Pfizer appears to have used the relabeling of its COX-2 inhibitor Celebrex to add class cardiovascular risk language as an opportunity to renegotiate the placement of safety data.

FDA approved ¹ revised labeling for celecoxib July 29. In the new label, gastrointestinal safety data from the Warnings section has been moved to the Clinical Trials section, and data on platelet aggregation has been shifted from the Precautions section to the Clinical Pharmacology section.

Following an advisory committee's recommendation, FDA requested that sponsors of non-steroidal anti-inflammatory drugs update labeling to include cardiovascular risk information; the agency issued a labeling template in conjunction with the request (² Also see "Celebrex, NSAID Labeling Must Include Skin Reactions Warning, FDA Says" - Pink Sheet, 20 June, 2005., p. 11).

Pfizer CEO Hank McKinnell has suggested that the addition of safety data from long-term trials would help differentiate celecoxib's risk/benefit profile with regard to cancer prevention (³ [A#00670210020], p. 6).

Celebrex labeling now includes safety data from two adenomatous polyp prevention studies, as well as more information from the Celebrex Long-Term Arthritis Safety Study (CLASS). The data appear under the Special Studies heading of the Clinical Studies section.

Of the three trials, language concerning the Adenoma Prevention with Celecoxib (APC) trial most clearly suggests a long-term cardiovascular risk with Celebrex.

"Preliminary safety information from this trial demonstrated a dose-related increase in cardiovascular events (mainly myocardial infarction) at Celebrex doses of 200 mg and 400 mg twice daily compared to placebo," labeling states.

In the APC trial, the relative risk of cardiovascular death, MI or stroke was 3.4 for 400 mg b.i.d. and 2.5 for the 200 mg b.i.d., labeling says. The absolute risk for CV events was 3% for the higher dose, 2.2% for the lower dose and 0.9% for placebo.

"The cumulative rates of serious cardiovascular thrombotic events began to differ between the Celebrex treatment groups and placebo after approximately one year of treatment," labeling notes.

The APC trial results prompted an FDA "Talk Paper" in December recommending limited use of COX-2 inhibitors (⁴ Also see "FDA Issues Caution On COX-2s, But Leaves Open Possibility For First-Line Use" - Pink Sheet, 3 January, 2005., p. 17).

A second long-term polyp study, Prevention of Colorectal Sporadic Adenomatous Polyps (PreSAP), "demonstrated no increased cardiovascular risk for the composite endpoint of cardiovascular death, MI or stroke," labeling states.

"The reason for the differing results for CV events in the APC and PreSAP trials is not known."

Cardiovascular safety data from the CLASS trial, which was previously placed under the heading "Safety Data from CLASS study" in the Adverse Events section of labeling, has been expanded and moved to the Special Studies section.

CLASS compared cardiovascular thromboembolic adverse event rates for Celebrex, diclofenac and ibuprofen. Patients in the study were allowed to take low-dose aspirin.

"The cumulative rates in all patients at nine months for Celebrex, diclofenac and ibuprofen were 1.2%, 1.4% and 1.1%, respectively," labeling states. The cumulative rate of CV events in non-aspirin users was less than 1% for each group.

Nine-month MI rates for patients not taking low-dose aspirin were "less than 0.2%" for each of the three treatment groups.

However, the new labeling notes that "there was no placebo group in the CLASS trial, which limits the ability to determine whether the three drugs tested had no increased risk of CV events or if they all increased the risk to a similar degree." Similar language did not appear in Celebrex' ⁵ previous label.

As outlined in FDA's labeling template, Celebrex labeling also now includes a "black box" warning concerning cardiovascular risk, a contraindication in coronary artery bypass graft patients, and gastrointestinal risk. The Warnings section added class language concerning cardiovascular risk, hypertension, and congestive heart failure and edema.

The new language is generally in line with FDA's NSAID labeling template. The most significant changes to previously existing safety data involved the placement and description of gastrointestinal risk information.

