Prescriptions for the drug increased from five million in 2005 to more than 14 million in 2009, a threefold increase, according to another FDA reviewer, Rajdeep Gill. The majority of prescriptions are for depression, but at least 14 percent are already for off-label use in pain treatment, he said.
"Approving the use of Cymbalta ... for chronic pain may increase the patient exposure and physician prescribing to an area that is already not uncommon," wrote Gill.
If approved, Cymbalta would be the first non-NSAID, nonopioid analgesic broadly indicated for the treatment of chronic pain.
Lilly previously filed an application for a chronic pain indication for Cymbalta in May 2008, but withdrew it the following November after the FDA disapproved of efficacy results. Results of an ongoing osteoarthritis trial were not available at that time.
Because it is a "nontraditional" analgesic where the mechanism of action is not well-defined, the agency may need greater evidence to support a chronic pain indication than it would for opioids and NSAIDS, wrote Rappaport.
The advisory committee will examine five new clinical trials from Eli Lilly: three testing various doses of Cymbalta versus placebo in 1,041 chronic low-back pain patients, and two comparing Cymbalta with placebo in 487 osteoarthritis patients.
All five trials were multicenter, randomized, double-blind trials with a treatment duration of at least 12 weeks. The primary efficacy endpoint for all the trials was the change from baseline to week 13 in pain intensity, which was measured by the 11-point Brief Pain Inventory and patient diaries.
In four of the five trials, Cymbalta -- given at between 60 mg and 120 mg -- demonstrated greater pain reduction compared with placebo.
However, FDA reviewers criticized the trial design because it assumed that patients dropped out of the trials at random, which was likely not the case.
"In analgesic trials, early discontinuations should be considered treatment failures; therefore, an improvement in a subject's pain score prior to dropping out due to an adverse event should not be credited in the analysis," the FDA reviewers wrote.
Patients in the Cymbalta trials were more likely to discontinue treatment because of adverse events, including nausea, sleep disturbances, dizziness, dry mouth, somnolence, constipation, and fatigue. Most of the events were dose-dependent.
Cymbalta use also appears to be associated with an increased risk of Stevens-Johnson syndrome, a skin disease that usually results from a drug reaction, as well as another form of the disease called toxic epidermal necrolysis. The FDA reviewers recommended that, if it is given the chronic pain indications, it should carry warning labels about the increased skin disease risks.
The FDA also has concerns about its risk of serious liver toxicity, already included in the drug's labeling.
No deaths were reported during the trials.
The safety findings aren't new or unexpected and are very similar to Cymbalta's current safety label, the FDA reviewers said.
The drug also carries a boxed warning for suicidality in children, adolescents, and young adults taking antidepressants for major depressive disorder and other psychiatric disorders.