Xigris Serious Bleeds Can Be Minimized If Invasive Procedures Avoided - Lilly

Serious bleeding events as a result of Xigris administration can be reduced by avoiding invasive procedures during the drug infusion period, Lilly indicates in advisory committee briefing documents for the sepsis agent.

"In future studies or clinical practice, avoidance of invasive procedures during the administration of drotrecogin alfa (activated) may reduce the potential risk of bleeding," Lilly maintained in documents prepared for the postponed committee meeting.

Patients given Xigris were more likely to experience at least one serious bleeding event compared to the placebo group. "The difference in the incidence of severe bleeding events between the two treatment groups occurred primarily during the study drug infusion period," Lilly said.

"The increased proportion of patients with at least one serious bleeding event in the drotrecogin alfa (activated) group was primarily related to traumatic injury or instrumentation of a major blood vessel or highly vascular organ," the documents say.

In a Phase III study, 20 Xigris patients reported a serious bleed during the infusion period (2.4%), compared to eight placebo patients (1%). Half of the events were procedure-related, according to Lilly. The most common bleeding events were gastrointestinal (5), intrathoracic (4) and retroperitoneal (3).

Xigris patients were also more likely to experience serious bleeds over the entire 28-day study period. Lilly reported 30 (3.5%) recombinant human activated protein C patients with at least one event, compared to 17 in the placebo group (2%).

Half of the Xigris events were procedure-related, Lilly said. The most common bleeding events in Xigris patients over the 28-day period were gastrointestinal (9), intrathoracic (6) and retroperitoneal (4).

To help reduce the incidence of gastrointestinal bleeds, Lilly recommends that "visualization of the suspected site of bleeding should be considered...even if drotrecogin alfa (activated) administration is permanently discontinued."

FDA's Anti-Infective Drug Advisory Committee meeting, scheduled for Sept. 13, was postponed due to the terrorist attacks on Washington, D.C. and New York City. A new date has not been determined.

Six patients died during the trial, five of which Lilly attributed to Xigris administration; one patient received placebo. Of the Xigris deaths, four were caused by bleeding, and the fifth by a neurologic adverse event.

Lilly Exec VP-Science and Technology August Watanabe, MD, remarked on what he characterized to be a "slight increase" in the rate of serious bleeding events during Banc of America's Annual Investment Conference in San Francisco Sept. 10.

"We expect FDA to focus on the risk of bleeding and whether there is a population of patients at a higher risk of bleeding," he said. "The risk of intracranial hemorrhage appears to be slightly higher in Xigris-treated patients with very low platelet counts and severe coagulopathy."

"However, these patients are the 'sickest of the sick' and are already at a markedly increased risk for severe bleeding complications. Because these patients are also at a very high risk of dying from their severe sepsis, physicians may want the option of using Xigris."

"These issues will be addressed to the satisfaction of the committee. As a result, we expect a favorable recommendation," he maintained, adding that Lilly "plans to launch Xigris later this year."

In its briefing documents, FDA expressed concern about the number of serious bleeding events in patients scoring in the first quartile of the Acute Physiology and Chronic Health Evaluation II sepsis scale. The agency also discussed the higher mortality rate in first-quartile Xigris patients (¹ (Also see "FDA Xigris Efficacy Analysis Suggests APACHE II Test Can Guide Therapy" - Pink Sheet, 17 Sep, 2001.)).

"There was a much higher bleeding SAE rate in the rhAPC-treated group in the first quartile compared to the placebo group," FDA said. Nine cases of serious bleeds were reported in first-quartile Xigris patients (4%), versus none in placebo. In the fourth quartile, both groups each had three events (1%).

Non-serious bleeding events were more prevalent across the APACHE II quartiles, FDA said. In the first, Lilly reported 38 cases in the rhAPC-treated group (17%), versus 17 in placebo (8%). "Other than bleeding events, there were no other patterns of adverse events noted in the rhAPC group compared to placebo," FDA concluded.