FDA Asks Panel To Consider If Salix's Xifaxan Caused Patient Deaths In Study

Salix's Xifaxan for hepatic encephalopathy faces questions over possible excess deaths and the subjectivity of the primary endpoint in the single pivotal trial.

An FDA advisory panel will have to grapple with concerns on both the safety and efficacy of Salix's Xifaxan (rifaximin) for maintenance of remission of hepatic encephalopathy; the agency's review reveals concerns about possible excess deaths and the subjectivity of the primary endpoint in the single pivotal trial.

FDA's medical officer is concerned by the possibility that several deaths of patients taking Xifaxan in the trial may have been related to the study drug, although the clinical investigator thought they may have been due to the underlying disease. "The contribution of potentially higher rifaximin systemic exposure in this population cannot be excluded," the reviewer writes in background materials for the Feb. 23 meeting of the Gastrointestinal Drugs Advisory Committee.

The death rate was more than twice as high for Xifaxan patients as for the placebo group (17.6 percent versus 7.1 percent) in one particular subgroup of patients known as Child's C, based on their classification in a system that categorizes liver dysfunction. "While the numbers are too small to permit any definite conclusions, this reviewer does not believe that the applicant has established that rifaximin does not have a negative impact on mortality in this subgroup of patients with significant hepatic impairment," FDA briefing documents state.

The FDA reviewer also is concerned by the fact that at least two patients had low baseline Model for End-Stage Liver Disease scores and subsequently showed sudden deterioration while on the study drug. There were more infections that constituted serious adverse events in the Xifaxan group, due mainly to an increased incidence of pneumonia and Clostridium difficile colitis as well as gastrointestinal disorders and general disorders (edema and pyrexia).

From Salix's standpoint, as expressed in its materials for the committee, most of the 47 patients who received Xifaxan and died during or after the Phase III trial did so because of their advanced liver disease and related health effects. "Survival probability for subjects in the primary studies was consistent with the natural history of advanced liver disease," the company argued. "Liver disease severity at baseline was the most important factor in determining the likelihood of survival for subjects in the study." The company also cited "experience in multiple clinical studies in HE and other indications with approximately 5000 subjects, as well as extensive postmarketing exposure (20+ years)" to confirm the drug's safety.

FDA is asking the advisory committee to consider whether the safety of Xifaxan at the proposed dose and duration has been adequately assessed in cirrhotic patients with a history of hepatic encephalopathy, as well as for the Child's Class C patients specifically. The committee will also assess the risk for development of resistant bacteria, C. difficile colitis and cardiac safety.

As to efficacy, Xifaxan did meet the primary endpoint, with statistical significance, of time to first breakthrough overt hepatic encephalopathy episode. Xifaxan-treated patients had a 58 percent lower risk on that measure, and a 50 percent lower risk of being hospitalized for an overt HE episode, Salix reports. Of the 140 subjects in the Xifaxan group, 31 (22 percent) experienced breakthrough overt HE episodes, compared with 73 of 159 subjects (46 percent) in the placebo group up to 170 days after randomization.

However, FDA notes that that the primary efficacy measure "was assessed utilizing the Conn score, which requires a subjective assessment and depends on clinician judgment ... [and] did not have to be assigned based on a direct assessment conducted during a clinic visit, but could also be assigned based on a telephone visit, caregiver assessment, or upon the basis of a non-investigator clinical report from a hospitalization."

Moreover, when FDA's Division of Gastroenterology Products called on CDER's Division of Neurology Products to assess the neurological aspects of the study, the latter concluded that Salix "did not provide enough evidence to establish that rifaximin is efficacious, based on the primary endpoint and the limitations of its assessment in this study." Neurology is relevant because hepatic encephalopathy is a deterioration of brain function that occurs because toxic substances normally removed by the liver build up in the blood and reach the brain.

FDA is seeking the panel's input on the efficacy measure, with a vote on whether an increase in time to breakthrough hepatic encephalopathy event as defined in the trial protocol constitutes a reduction in the risk of developing episodes of overt hepatic encephalopathy. The agency is also interested in whether this represents an actual clinical benefit.

As this Xifaxan application is the first indication sought for hepatic encephalopathy, there is no precedent for study design. According to the Merck manual, hepatic encephalopathy is currently treated with the synthetic sugar lactulose, or, for patients who cannot tolerate that, an oral antibiotic. But other drugs are in development, and FDA is also seizing the opportunity of the advisory committee to solicit input on how future trials of drugs for hepatic encephalopathy should be conducted.

Xifaxan, which was approved in the U.S. in 2004 to treat travelers' diarrhea caused by noninvasive strains of E. coli, is also being studied for non-constipation irritable bowel syndrome and Clostridium difficile infections (Pharmaceutical Approvals Monthly, December 2009). Salix expects to submit an NDA for the IBS indication in the first half of 2010.

Xifaxan's original priority review user fee date of Dec. 24, 2009 has been extended to March 24 (Pharmaceutical Approvals Monthly, January 2010).

-Martin Berman-Gorvine ( [email protected] )