AHA: Study Finds Celebrex Safe for Heart

December 2, 2016

Article

Results of the PRECISION study are embraced by some and soundly criticized by others. What would you tell your osteoarthritis patients?

But naproxen not as heart safe as believed

NEW ORLEANS -- After 10 years during which 24,000 osteoarthritis (OA) and rheumatoid arthritis (RA) patients were randomized, the verdict on cardiovascular safety of NSAIDs is in and it's not all that surprising: celecoxib (Celebrex) is no more likely to cause heart attacks or stroke than naproxen or ibuprofen.

In fact the surprise in the long-awaited PRECISION study was not about the cardiovascular safety of the selective cyclooxygenase-2 (COX-2) inhibitor, but rather about the non-selective agent: naproxen.

"These findings challenge the widely held view that naproxen provides superior cardiovascular safety," Steven E. Nissen, MD, of the Cleveland Clinic, reported here in the first late-breaking clinical trials session at the American Heart Association annual meeting.

The PRECISION study results also were published simultaneously online in the New England Journal of Medicine.

In an intention to treat analysis, 188 patients in the celecoxib group reached the endpoint of death, non-fatal myocardial infarction (MI), or non-fatal stroke compared with 201 patients in the naproxen arm and 218 patients who received ibuprofen (P<.001 for noninferiority).

Click here for more video comments from authors of the Hot Lines trials, and leading cardiologists from around the world providing daily commentary.

A per protocol analysis had similar results: 134 celecoxib patients versus 144 in the naproxen group and 155 ibuprofen patients (P<.001 for noninferiority).

As expected, there were fewer gastrointestinal events -- including GI bleeds -- among celecoxib patients than naproxen and ibuprofen, and the risk of renal events were lower with celecoxib than with ibuprofen, which was significant. But there was no significant renal protection for celecoxib compared with naproxen.

About two-thirds of the 23,081 randomized patients were women and the average age was 63. The majority of patients were white (75%) and about 90% had OA. They were randomized to celecoxib (100 mg b.i.d.), ibuprofen (600 mg t.i.d.) or naproxen (375 b.i.d.). All of the participants had either existing cardiovascular disease or were at increased risk.

Former AHA president Elliot Antman, MD, of Brigham and Women's Hospital and Harvard Medical School in Boston, had a more critical view of the trial. Antman, who served as discussant for Nissen's presentation, said most of the patients were really low risk for cardiovascular events.

"These are not the people that are a concern with COX-2 agents" he told MedPage Today.

Nissen countered noting that "if we had enrolled only people who had a history of heart attacks, we would still be recruiting for this study." He pointed out that about a third of the participants had diabetes, which is a recognized cardiovascular risk factor.

Backstory

The study has an interesting history: it was ordered by the FDA in response to concern about the safety of nonsteroidal anti-inflammatory drugs (NSAIDs) -- and especially concern about the safety of COX-2 inhibitors, triggered by the finding that rofecoxib (Vioxx) appeared to increase the risk of MI and stroke when compared with naproxen.

That risk was first reported in a 2001 pooled analysis published in the Journal of the American Medical Association, and Nissen was as a co-author of that paper. The initial response to that report was the claim that comparing the COX-2 drugs with naproxen introduced bias because naproxen was cardio-protective, critics claimed.

But doubt had been raised, and in 2004, a study investigating rofecoxib for chemoprevention of colorectal cancer was halted due to excess heart attacks and stroke, Merck pulled Vioxx from the market.

In April 2005, the FDA ordered another COX-2, valdecoxib (Bextra), pulled from the market, and added black box warnings about cardiovascular risks to all NSAID labels. Aspirin was the sole exception to the black box label.

Celecoxib was allowed to remain on the market, but to gain that reprieve, Pfizer was ordered to conduct PRECISION, a cardiovascular safety trial.

Moreover, going forward the FDA ordered drugmakers to include cardiovascular safety testing in all drugs, not just drugs designed to treat cardiovascular conditions.

'Practice-Changing'?

"It [the COX-2 scare] changed a lot of things" said Milton Packer, MD, of Baylor University Medical Center in Dallas. "It confirmed that drugs for non-cardio indications could have important cardiovascular effects."

Packer praised the PRECISION trial as an example of the what needs to be done "to get the right answer and sometimes that takes 10 years."

