Aetna Plans “Clinical Alert” On Plavix Use With PPIs Based On Claims Review

Aetna is developing a clinical alert for physicians and pharmacists to warn about potentially reduced effectiveness for Plavix (clopidogrel) when the antiplatelet drug is used concurrently with a proton pump inhibitor, such as Prilosec (omeprazole) or Nexium (esomeprazole).

Plans for the alert are based on a retrospective analysis by the insurer using its pharmacy and medical claims databases. In the analysis of over 1,000 members covered by its plans, Aetna found higher rates of myocardial infarction among those taking both drugs concurrently.

The results of the analysis were published as a letter to the editor in the Sept. 16 Journal of the American College of Cardiology. Aetna also announced the results in a press release and informed FDA about its analysis. The agency says it is reviewing the findings.

PPIs are prescribed in almost half of patients taking clopidogrel, Aetna said. PPIs are frequently prescribed to patients on clopidogrel to avoid serious gastrointestinal bleeding. Bristol-Myers Squibb and Sanofi-Aventis market Plavix. AstraZeneca markets Nexium as well as prescription Prilosec, which is also sold in an OTC formulation and in generic versions. [Editor's Note: this paragraph was updated Oct. 8 to correct marketing information.]

The review was conducted by a recently-formed pharmacy outcomes research team within the insurer's internal pharmacy benefit management unit, Aetna Pharmacy Management. Results were "confirmed" by the company's health data analytics group, Aetna Informatics, a spokesman said.

The letter published in JACC was co-authored by Edmund Pezalla, national medical director for Aetna Pharmacy Management and David Day, head of pharmacy outcomes research for Aetna.

Promoting Safe Prescribing

In both the letter and the release, Aetna acknowledges that its review has some limitations but maintains its findings are important enough to warrant an alert.

The letter acknowledges that claims-based analyses "are limited" and cannot control for variables such as "years of risk factor presence, weight gain, smoking history, and family history of coronary heart disease."

Nevertheless, the authors maintain, "we feel that evidence is pointing toward a potentially significant interaction between PPIs and clopidogrel that may decrease the ability of clopidogrel to prevent acute MI events."

"This is one study, one piece of the data puzzle, but it certainly suggests that we may be able to promote improved patient safety through outreach and education," Pezalla said in the release.

Need For Caution

Members of the drug industry are wary of attaching too much significance to the claims review. Plavix marketer Bristol-Myers Squibb said it is "evaluating the preliminary data" and is interested in further details on how the review was carried out.

The company is concerned that the retrospective analysis was not able to control for all potential confounding variables or definitively establish causality.

PPI researcher David Johnson also emphasized the need for caution in interpreting results of such an analysis. Johnson is a professor at Eastern Virginia School of Medicine and immediate past president of the American College of Gastroenterology. He also has led studies on Nexium.

Johnson suggested that the "worst thing" that could happen as a result of the Aetna analysis is that patients on the concurrent therapies would stop taking their PPI.

Aetna reviewed claims filed during 2006 for 1,010 members younger than 65 who were newly started on clopidogrel at the beginning of that year and were also taking either omeprazole or esomeprazole.

Those concurrently taking a PPI were divided into a low exposure group (on the drug for less than six months) and a high exposure group (for more than six months). Members taking clopidogrel but not a PPI were assigned to a control group.

Aetna found one-year acute MI rates of 1.38 percent in the control group, 3.08 percent in the low PPI exposure group, and 5.03 percent in the high PPI exposure group, the researchers reported.

After adjusting for differences in comorbidities between the control group and high PPI group, "differences in acute MI rates between the control and high PPI groups remained significant," the researchers reported. "Relative risk for acute MI in the high PPI exposure group was 337 percent greater than in the control group."

A Data-Mining Trend

The move is a novel one for the insurer but reflects Aetna's interest in promoting cost-effective care, Pezalla said.

It also is in line with activities by other large insurers that have been spurred by FDA's plan for a comprehensive post-marketing drug safety initiative - called Sentinel - using a national network of electronic databases. Aetna is interested in working with the agency on that initiative, Pezalla said.

WellPoint has expressed a similar interest. That firm announced in the spring it is working on its own drug safety monitoring system (called Safety Sentinel) designed to continually draw data from its integrated medical and pharmacy claims databases ¹ (Also see "WellPoint Drug Safety Monitoring System Expected To Go Live In Mid-2009" - Pink Sheet, 21 Apr, 2008.), p. 18). The system is expected to begin operation in 2009.

Aetna's analysis raises some of the key drug industry concerns with FDA's Sentinel project, including how much weight should be given to such retrospective analyses and how much they should influence patient care. The concerns were aired during a recent public conference on Sentinel (² (Also see "FDA’s Sentinel Framework Should Preclude “Unbridled” Data Mining – Platt" - Pink Sheet, 23 Jun, 2008.), p. 11).

Aetna's claims review was prompted by a small clinical trial published in January 2008 in JACC that found omeprazole "significantly decreased clopidogrel's inhibitory effect on platelet P2Y12." Involving 124 patients, the study was conducted by Martine Gilard, Brest University Hospital, Brest, France, et al.

In a response letter published along with Aetna's analysis, Gilard et al., said that "additional randomized trials with clinical endpoints must be performed. Nevertheless, biological data from our randomized double-blind trial, confirmed by clinical results of registries such as the one reported ... by Dr. Pezalla ... should lead us to avoid systematic addition of PPIs when clopidogrel is prescribed."

- Cathy Kelly ([email protected])