What impact does Merck's withdrawal of rofecoxib have on AD patients?

Some AD patients have been taking rofecoxib and other NSAIDs to treat the AD. There is little justification for this. Each large trial of rofecoxib for the treatment of AD or MCI has been unequivocally negative. Those nonetheless taking rofecoxib for AD should certainly stop.

Many AD patients have been taking rofecoxib appropriately for treatment of arthritis or pain. At this point, such patients should discuss alternatives, such as acetaminophen, non-selective NSAIDs or COX-2 selective drugs, with their physicians. In my view, acetaminophen is the least risky. COX inhibitors carry important risks to the gastrointestinal tract (this is particularly true of non-selective NSAIDs) and to the kidneys.

Does the apparent increased risk of cardiovascular events with rofecoxib carry over to other COX-2 inhibitors such as celecoxib? This is unclear. While it is physiologically plausible that any COX-2 inhibitor may carry some risk of thromboembolism, I am not aware that there is any evidence of...  Read more

What impact does Merck's withdrawal of rofecoxib have on AD patients?

Some AD patients have been taking rofecoxib and other NSAIDs to treat the AD. There is little justification for this. Each large trial of rofecoxib for the treatment of AD or MCI has been unequivocally negative. Those nonetheless taking rofecoxib for AD should certainly stop.

Many AD patients have been taking rofecoxib appropriately for treatment of arthritis or pain. At this point, such patients should discuss alternatives, such as acetaminophen, non-selective NSAIDs or COX-2 selective drugs, with their physicians. In my view, acetaminophen is the least risky. COX inhibitors carry important risks to the gastrointestinal tract (this is particularly true of non-selective NSAIDs) and to the kidneys.

Does the apparent increased risk of cardiovascular events with rofecoxib carry over to other COX-2 inhibitors such as celecoxib? This is unclear. While it is physiologically plausible that any COX-2 inhibitor may carry some risk of thromboembolism, I am not aware that there is any evidence of increased cardiovascular risk with celecoxib.

What is the implication of the Merck announcment for the ADAPT study? Not much, in my view. Again, I am not aware of any evidence of increased cardiovascular risk with celecoxib. Recent studies have suggested that celecoxib may be associated with lower risk of hypertension and congestive heart failure than rofecoxib. I do not see a valid reason to alter the ADAPT trial based on the new data on rofecoxib.

View all comments by Paul Aisen

Alzforum subscribers have no doubt heard that Merck has withdrawn its blockbuster COX-2 inhibitor Vioxx (rofecoxib) from the market because of increased risks of heart attack and stroke. Most are aware also that the COX-2 inhibitors, including Celebrex (celecoxib, Pfizer), and Bextra (valdecoxib) as well as Vioxx, are a newer class of nonsteroidal antiinflammatory drugs (NSAIDs). Because of the sustained interest in NSAIDs as a possible way to protect against Alzheimer’s disease (AD), the withdrawal of Vioxx is of probable interest to readers, and the Alzforum editors have asked me to comment.

First, any person currently taking Vioxx should discontinue the drug. This applies with full force to those who have used the drug hoping for protection against AD.

Second, there are published, definitive results that disprove any benefit of Vioxx for disease modification (slowed progression of symptoms) in those with Alzheimer’s dementia. Even before the Merck withdrawal, therefore, there was no justification for the use of Vioxx to protect against the progression of dementia....  Read more

Alzforum subscribers have no doubt heard that Merck has withdrawn its blockbuster COX-2 inhibitor Vioxx (rofecoxib) from the market because of increased risks of heart attack and stroke. Most are aware also that the COX-2 inhibitors, including Celebrex (celecoxib, Pfizer), and Bextra (valdecoxib) as well as Vioxx, are a newer class of nonsteroidal antiinflammatory drugs (NSAIDs). Because of the sustained interest in NSAIDs as a possible way to protect against Alzheimer’s disease (AD), the withdrawal of Vioxx is of probable interest to readers, and the Alzforum editors have asked me to comment.

First, any person currently taking Vioxx should discontinue the drug. This applies with full force to those who have used the drug hoping for protection against AD.

Second, there are published, definitive results that disprove any benefit of Vioxx for disease modification (slowed progression of symptoms) in those with Alzheimer’s dementia. Even before the Merck withdrawal, therefore, there was no justification for the use of Vioxx to protect against the progression of dementia. In fact, efforts to show such a protective effect with numerous other antiinflammatory treatments have also failed. These other drugs include prednisone, Celebrex, naproxen or Aleve (naproxen, Bayer) and hydroxychloroquine, an anti-malarial with potent antiinflammatory action. It appears that antiinflammatory agents have no place in the treatment of AD.