He said the finding on naproxen "should be practice-changing, if people read these findings and understand them. Based on this [PRECISION], it appears that the best choice for patient who needs an NSAID is celecoxib," Packer said.

But understanding could be difficult because the trial is packed with so many comparisons -- celecoxib versus both naproxen and ibuprofen; celecoxib versus each of the agents; naproxen versus ibuprofen.

That said, Packer said there was a very straightforward and important finding: "the major cardiovascular risk with NSAIDs is not their effect on platelets, but the effect on blood pressure."

In PRECISION, celecoxib was associated with the lowest average increase in blood pressure (2.3%) versus ibuprofen (3.1%) and naproxen (2.5%). Noting that NSAIDs are used chronically by people with RA or OA, Packer said "if blood pressure increases by 4 to 6 mmHg over a period of years, you are going to hurt people ... you make heart failure worse, raise the risk for MI and stroke."

Nissen agreed that blood pressure is a concern, but noted that hypertension hospitalizations were higher in both the ibuprofen and naproxen arms compared with the celecoxib group.

"The hazard ratio for hospitalization with hypertension was 60% higher with ibuprofen," he said.

The evidence linking blood pressure to cardiovascular risk, "destroys the FitzGerald hypothesis that inhibiting COX-2 disturbs the balance between prostacyclin and thromboxane. That hypothesis, which is not correct, which has driven cardiovascular thinking for 20 years is no longer supported," he added.

'Mostly Low Risk OA Patients'

Packer was referring to the work of Garret A. FitzGerald, MD, of the Perelman School of Medicine at the University of Pennsylvania.

FitzGerald disagreed with that assessment. He took exception to most of PRECISION in a tartly worded commentary piece that will be published online in Circulation.

Nissen's group was wrong on a number of counts, according FitzGerald, who charged that PRECISION "is not a study of arthritis patients at high cardiovascular risk" but rather was a trial of mostly low risk OA patients who lacked the "underlying cardiovascular risk substrate." This was a fatal flaw because that underlying risk is key to NSAID mediated harm, he stated.

"Furthermore, this is a nonâinferiority trial. Due to the low number of events accruing, the statistical upper bound was relaxed during the trial from 1.3 to 1.4 (with a power of only 80%). How likely would celecoxib be found inferior? The RR for serious vascular events from celecoxib is 1.36," Fitzgerald wrote.

Fitzgerald questioned the drug exposure, pointing out that although physicians could increase the doses of ibuprofen or naproxen, the daily celecoxib dose was limited to just over 209 mg. "Indeed, the higher rates of hypertension and renal effects of the other two drugs [ibuprofen and naproxen] are also consistent with lower drug exposure in the celecoxib group," he wrote.

Antman, too, faulted the PRECISION investigators on this point, claiming that the celecoxib arm had worse pain control and a higher drop out rate. Nissen disputed both claims saying the pain control was equal across all three arms and there was no significant difference in drop-out rates.

In addition to an interesting history, PRECISION had a number of difficulties during its 10-year history, and the greatest of those was a difficulty keeping patients and following them. The study authors acknowledged that "68.8% of the patients stopped taking the study drug, and 27.4% of the patients discontinued follow-up."

This point was not lost on FitzGerald, who also slammed the study for the high dropout rate and poor follow-up rate.

"In summary, there are so many problems with the interpretation of PRECISION that it fails to inform clinical practice. Thus, despite the enrollment of >24,000 patients and more than a decade of study, we are no closer to being able to advise the millions of patients with chronic arthritic pain regarding relative efficacy and safety of the treatments available to them," he concluded.

The PRECISION trial was funded by Pfizer.

Nissen disclosed a relevant relationships with Pfizer.

Packer disclosed relevant relationships with Admittance, Amgen, AstraZeneca, Bayer, BioControl, Boehringer Ingelheim, Boston Scientific, Celyad, Cardiorentis, Daiichi Sankyo, GlaxoSmithKline, Novartis, NovoNordisk, Relypsa, Takeda, and ZS Pharma.

Fitzgerald disclosed a relevant relationship with Pfizer.

last updated 11.15.2016

This article was first published on MedPage Today and reprinted with permission from UBM Medica. Free registration is required.