By Rick Weiss
Washington Post Staff Writer
Sunday, March 4, 2007
The government is on track to approve a new antibiotic to treat a pneumonia-like disease in cattle, despite warnings from health groups and a majority of the agency's own expert advisers that the decision will be dangerous for people.
The drug, called cefquinome, belongs to a class of highly potent antibiotics that are among medicine's last defenses against several serious human infections. No drug from that class has been approved in the United States for use in animals.
The American Medical Association and about a dozen other health groups warned the Food and Drug Administration that giving cefquinome to animals would probably speed the emergence of microbes resistant to that important class of antibiotics, as has happened with other drugs. Those super-microbes could then spread to people.
Echoing those concerns, the FDA's advisory board last fall voted to reject the request by InterVet Inc. of Millsboro, Del., to market the drug for cattle.
Yet by all indications, the FDA will approve cefquinome this spring. That outcome is all but required, officials said, by a recently implemented "guidance document" that codifies how to weigh the threats to human health posed by proposed new animal drugs.
The wording of "Guidance for Industry #152" was crafted within the FDA after a long struggle. In the end, the agency adopted language that, for drugs like cefquinome, is more deferential to pharmaceutical companies than is recommended by the World Health Organization.
Cefquinome's seemingly inexorable march to market shows how a few words in an obscure regulatory document can sway the government's approach to protecting public health.
Industry representatives say they trust Guidance #152's calculation that cefquinome should be approved. "There is reasonable certainty of no harm to public health," Carl Johnson, InterVet's director of product development, told the FDA last fall.
Others say Guidance #152 makes it too difficult for the FDA to say no to some drugs.
"The industry says that 'until you show us a direct link to human mortality from the use of these drugs in animals, we don't think you should preclude their use,' " said Edward Belongia, an epidemiologist at the Marshfield Clinic Research Foundation in Wisconsin. "But do we really want to drive more resistance genes into the human population? It's easy to open the barn door, but it's hard to close the door once it's open."
The FDA knows how hard it can be to close that door. In the mid-1990s, overriding the objections of public health experts from the Centers for Disease Control and Prevention (CDC), the drug agency approved the marketing of two drugs, Baytril and SaraFlox, for use in poultry. Both are fluoroquinolones, a class of drugs important for their ability to fight the bioterror bacterium that causes anthrax and a food-borne bacterium called campylobacter, which causes a serious diarrheal disease in people.
Before long, doctors began finding fluoroquinolone-resistant strains of campylobacter in patients hospitalized with severe diarrhea. When studies showed a link to poultry, the FDA sought a ban. But while Abbott Laboratories, which made SaraFlox, pulled its product, Baytril's manufacturer, Bayer Corp., pushed back.
"They fought this tooth and nail. It took years," said Kirk Smith, an epidemiologist at the Minnesota Department of Health.
Finally, late in 2005, Bayer gave up, but not before fluoroquinolone resistance had spread even further.A Question of Resistance
Microbes are constantly mutating, and some of those mutations happen to confer immunity to one drug or another. Exacerbating the problem, bacteria constantly exchange bits of DNA with each other, spreading that resistance.
Given those realities, experts agree that all antibiotics should be used judiciously.
"If a drug is used less, then less resistance emerges," said Patricia Griffin, chief of intestinal disease epidemiology for the CDC.
Prudence is especially important for medicines of last resort, which is why the cefquinome application stirred such a storm.
Cefquinome is a fourth-generation cephalosporin, the most recent of several steadily improving versions of the cephalosporin family of antibiotics. Only one medicine from that family has been approved in the United States -- a powerful human drug called cefepime (brand name Maxipime), which is the only effective treatment for serious infections in cancer patients and a reliable lifesaver against several other nearly invincible infections.
InterVet developed cefquinome to treat bovine respiratory disease, the most common disease in cattle. Recognizing the potential public health implications of using a close cousin of cefepime in animals, the FDA's Center for Veterinary Medicine, which oversees animal drug approvals, convened its expert advisers in September.
One of the first things the group learned was that more than a dozen medicines are already on the market for the respiratory syndrome, and all are still effective.
"If we have no susceptibility problem, why do we need one more new drug?" asked James E. Leggett Jr., a professor of medicine at Oregon Health & Science University, whom the FDA brought in as a consultant on the cefquinome question.
The panel also learned that the disease would be a relatively minor issue but for the stressful conditions under which U.S. cattle are raised, including high-density living spaces and routine shipment on crowded trains for hundreds or thousands of miles. Those "production dynamics" suppress the animals' immune systems, explained feedlot consultant Kelly Lechtenberg of Oakland, Neb., and virtually guarantee that bovine respiratory disease will be a major problem.
Yet Stephen Sundlof, head of the FDA's Veterinary Medicine Center, told the panel members that under agency rules they should ignore those issues and consider only the language in Guidance #152.Flaws Seen in Rules
Guidance #152 is essentially a checklist of points to consider when weighing the potential human impact of a new animal drug.
After the Baytril debacle, the public health community embraced the idea of a guidance document. A formalized risk-assessment process promised to minimize the chances of making a bad regulatory call.
But a struggle ensued when the FDA hosted meetings to spell out the criteria to be used for measuring risk, often with veterinarians and veterinary drug companies on one side and doctors and public health experts on the other.
