Phase III data of topical NSAID show no hepatotoxicity

June 15, 2005

Washington, DC—Safety monitoring in the phase III clinical studies of topical bromfenac sodium 0.09% (Xibrom, ISTA Pharmaceuticals) shows no detectable systemic absorption or evidence for treatment-related liver toxicity in patients treated for up to 2 weeks with this twice-daily ophthalmic nonsteroidal anti-inflammatory drug (NSAID), reported Eric D. Donnenfeld, MD, at the annual meeting of the American Society of Cataract and Refractive Surgery.

"Based on the results of these studies and clinical experience with this product in Japan, where it has been available since 2000, topical bromfenac appears to be well-tolerated, safe, and effective. However, any time a new medication appears on the market, and especially a new NSAID, it has to be closely scrutinized," said Dr. Donnenfeld, assistant clinical professor of ophthalmology, New York University Medical Center, New York.

"Additional studies are needed and will be performed not just to look at systemic side effects, which are unlikely, but to evaluate any potential surface toxicity for this compound and see how it compares with other drugs in this class," he continued.

Liver function testing The phase III studies of bromfenac included extensive liver function testing because oral bromfenac (Duract) was withdrawn from the market by its manufacturer about a year after it was approved due to postmarketing reports of rare severe hepatitis and liver failure that even required transplantation in some patients. The panel of liver function tests (LFTs) in the ophthalmic bromfenac studies included measurement of AST, ALT, GGT, total bilirubin, direct bilirubin, and alkaline phosphatase.

Two identical phase III studies of ophthalmic bromfenac were conducted at 39 investigational centers to evaluate the drug's efficacy in treating inflammation postcataract surgery. The trials had a randomized, double-masked, placebo-controlled design and included 527 subjects in the intent-to-treat population, of whom 356 patients received bromfenac and 171 were treated with placebo. The assigned study medication was administered twice daily for 14 days.

Eligible patients had to have a summed inflammation score 3 at a postoperative visit conducted 16 to 32 hours after surgery. In addition, inclusion criteria required patients to have LFTs less than grade 1 using the Common Toxicity Criteria (CTC) of the World Health Organization. Masked results from study subjects were evaluated by Mitchell Shiffman, MD, a board-certified gastroenterologist specializing in liver disease.

Patients in the bromfenac and control groups had similar demographic characteristics. Overall, the study group had a mean age of about 70 years, consisted of about half males and half females, and was predominantly Caucasian (82%).

LFTs were repeated at the last visit when treatment ended at day 29, or earlier for subjects who discontinued treatment prematurely. Within the bromfenac group, 81.5% of patients received all 28 doses of study medication.

The key hepatic safety endpoint for the study was a post-treatment LFT of CTC grade 1 or higher. Although some individual LFTs were reported that were outside the upper limits of reference ranges, the percentage and distribution of subjects with LFTs of grade 1 were equal in the placebo and bromfenac groups.

"In this elderly population, it is not surprising to see sporadic elevated LFTs. However, LFT values remained within the normal range limits for more than 90% of subjects, and the masked review did not reveal any clinically significant increases in any LFT parameter," Dr. Donnenfeld said.

Safety assessments also included analyses of ocular adverse events, both from patient diaries and investigator-determined, as well as systemic adverse events. Rates of ocular adverse events were generally more common in the placebo group than among bromfenac-treated patients, and there was no evidence of systemic side effects.

Analysis of systemic drug absorption using an assay with a sensitivity limit of 50 µg/ml showed no detectable bromfenac in the plasma.