Aventis' existing labeled warnings for Arava regarding liver toxicity do not need to be strenghened, members of FDA's Arthritis Advisory Committee told the agency at its March 5 safety review of the arthritis drug.

"I have not seen anything presented here that was surprising, that was not already being monitored for," committee member James Williams, MD, University of Utah, said. "I would not think that any labeling change would be necessary."

FDA asked the committee to weigh the benefit/risk of Arava (leflunomide) considering reports of serious liver injury in patients taking the drug. The agency's investigation into Arava's adverse event profile was prompted by a Public Citizen Health Research Group petition in March 2002 requesting that Arava be withdrawn from the market (¹ [A#00640130023], p. 29).

The committee voted unanimously that the benefit-to-risk profile for leflunomide is acceptable.

Anti-Inflammatory, Analgesic & Ophthalmic Drug Products Division Director Lee Simon, MD, told the committee that FDA is likely to make some changes to the hepatotoxicity language in labeling. "We do believe that the labeling needs to be changed slightly, that there needs to be a little bit more emphasis to the potential liver toxicity," he said.

Hepatotoxicity is discussed in the warnings section of Arava labeling, which notes that Arava was associated with elevated liver enzymes in clinical trials and recommends monitoring ALT levels. Arava labeling includes a "black box" warning, but it discusses only the need to avoid pregnancy while taking the drug.

Voting committee consultant James Lewis, MD, Georgetown University, said the Arava label is "satisfactory" and would change it only "if there's a fatal case or a transplant case that is unequivocal. One case like that would warrant putting it in the label."

Lewis was one of the outside consultants FDA asked to review hepatotoxicity reports submitted to the Adverse Events Reporting System. Of the 18 cases Lewis reviewed that mentioned acute liver failure or death, he did not find any to be probably or definitely related to Arava, although he said five cases were possibly related to the arthritis drug.

Voting committee consultant James Fries, MD, Stanford University, urged FDA not to mention liver injury in the label to avoid "the problem of a false positive....Until we actually have something that rises up out of background, I don't think I would mention it."

Fries noted that when his patients read news coverage of HRG's Arava petition, several said they wanted to stop or did not want to start taking the drug. He expressed disappointment with HRG for "hurting the health of the public by raising what turned out to be false-positive red flags."

Simon asked the committee whether FDA should notify health care providers through a "Dear Doctor" letter or other means in the event FDA requests labeling changes.

"I would say no," Lewis responded. "If you change the labeling, you put in one more layer of liver toxicity, it's already pretty well replete with things that happen in the liver."

"If you're going to go to one case of liver failure that's been reported...or one transplant has occurred, I don't think that rises to the level that I would want to see a 'Dear Doctor' letter or anything else about that. If you accumulate additional information, that's different. But, on the basis of what we discussed today, I don't think it would be necessary."

Acting Committee Chairman Steven Abramson, MD, Hospital for Joint Diseases (New York City), suggested that FDA notify physicians about the committee's finding regarding Arava's hepatotoxicity risk.

"It's important that information be communicated to the other side...that what we heard today...has not seemed to indicate an enhanced risk for this drug," Abramson said.

FDA appears to be leaning toward developing a risk communication program for Arava. In a Feb. 3 memo to Simon, Anti-Inflammatory Division Deputy Director Lawrence Goldkind, MD, said that given the severity of the rare hepatotoxicity cases, "risk communication should be expanded to characterize the risks."

Aventis presented several safety analyses using different sources. In addition to data from pooled clinical trials and postmarketing surveillance, which were included in Aventis' August response to HRG's petition, the company presented results from two studies that relied on health plan patient databases (² Also see "Aventis Arava Labeling Update Expected Based On Methotrexate Combo Study" - Pink Sheet, 16 September, 2002., p. 22).

A retrospective cohort study using Aetna's claims database included 40,594 rheumatoid arthritis patients and over 5,000 leflunomide patients. Follow-up was conducted from September 1998 to December 2000, for 83,143 person-years. Aventis deemed it the "largest RA cohort study ever performed."

The incidence rates for severe hepatic events were 11 per 1,000 person-years for leflunomide, compared to 40 for methotrexate and 120 for other DMARDs, Aventis told the committee. Acute hepatic necrosis occurred in one per 1,000 person-years for leflunomide, compared to two for methotrexate and seven for DMARDs.

A nested case-control study relied on the Protocare Sciences' and PharMetrics' claims databases, with follow-up from September 1998 to December 2001. Follow-up included 41,885 patients who received DMARDs over 51,315 person-years.

Using methotrexate as the reference drug, Aventis found the adjusted rate ratios for leflunomide were not significantly different from one for any serious adverse event, serious hepatic events, serious hematologic events, serious pancreatitis events or serious opportunistic infections and septicemia. "Of note were the generally elevated RRs for the biologic DMARDs, especially for any event, serious liver events, and opportunistic infections and septicemia events."

Prior to the committee meeting, FDA appeared to be leaning toward a finding of an acceptable benefit/risk profile for Arava rather than withdrawing the drug, according to a memo provided as part of the ³ briefing documents for the meeting.

"Leflunomide offers similar advantages as the other therapies, and there is no evidence of increased overall risk associated with leflunomide as compared to the other DMARDs," a Feb. 7 memorandum from FDA's Simon to Office of Drug Evaluation V Director Jonca Bull, MD, states.

Simon's memo says HRG's claim that Arava offers no advantages to patients based on a lack of increased efficacy and an apparent increased risk of serious adverse events "is predicated on inaccurate evidence."

The memo refers to Goldkin's Feb. 3 safety review memo.

The databases and the clinical studies included over 16,000 patients treated with leflunomide and more than 37,000 treated with methotrexate, Goldkind's review notes. "There was no consistent signal for higher rates of serious liver injury in leflunomide-treated patients compared to methotrexate-treated patients."

In addition, Goldkind said he "identified only two cases of probable [acute liver failure] associated with leflunomide" among the postmarketing reports, one of which was associated with an unintentional overdose. Two outside experts also "identified no definite cases of leflunomide-induced ALF," although all three reviewers "considered some cases to be possibly related to leflunomide."

Goldkind concluded that rare cases of acute liver failure "while important to the characterization of drug toxicity profile do not outweigh the potential benefit of this product."

While anti-inflammatory division officials are recommending against withdrawal, an earlier safety review by the Office of Drug Safety reached a different conclusion.

A Nov. 7 memo from Drug Risk Evaluation Division Safety Evaluator Renan Bonnel and ODS Associate Director of Science David Graham, MD, recommended that "leflunomide be withdrawn from the market."

Goldkind disagreed, however, with the methodology used in the review. ODS Director Victor Raczkowski, MD, and Office of Pharmacoepidemiology & Statistical Science Director Paul Seligman, MD, also issued separate memos disagreeing with Bonnel's and Graham's findings.

Bonnel and Graham did not appear at the March 5 meeting, prompting HRG Director Sidney Wolfe, MD, to maintain during the open public hearing that the review division is attempting to "whitewash" the opinions of the drug safety staff.

The Arthritis Advisory Committee also voted in favor of approving an Arava supplemental NDA for improvement in physical function in rheumatoid arthritis patients (see ⁴ Also see "Arthritis “Physical Function” Claim Can Be Shown In 6-12 Months, Cmte. Says" - Pink Sheet, 10 March, 2003. ).

Aventis Arava Liver Toxicity Warning Is Satisfactory, FDA Arthritis Cmte. Says

More from Archive

More from Pink Sheet