When differences could not be resolved after repeated drafts and months of work, the agency sidestepped some tough issues and adopted language that both sides agree can block approval of the most worrisome drugs -- those such as Baytril that are put in animal feed or water, and so are easily overused. But public health experts say the wording tilts the playing field toward industry for other kinds of drugs. They want to see it revised.
Most glaring, they say, is that the guidance makes it almost impossible to say no to a new animal drug unless it is likely to threaten the effectiveness of an antibiotic that is a critical player against food-borne illnesses. By contrast, the World Health Organization recommends saying no if approval would spur resistance to any antibiotic that is important for fighting "serious human disease" -- not just food-borne illnesses.
Cefquinome's primary threat is that it may undermine the usefulness of the closely related human drug, cefepime. But as it turns out, the FDA does not consider cefepime a front-line drug against food-borne infections. So although it is a highly important drug in human medicine generally -- and although the Infectious Diseases Society of America even recommends it against some food-borne bacteria -- that risk does not count under the terms of Guidance #152.
A related problem is that the guidance's definition of "food-borne" is conservative, said Margaret Mellon of the Union of Concerned Scientists, a science policy advocacy group. For example, most urinary tract infections are caused by intestinal bacteria acquired from food, and cefepime is prescribed for those infections. If the FDA counted those infections as food-borne, then the guidance's formula would call for rejecting cefquinome for cattle.
"But FDA didn't do that," Mellon said. "That restricted the analysis right there."
Moreover, the guidance does not take into account that when microbes become resistant to fourth-generation cephalosporins, they often gain resistance to third-generation versions, too.
Third-generation cephalosporins are among the only effective therapies for serious gastrointestinal diseases in children and are the sole therapies for many cases of meningitis. That means the emergence of resistance to fourth-generation cephalosporins "could have a much more far-reaching effect" than is considered under the terms of Guidance #152, John H. Powers, a medical officer at the FDA's Center for Drug Evaluation and Research, told the agency's panel of experts.How Great a Risk?
Richard Carnevale, vice president for scientific and regulatory affairs at the Animal Health Institute, which represents veterinary drugmakers, said critics should not presume that a dozen drugs effective against bovine respiratory syndrome are enough.
"It's not a question of whether there is a need or not," Carnevale said. "The answer is, there's always a need."
The institute contends that the risk to human health posed by animal antibiotics has been overblown.
Officials at InterVet declined several requests to be interviewed. In a statement, the company said it "fully supports the prudent use of antibiotics in animals."
The statement also said that in Europe, fourth-generation cephalosporins similar to cefquinome have been used in animals for the past decade "without compromising the interests of public health."
Yet recent European data indicate that resistance against this class of antibiotics is on the rise.
An analysis of E. coli bacteria in pigs and other animals in Spain, published in December, found high levels of the resistance that renders fourth-generation cephalosporins useless. A January report from Britain documented similar resistance patterns emerging at 10 farms.
Microbes resistant to fourth-generation cephalosporins have also begun to pop up in European patients. Such resistance is virtually unknown in the United States, where fourth-generation cefepime has been used in patients since 1997. That suggests that the resistance emerging in Europe is a result of veterinary use, said Steve Roach of the Food Animal Concerns Trust, a Chicago public interest group.
Roach says he is concerned that history is about to repeat itself. U.S. cattle were free of bacteria resistant to third-generation cephalosporins in 1997, but by 2003 one of every five samples was resistant. "This is exactly what should be avoided with cefquinome," he said.Merely Suggestions
At the FDA advisory meeting in September, the agency's experts defied Guidance #152 and voted 6 to 4 against approval of cefquinome. But that day, and in follow-up interviews, Sundlof, the agency's veterinary chief, made it plain that the vote was "not binding."
"I think we all agreed . . . that Guidance for Industry #152 would be the criteria against which we would base our decisions on safety," Sundlof said at the meeting.
Concerned that the FDA is poised to approve cefquinome, Congress's only microbiologist recently wrote to the agency.
"Given the recent outbreaks of E. coli and other food borne illnesses across the nation, it is hardly the time to ignore the advice of scientists, and potentially impair our ability to treat deadly infections," wrote Rep. Louise M. Slaughter (D-N.Y.), who chairs the House Rules Committee.
Yet, being realists, the FDA advisers in September said more than just no. They told the FDA that if it approves cefquinome, then it should at least impose limits to minimize the potential consequences. One suggestion was to explicitly preclude "off label" use of the drug -- that is, to tell veterinarians that it can be used only for bovine respiratory disease.
But Sundlof said that, under FDA rules, those decisions must be left up to veterinarians unless there is clear evidence that wider use is causing harm.
"We have to take a fairly legal interpretation," Sundlof said in an interview. "If we have no evidence of a problem, or sparse evidence, we would not be able to make the prohibition prior to approval."
However, raising a point that opponents do not find very reassuring, he noted: "As soon as we have the first glimpse of evidence that off-label use of a drug is causing resistance, we have the authority to prohibit off-label use."
The advisers also urged that as a condition of approval, the FDA should demand that InterVet provide annual reports on how much cefquinome was used and in which animals -- data that would help scientists detect links between the drug's use and patterns of resistance that emerge in people.
"Without reliable, meaningful data on the quantity of use, the purpose of use, the type, number and location of animals treated, it will be exceedingly difficult to interpret fluctuations in rates of resistance," said Susan Prolman of the Union of Concerned Scientists.
But Sundlof offered little hope for that outcome.
"That is information that would be useful to have," he said. But the agency does not have the authority to demand it.