the
        long duration of the study.

Methods

Data were gathered from a 52-week, open-label, multinational study of 1,279
adult outpatients who met the criteria for major depressive disorder. Patients
were administered duloxetine 80 to 120 mg/day as two equal doses of 40 mg or 60
mg.

Efficacy was assessed using the Clinical Global Impression-Severity
(CGI-Severity) scale, the Hamilton Depression Rating Scale (HAMD(17)), the Beck
Depression Inventory-II and the Patient Global Impression-Improvement
(PGI-Improvement) scale. Patient-rated quality was evaluated with the Sheehan
Disability Scale. Response was defined as a 50 percent decrease from baseline on
the HAMD(17) total score and remission was defined as a HAMD(17) total score of
less than or equal to 7.

Mean changes from baseline to last observation in laboratory analyses, vital
signs and electrocardiograms intervals were assessed using an analysis of
variance with models that included the investigator. Longitudinal mean changes
and categorical changes were assessed using likelihood-based, mixed-models
repeated measures and last observation carried forward.

Tolerability/Safety

Discontinuation due to adverse events over the entire 52-week study was 17
percent -- a favorable rate given the long duration of the study. The most
common reasons for discontinuation included adverse event (17 percent), personal
conflict/other reasons (10.2 percent), and lost to follow-up (9.3 percent).

The most common treatment-emergent adverse events in the study included nausea
(34 percent), somnolence (29.8 percent), insomnia (31.3 percent), headache (30.4
percent), dry mouth (23.5 percent), constipation (21.3 percent) and dizziness
(23.3 percent). Most side effects occurred early in the study and generally
dissipated over time. Only one side effect (headache) occurred in more than 10
percent of patients in weeks 9-52 of the study.

The rate of serious adverse events per patient-year-exposure was low -- roughly
1 event per 13 years of exposure. A total of 64 patients experienced serious
adverse events, but investigators considered most unrelated to the study
medication.

Efficacy/Remission

Efficacy was demonstrated on all assessed measures, both clinician-and
patient-rated. The high rates of improvement at week one and two, while
difficult to define and assess, were consistent with results from double-blind,
placebo-controlled studies.(1-3)

Accumulating evidence suggests complete resolution of disease symptoms, or
remission, rather than simple treatment response, should be the primary goal of
depression treatment. Current medical literature suggests that depressed
patients frequently experience lingering symptoms, such as persistent
unexplained pain, putting them at a higher risk for relapse or recurrence.(4)

Although interpreting results in an open-label study can be problematic, the
remission rates in this 52-week study were high, implying Cymbalta was effective
in relieving the symptoms of major depression in these patients.

"By treating a broad spectrum of depressive symptoms - emotional and physical -
Cymbalta, once approved by the FDA, may help more patients achieve remission,"
Raskin said. "These data show that these patients were able to easily tolerate
Cymbalta over the long-term, enabling them to achieve even higher rates of
response and remission the longer they remain on therapy."

About Cymbalta

In placebo-controlled clinical trials for major depressive disorder, Cymbalta
had a favorable safety profile. The most commonly observed adverse events
(greater than or equal to 5% and at least twice placebo) for Cymbalta vs.
placebo (n = 1,139 vs. 777) were: nausea (20% vs. 7%), dry mouth (15% vs. 6%),
constipation (11% vs. 4%), decreased appetite (8% vs. 2%), fatigue (8% vs. 4%),
somnolence (7% vs. 3%) and increased sweating (6% vs. 2%). In these studies,
anxiety was reported as an adverse event (3% vs 2%). The overall discontinuation
rate due to adverse events for Cymbalta vs. placebo was 10% vs. 4%. Nausea was
the only common adverse event reported as a reason for discontinuation and
considered to be drug related (1.4% vs. 0.1 %).

The US Food and Drug Administration issued an approvable letter for Cymbalta
(duloxetine for depression) in September 2003 based upon eight- and nine-week
trials. Duloxetine hydrochloride also is being studied by Lilly for treatment of
stress urinary incontinence, a condition mediated by serotonin and
norepinephrine.

About Lilly

Lilly, a leading innovation-driven corporation, is developing a growing
portfolio of best-in-class pharmaceutical products by applying the latest
research from its own worldwide laboratories and from collaborations with
eminent scientific organizations. Headquartered in Indianapolis, Ind., Lilly
provides answers - through medicines and information - for some of the world's
most urgent medical needs. Additional information about Lilly is available at
www.lilly.com.

This press release contains forward-looking statements about the potential of
the investigational compound Cymbalta for the treatment of major depressive
disorder and reflects Lilly's current beliefs. However, as with any
pharmaceutical product under development, there are substantial risks and
uncertainties in the process of development and regulatory review. There is no
guarantee that the product will receive regulatory approvals, or that the
regulatory approval will be for the indication(s) anticipated by the company.
There is also no guarantee that the product will prove to be commercially
successful. For further discussion of these and other risks and uncertainties,
see Lilly's filings with the United States Securities and Exchange Commission.
Lilly undertakes no duty to update forward-looking statements.

    1. Detke MJ, Lu Y, Goldstein DJ, et al. Duloxetine, 60 mg once daily, for
        major depressive disorder: a randomize double-blind placebo-controlled
        trial. J Clin Psychiatry 2002;63: 308-315

    2. Detke MJ, Goldstein DJ, et al. Duloxetine 60 mg once daily dosing versus
        placebo in the acute treatment of major depression. J Psychiatry Res
        2002;36: 383-390

    3. Goldstein DJ, Mallinckrodt C, Lu Y, et al. Duloxetine in the treatment of
        major depressive disorder: a double-blind clinical trial. J Clin
        Psychiatry 2002;63: 225-231

    4. Paykel ES, Ramana R, Cooper Z, et al. Residual symptoms after partial
        remission: an important outcome in depression. Psychol Med 1995;25:
        1171-1180