Some hope has remained that NSAIDs or other antiinflammatory treatments could stop progression to dementia in those with mild cognitive impairment (MCI). Here the data are more sparse, but there is at least one publicly disclosed trial using Vioxx for this purpose in patients with MCI. The trial produced the disappointing result that the incidence of AD was higher in those assigned to Vioxx than in those receiving placebo. It is thus clear that Vioxx cannot prevent dementia in those with MCI, and may actually promote it. The increased risk of dementia onset with Vioxx may well reflect its now-known thrombogenic properties, because cerebrovascular disease is a known risk factor for the expression of clinical dementia in people with AD pathology. There are no other published or reported trials of any NSAIDs for prevention of dementia in individuals with MCI, so we don’t know for certain whether these might work. However, I can think of no reason to expect that any other COX-2 inhibitor would be effective for this purpose when Vioxx was not. This leaves open the possibility that a conventional “dual inhibitor” NSAID such as ibuprofen or naproxen might prevent dementia in those with MCI. Unfortunately, the epidemiological data argue also against this possibility. Both the Rotterdam Study and the Cache County results suggested that any protective effect of conventional NSAIDs is lost some two or three years prior to the onset of AD, i.e., at the time when many incipient cases would show MCI.

So why all the fuss about NSAIDs for protection against AD? The epidemiological data suggest that prolonged use of NSAIDs can result in a dramatic decrease in the incidence of AD after a lag period of two or three years. In other words, the most promising indication for use of NSAIDs to protect against AD is in a “primary prevention” strategy. Peter Zandi’s paper at the recent World Alzheimer Congress showed that several different conventional NSAIDs, including both “Aβ lowering” drugs such as ibuprofen, and “non-Aβ-lowering” compounds such as naproxen, are associated with a similar reduction in AD incidence. It is also noteworthy that aspirin, a “non-Aβ-lowering” dual-inhibitor NSAID with strong COX-1 activity, was also associated with reduced risk of AD in the Cache County Study.

Unfortunately, there are no published data on the potential of selective COX-2 inhibitors for primary prevention of AD. The Alzheimer’s Disease Antiinflammatory Prevention Trial (ADAPT) is currently testing both the COX-2 inhibitor Celebrex and the dual inhibitor naproxen as agents for primary prevention. Upon completion, ADAPT should inform us whether a COX-2 inhibitor can prevent AD when given to asymptomatic individuals. This would be an important result because it would show that the action of NSAIDs is mediated by their effects on COX-2, and also because it is likely that—notwithstanding the Vioxx findings—selective COX-2 inhibitors are “the wave of the future” for NSAID therapy.

Should ADAPT participants be concerned by the recent Vioxx debacle? The FDA has noted that it will continue to monitor closely the possible thrombogenic effects of other COX-2 inhibitors. Indeed, it is only prudent that they should. It is important to note, however, that there are no preliminary data to suggest an increased risk of cardiovascular disease in users of celecoxib. In fact, a recently presented paper suggested a dichotomy between Vioxx and Celebrex as risk factors for myocardial infarction. The paper reported a large observational study in the Kaiser-Permanente database suggesting that users of Vioxx (especially users of very high doses) were at substantially increased risk of heart attack and stroke. In these same analyses, users of Celebrex showed risks that were lower than those in an untreated control group.

Thus, we are advising ADAPT participants that there is no cause at present to be concerned about increased cardiovascular risks in those assigned to Celebrex. Like the FDA, we will monitor ADAPT participants closely for any evident increase in cardiovascular risks. However, the trial’s safety monitoring committee has thus far found no reason to express concerns about increased risks in ADAPT participants assigned to any treatment.

I know that some people may use NSAIDs “off label” in hopes these drugs may prevent Alzheimer’s disease. Except for those enrolled in trials, I do not recommend this practice, and I urge that those who follow it nonetheless do so only in consultation with a physician. The doctor can monitor for the known risks of NSAID use (e.g., gastrointestinal bleeding, diminished kidney function), as we do in ADAPT. In view of current concerns about COX-2 inhibitors, and because there are no epidemiological data to suggest their utility for AD prevention, I urge specifically against the use of COX-2 inhibitors “off label” for AD prevention (unless, of course, you are an ADAPT participant!)

View all comments by John Breitner