GI safety data from the CLASS study were moved from the Warnings section to the Special Studies heading of the Clinical Studies section. A table presenting Kaplan-Meier nine-month complicated and symptomatic ulcer rates in patients receiving Celebrex 400 mg b.i.d. from CLASS was also moved to the Clinical Studies section.

The Endoscopic Studies heading was moved to the bottom of the Clinical Studies section.

A statement noting that "the correlation between findings of short-term endoscopic studies with Celebrex and the relative incidence of clinically significant upper GI events with long-term use has not been established" was moved from near the bottom of the Endoscopic Studies heading to the top.

A table on "Incidence of Gastroduodenal Ulcers from 3-month Serial Endoscopy Studies in OA and RA" patients was removed. Instead, the key findings from the serial endoscopy studies are now reported via text.

Labeling also adds a statement that "there have been no large, clinical outcome studies to compare clinically relevant GI outcomes with Celebrex and naproxen."

Whereas labeling previously noted that "serious clinically significant upper GI bleeding has been observed in patients receiving Celebrex in controlled and open-label trials, albeit infrequently," the qualifier "albeit infrequently" was dropped in revised labeling.

A paragraph concerning Celebrex' effects on platelet aggregation and bleeding time was moved from the Precautions section to the Clinical Pharmacology section.

A new sentence, added at the bottom of the paragraph on platelet aggregation, states "it is not known if there are any effects of Celebrex on platelets that may contribute to the increased risk of serious cardiovascular thrombotic adverse events associated with the use of Celebrex."

A paragraph concerning fluid retention was also added to the Clinical Pharmacology section. Information concerning fluid retention previously appeared under the heading Fluid Retention, Edema and Hypertension in the Precautions section.

Data from the CLASS trial concerning hypertension and edema were included at the bottom of the new class warnings for hypertension and congestive heart failure and edema.

Consistent with FDA's labeling template, a warning for advanced renal disease was moved ahead of a warning concerning anaphylactoid reactions. The language of the warning was unchanged.

Celebrex' new labeling also corresponds to the NSAID template in noting that another COX-2 selective NSAID has been associated with increased risk for cardiovascular events and stroke in clinical trials in coronary artery bypass graft patients.

"Two large controlled, clinical trials of a different COX-2 selective NSAID for the treatment of pain in the first 10-14 days following CABG surgery found an increased incidence of myocardial infarction and stroke," labeling states.

Two studies of Pfizer's withdrawn COX-2 inhibitor Bextra (valdecoxib) and the Bextra pro-drug parecoxib showed a numerically increased risk of cardiovascular thromboembolic adverse events in CABG patients (⁶ Also see "Bextra Meta-Analysis Affirms Safety Profile, Pfizer Tells Physicians" - Pink Sheet, 6 December, 2004., p. 10).

The Celebrex cardiovascular risk data were added to labeling in conjunction with FDA approval of a new celecoxib indication for "relief of signs and symptoms of ankylosing spondylitis."

Celebrex was previously approved for relief of osteoarthritis and rheumatoid arthritis, management of acute pain, treatment of dysmenhorrhea and reduction of adenomatous colorectal polyps.

Merck initially cited the desire to add data on ankylosing spondylitis as the reason for its March 2002 withdrawal of the NDA for Arcoxia (etoricoxib). The company later acknowledged that the delay was due to the need to supply additional CV safety data to FDA (⁷ Also see "Merck Arcoxia 6,000-Patient Trial Will Study GI, Cardiac Events" - Pink Sheet, 17 June, 2002., p. 15).

Merck resubmitted the Arcoxia NDA with efficacy data on ankylosing spondylitis in December 2003; FDA deemed the drug "approvable" in October 2004, weeks after the withdrawal of Merck's COX-2 inhibitor Vioxx (rofecoxib).

Wyeth/Amgen and Johnson & Johnson's respective TNF inhibitors Enbrel (etanercept) and Remicade (infliximab) are approved for ankylosing spondylitis; Abbott is also pursuing an ankylosing spondylitis indication for its TNF inhibitor Humira (adalimumab).

Pfizer Celebrex Revised Labeling Adds Class Cardio Risk, Moves Safety Data

More from Archive

More from Pink Sheet