Vioxx: what did Merck know and when did it know it?

Merck may have been aware of the heightened cardiovascular risk of rofecoxib (Vioxx) years before it withdrew the drug from the market in 2004, according to an article appearing in the Archives of Internal Medicine.

Joseph Ross and colleagues analyzed data available to the manufacturer from 30 randomized trials and enrolling over 20,000 subjects. As early as December 2000, only a year and a half after the drug’s approval in May 1999, the data from the trials available at the time gave a strong cardiovascular safety signal (RR= 2.18, CI=0.93-5.81, p=0.07). The signal grew increasingly convincing, reaching significance (p=0.05) by June 2001 with a 35% increased risk of cardiovascular thromboembolic events or death. The signal continued to grow stronger: by September 2004 the use of rofecoxib was associated with a highly significant 43% elevated risk (RR=1.43, CI=1.16-1.76, p<0.001).

“Physicians and the public deserve to be in a position to make informed choices about risks and benefits, and the disclosure and dissemination of information about potential risk immediately after its recognition is absolutely essential. Our study provides insight into what should have been known about the risks of rofecoxib,” write the authors in their conclusion. “If we are to detect harms early and protect the public’s health, while ensuring the availability of new, clinically effective therapeutics, a system must be established that makes full use of all existing evidence.”

In an “Invited Commentary,”  Steven Woloshin and Lisa Schwartz write that “the Vioxx story really highlights the difference between marketing and informing…. The problem is that when it comes to prescription drugs, a lot more effort goes into marketing than informing.” They write that “Merck spent $500 million promoting Vioxx in 2003 along– an amount equal to the FDA’s total annual budget for all human drugs.” They conclude that  FDA needs to mandate “widespread dissemination” of available data to help balance the odds.

In a statement to the press (see below, along with a more complete response from Harlan Krumholz, senior author of the paper), Merck says the Archives article “used unreliable methods and reached incorrect conclusions.” Krumholz provided the following response to CardioBrief:

The methods we employed represent a well accepted approach that passed rigorous peer review and demonstrates the harm of Vioxx that eventually led to the withdrawal of Vioxx from the worldwide market. Merck may criticize the methods but what we found beginning with data available as early as 2000 was what they eventually admitted to be true – that Vioxx increases risk.

Here is the press release from Archives of Internal Medicine:

Adverse Heart Effects of Rofecoxib May Have Been Identified Years Earlier
CHICAGO—Clinical trial data indicated an association between the anti-inflammatory medication rofecoxib and cardiovascular risk as early as December 2000, before the product was taken off the market in September 2004, according to a report in the November 23 issue of Archives of Internal Medicine, one of the JAMA/Archives journals.

Rofecoxib was introduced to the market in May 1999 and quickly became a commercial success, with sales reaching $2 billion annually, according to background information in the article. The manufacturer marketed the product (with the brand name Vioxx) as a safer alternative to traditional nonsteroidal anti-inflammatory drugs. However, concerns about its cardiovascular adverse effects reportedly existed during the drug development process. In September 2004, the manufacturer voluntarily withdrew the product from the market after one large trial was terminated early due to an increased risk of cardiovascular events.

In November 2004, the manufacturer’s chief executive testified before a U.S. Senate committee that until the halted trial, combined data from all randomized controlled clinical trials showed no difference in the risk of confirmed heart events between patients taking rofecoxib and those taking placebo. To assess whether and when analysis of published and unpublished clinical trial data could have revealed the cardiovascular risks of rofecoxib, Joseph S. Ross, M.D., M.H.S., of Mount Sinai School of Medicine, New York, and colleagues conducted a pooled analysis of all such trials conducted by the manufacturer before September 2004.

The researchers identified 30 randomized, placebo-controlled trials that enrolled a combined 20,152 individuals, lasted from four weeks to four years and assigned a range of 17 to 2,586 participants to take doses of rofecoxib ranging from 12.5 milligrams to 50 milligrams. The authors pooled the data from these studies and analyzed the cumulative results.

“As of December 2000, 21 of these trials had been completed (70 percent) and the risk of a cardiovascular thromboembolic [heart- or blood clot-related] adverse event or death was greater among subjects assigned to the rofecoxib group, raising concerns from a safety standpoint,” the authors write. “Subsequently collected data through June 2001 showed that rofecoxib was associated with a 35-percent increased risk of a cardiovascular thromboembolic adverse event or death.” The association strengthened as additional data became available—as of April 2002 the pooled analysis showed a 39-percent increased risk, and as of September 2004, a 43-percent increased risk.

The analyses provide a roadmap for how drug safety can be assessed after a product has been introduced into the market, the authors note. New legislation requiring the public disclosure of trial results in the ClinicalTrials.gov database will make available substantial data that has not previously been used to understand drug safety or efficacy. Independent investigators will now be able to conduct comprehensive meta-analyses that can complement and corroborate surveillance done by the U.S. Food and Drug Administration.

“Physicians and the public deserve to be in a position to make informed choices about risks and benefits, and the disclosure and dissemination of information about potential risk immediately after its recognition is absolutely essential. Our study provides insight into what should have been known about the risks of rofecoxib,” the authors conclude. “If we are to detect harms early and protect the public’s health, while ensuring the availability of new, clinically effective therapeutics, a system must be established that makes full use of all existing evidence.”
(Arch Intern Med. 2009;169[21]:1976-1985.

Editor’s Note: This project was not directly supported by any external grants or funds. However, Dr. Ross is currently supported by a National Institute on Aging grant and by the American Federation of Aging Research through the Paul B. Beeson Career Development Award Program. Please see the article for additional information, including other authors, author contributions and affiliations, financial disclosures, funding and support, etc.

