|
Medical errors and the risks of pharmaceuticals have been in the news lately. The diabetes drug troglitazone (Rezulin) was withdrawn in late March following reports of 63 deaths and 90 cases of liver failure linked to the drug. Cisapride (Propulsid), which has been used to treat severe heartburn and gastroesophageal reflux disease, is being voluntarily withdrawn by Janssen Pharmaceuticals, following 341 reports of arrhythmia and 80 deaths. Late in 1999, the rotavirus vaccine Rotashield was withdrawn after 99 reports of bowel obstruction (intussusception) and the deaths of two infants. Since withdrawal of its drugs fenfluramine and dexfenfluramine (Redux) in 1997, American Home Products continues to seek settlements in more than 4000 lawsuits relating to heart valve abnormalities linked to the fen–phen combination of anti-obesity drugs. To Err Is Human, the Institute of Medicine’s (IOM’s) report on medical errors released late last year, drew public attention to the economic and human costs of medical errors, which the report says are responsible for an estimated 44,000 to 98,000 deaths annually. Among other proposals, the IOM report called for the creation of a national patient safety center, with an initial budget of $30–35 million and expanding to $100 million, that would develop the tools and systems necessary to prevent medical errors. Creation of the CERTs Four CERTs have now been established under the administrative oversight of the Agency for Health Research and Quality (AHRQ), formerly the Agency for Health Care Policy and Research. Each CERT is conducting multiple postmarketing drug safety and efficacy studies within a broad area of focus (such as cardiovascular disease or arthritis), typically involving a vulnerable, understudied population, such as children, women, or the elderly. With the release of the IOM report, the recent product removals, and heightened consumer interest in safety issues, “there’s a confluence of priorities right now, and the CERTs may be the right thing at the right time,” according to Peter Honig, director of the Office of Postmarketing Drug Risk Assessment at the FDA. AHRQ Director John Eisenberg concurs, saying, “In some studies about 19% or 20% of the medical errors that occur are related to the prescribing, dispensing, or taking of pharmaceuticals, and so we see the CERTs as a major component of a nationwide effort to improve patient safety.” Man proposes, Congress disposes The four CERTs established under FDAMA resemble Woosley’s proposal in name and mandate, but not in budget. He had proposed a network of 15 regional centers with a total budget of $75 million. To fulfill their ambitious program of postmarketing research and provider education, the CERTs are encouraged by AHRQ to seek additional funding through partnerships with for-profit and nonprofit organizations. But such partnerships may appear to create conflicts of interest if the CERTs make recommendations—such as how to use drugs more appropriately—that benefit some companies more than others, according to Peter Honig, who represents the FDA on the CERT steering committee. “The CERTs are wrestling with the issue of having relatively small amounts of appropriated money and trying to exercise a broad mandate, so they are looking for the most productive and appropriate way of interacting with industry. There is a fine line to walk here.” When good drugs go bad The FDA’s voluntary reporting system (www.fda.gov/medwatch) has proved effective in picking up rare but serious events suggesting drug toxicity, such as agranulocytosis or liver failure (6). But, “unless the increase [in events] is really huge, spontaneous reporting is not very good at discovering events that are common in the population, like heart attacks or stroke,” said Robert Temple, director of the Office of Medical Policy at the FDA’s Center for Drug Evaluation and Research. “And unless the difference is large, epidemiologic methods also are not very credible and frequently make errors.” Temple cites the putative link between reserpine and breast cancer, reported in one well-designed study and confirmed by two others. Yet later studies were unable to replicate the finding. One postmarketing monitoring option, according to Temple, involves record linkage studies using large Medicaid and HMO databases that offer the capability of linking diagnoses with medication history. Aside from incomplete data, a major limitation of record linkage studies is the possibility of selection bias. Patients who received a particular drug might have been different in some fundamental way from those who did not, putting them at higher risk of the event being studied. Other postmarketing monitoring efforts are attempting to locate evidence of possible adverse effects in disease and drug registries, such as one maintained by a consortium of liver transplant surgeons. “When people have liver damage because of a drug, they either die or get transplanted. We’re hoping this will be another signaling system for important hepatotoxins,” said