Here is Merck’s response to the article:

Merck Statement on Article in The Archives of Internal Medicine Involving VIOXX®

WHITEHOUSE STATION, N.J., November 23, 2009 – Merck believes the article published in The Archives of Internal Medicine in today’s issue related to VIOXX used unreliable methods and reached incorrect conclusions.

The scientific exercise of examining what the data showed about VIOXX’s cardiovascular safety over time is an important one. Indeed, Merck scientists did exactly that, in real time and with pre-specified rules while VIOXX was on the market. These regularly updated analyses of a large body of data from randomized clinical trials, conducted in consultation with outside experts and with regulatory agencies, did not demonstrate an increased risk of thrombotic events when VIOXX was compared to placebo while VIOXX was on the market.

The first time Merck observed a difference in a placebo-controlled study was when it learned the results of the APPROVe study in September 2004. We voluntarily withdrew VIOXX from the market within a week of those results.

The methods used by the authors are not valid for a number of reasons, including:

• The authors relied exclusively on “investigator-reported” cardiovascular (CV) events. Merck took the same investigator-reported events and put them through a rigorous cardiovascular adjudication process put in place in 1999 in response to suggestions by its outside scientific consultants. The cardiovascular adjudication process was created to improve upon the accuracy of investigator-reported thrombotic CV events by sending patients’ medical records to external blinded experts to determine if the events reported by investigators were in fact thrombotic. Merck’s real time analyses of VIOXX’s cardiovascular safety focused on the events that were confirmed by these blinded external adjudicators as thrombotic. This procedure was more rigorous than that used by the authors of the Archives paper.
• The authors then combined this less reliable endpoint with “all cause” deaths. We do not believe deaths from all causes — including, in the case of VIOXX, electrocutions, infections and trauma — are the most reliable way to examine a question about cardiovascular safety.
• The authors also included events that occurred after a patient had stopped taking any study drug. For example, many of the Alzheimer’s disease patients were followed for up to a year or longer after they had stopped the study drug. Because the question being considered was the effect of VIOXX on thrombotic events while patients were taking the medicine, it does not make sense from a medical perspective to include events that occurred long after patients had stopped taking any medicine.

Merck acted responsibly – from researching VIOXX prior to approval in studies with approximately 10,000 patients – to monitoring the medicine while it was on the market — to voluntarily withdrawing the medicine when it did. Our decisions were based on the data from well-controlled clinical trials.

Status of Litigation

Merck voluntarily withdrew VIOXX from the market on September 30, 2004. In November of 2007, Merck entered into an agreement to resolve state and federal myocardial infarction and ischemic stroke personal injury claims filed or tolled by Nov. 9, 2007. More than 99 percent of all eligible personal injury claimants enrolled in the program, and the program is proceeding as scheduled.

About Merck

Today’s Merck is working to help the world be well. Through our medicines, vaccines, biologic therapies, and consumer and animal products, we work with customers and operate in more than 140 countries to deliver innovative health solutions. We also demonstrate our commitment to increasing access to healthcare through far-reaching programs that donate and deliver our products to the people who need them. Merck. Be Well. For more information, visit www.merck.com

Forward Looking Statement

This news release includes “forward-looking statements” within the meaning of the safe harbor provisions of the United States Private Securities Litigation Reform Act of 1995. Such statements may include, but are not limited to, statements about the benefits of the merger between Merck and Schering-Plough, including future financial and operating results, the combined company’s plans, objectives, expectations and intentions and other statements that are not historical facts. Such statements are based upon the current beliefs and expectations of Merck’s management and are subject to significant risks and uncertainties. Actual results may differ from those set forth in the forward-looking statements.

The following factors, among others, could cause actual results to differ from those set forth in the forward-looking statements: the possibility that the expected synergies from the merger of Merck and Schering-Plough will not be realized, or will not be realized within the expected time period, due to, among other things, the impact of pharmaceutical industry regulation and pending legislation that could affect the pharmaceutical industry; the risk that the businesses will not be integrated successfully; disruption from the merger making it more difficult to maintain business and operational relationships; Merck’s ability to accurately predict future market conditions; dependence on the effectiveness of Merck’s patents and other protections for innovative products; the risk of new and changing regulation and health policies in the U.S. and internationally and the exposure to litigation and/or regulatory actions.

Merck undertakes no obligation to publicly update any forward-looking statement, whether as a result of new information, future events or otherwise. Additional factors that could cause results to differ materially from those described in the forward-looking statements can be found in Merck’s 2008 Annual Report on Form 10-K, Schering-Plough’s Quarterly Report on Form 10-Q for the quarterly period ended September 30, 2009, the proxy statement filed by Merck on June 25, 2009 and each company’s other filings with the Securities and Exchange Commission (SEC) available at the SEC’s Internet site: www.sec.gov.

# # #

Here is a more detailed response to the Merck statement from Harlan Krumholz on behalf of the study authors:

In response to the Merck criticisms, adjudicated endpoints were available for only a fraction of the trials and adjudications were conducted under the auspices of Merck, so we felt it would be better to take the direct reports of the study physicians and be able to include all the endpoints and not be concerned about whether there was a problem in the adjudication process. We did include patients who had discontinued Vioxx, an approach that is called intention-to-treat, which is widely accepted as the gold standard approach in clinical trials analysis. We did include all deaths as the underlying cause of the death is often difficult to ascertain and the use of all deaths ensures that there is no opportunity to manipulate the findings through the adjudication process. Moreover, many cardiology trials use this approach. The use of investigator reported events, the use of intention to treat and the inclusion of all deaths would tend to make it more difficult to see a safety problem. If this analysis were negative then other analyses might be conducted to explore further. In this case, we found a signal early despite using this very conservative approach.