Temple. Aside from the liver transplant registry and some pregnancy outcome registries, these sorts of databases are not easy to come by. Nonetheless, they offer promise as a source of postmarketing safety data. “The problem with spontaneous reports is that you don’t get a very good idea of what the rate of the event is. It’s possible with these registries that we’ll be better able to get a quantitative estimate—to say, okay, this represents one-eighth of all events of this kind that happen in the United States,” Temple said. CERTs research programs Under principal investigator Raymond Woosley, research at the Georgetown CERT will focus on drug interactions and adverse effects in women. “One of the things we’re beginning to realize is that many of the drugs being taken off the market are being withdrawn because of harm to women. Seldane, Hismanal, Posicor, Raxar, Propulsid—all had more than twofold greater harm in women,” commented Woosley, who chairs the pharmacology department at Georgetown. “There are another 40 drugs still out there that potentially have this same problem of causing arrhythmias in women, often due to drug interactions but maybe not.” The Georgetown CERT is tackling this issue by setting up a registry for drug-induced arrhythmias and maintains two related Web sites, www.qtdrugs.org and www.druginteractions.com. The Georgetown investigators are also investigating possible interactions between the breast cancer drug tamoxifen, which is metabolized by a P450 liver enzyme, and the antidepressant fluoxetine (Prozac), which inhibits that enzyme. This is another fall-through-the-cracks issue that has gone unstudied until now. “It’s never been looked at because tamoxifen is an old drug. There are all sorts of potential complications that would result from it, but they are all theoretical. It could make the drug better, or it could make it worse,” Woosley said. These various drug safety studies typify the type of research to be undertaken by the CERTs—“the kind of research clinicians and patients need that isn’t funded by industry,” as Eisenberg puts it. Funding remains a major issue. The Vanderbilt and Georgetown centers have both obtained parallel grants from the NIH, and the Vanderbilt CERT has some additional funding through the FDA. Woosley and Ray hesitate over the prospect of seeking industry funding. “I have always emphasized that these centers have to be independent,” said Woosley. “They cannot be beholden to someone else, or they will fail to serve the public interest.” Ray does not entirely dismiss the possibility of industry funding, noting that “if the right study comes along that meets the interests of both the company and ourselves, then we would consider it.” Rather than seeking funding from multiple private and public sources, Ray hopes for legislation that would establish regular funding by levying a fee upon the pharmaceutical industry, perhaps comparable to the user fees for New Drug Applications. On the face of it, paying a tax to fund studies investigating possible adverse reactions might not seem like an attractive proposal to pharmaceutical company executives. Ray argues, however, that better postmarketing safety surveillance and provider education would benefit industry. Systematic study of the fen–phen combination, he says, might have identified heart valve damage sooner, before so many people were exposed and so many lawsuits filed against American Home Products. Systematic mechanisms for monitoring provider behavior and communicating with providers about contraindications for cisapride and the need for liver function tests with troglitazone therapy might have allowed those two drugs to be used safely and to stay on the market, Ray said. As Peter Honig points out, the ‘E’ in CERT stands for education. The centers also have programs focused on communicating with healthcare providers and pharmacists to promote rational therapeutics and improve awareness of known drug toxicity and drug interactions. “One of the real problems today is we’re having to take safe drugs off the market because they are misused. We can help the drug industry and society if we can make sure these drugs are used better and more appropriately. Then everybody wins,” said Woosley. According to Ray, effective mechanisms for provider education do exist. “The challenge is how to integrate that into the daily operation of the health care system.” Meeting that challenge—and doing the studies to detect or confirm suspected safety issues—will take money. “It could easily jump 100-fold” over the current CERT funding, Ray says. For now, the CERTs must address a very big need with a relatively modest supply of funds. References
Carol Hart is a science writer based in Narberth, PA. Comments and questions for the author may be addressed to the Editorial Office by e-mail at mdd@acs.org, by fax at 202-776-8166 or by post at 1155 16th Street, NW; Washington, DC 20036. Top || Modern Drug Discovery Home Page
|
The Centers for Education and Research in Therapeutics have small budgets, but big goals.