Are We Running Out of Antibiotics?

Hardly any doctors still practicing can remember life before antibiotics, when people were routinely hospitalized for common infections, and the threat of deadly Staphylococcus shadowed even the simplest surgery. But infectious-disease specialists like Brad Spellberg of UCLA's David Geffen School of Medicine have been reading up on those days because of a growing fear they are not all in the past. Wealthy countries take for granted the triumph of science over bacteria, but increasingly doctors are coming up against infections that can be quelled only by the most powerful antibiotics known to medicine—or by none of them. "It's already happening," says Spellberg, to the tune of roughly 100,000 deaths a year from antibiotic-resistant infections in the United States alone. "But it's going to become much more common." Imagine a world in which antibiotics resemble chemotherapy drugs—producing toxic side effects and unpredictable outcomes instead of the guaranteed cures we have come to expect—and you can understand what keeps Spellberg awake at night.

In the future, historians of science may debate whether victory over bacteria was ever within our grasp. But it seems almost certain that the 60 or so years after penicillin came to market will eventually be viewed as just an interlude in the eternal war between us and them. We are multicelled animals of astonishing complexity and delicacy, moving through a world in which they vastly outnumber us. They are single-celled organisms so primitive they lack even a nucleus, marvelously adapted to multiply inside us—under the right circumstances, to consume our flesh and poison us with their waste. For a few decades we gained the upper hand through the use of antibiotics, natural substances that are as toxic to germs as germs are to us. But our ingenuity is in a desperate race against their ability to reproduce. More and more strains of bacteria are developing biological countermeasures to antibiotics—cell membranes that won't let them in, tiny pumps that push them back out, biochemical tweaks that make them harmless. Evolution is a process that has been at work on earth for hundreds of millions of years; modern biological science has been around for less than a century and a half. Which would you bet on?

We have handicapped ourselves in this race, partly through carelessness and partly as an outcome of the complex politics and economics of drug policy. We've squandered our antibiotics through overuse—in animal feed, or on diseases they can't cure, such as influenza—and, paradoxically, by under-use. Particularly among the poor and illiterate, it's not uncommon for patients to stop taking antibiotics as soon as they feel better—leaving behind a residual population of resistant bacteria to multiply and spread.

For a while this didn't matter, because there were always new antibiotics being discovered. Beginning in the 1940s, when penicillin first hit pharmacy shelves, humanity embarked on a decades-long quest to collect as many soil samples as possible and probe them for potential miracle cures. Allied soldiers scooped up dirt from the African front, the National Geographic Society collected samples from the top of the Himalayas, and schoolchildren everywhere dug up shovelfuls from parks and fields. Pharmaceutical companies led the charge in harvesting this vast collection, ultimately producing some 200 new drugs in a mere three decades.

But by the mid-1980s, the discoveries had slowed to a trickle. "It's a bit like oil," says Spellberg, author of the book Rising Plague. "There's still a lot out there, but we've harvested all the easy gets. There are very few places left where you can just tap the ground and find some bountiful reserve." For a time, scientists thought the molecular wizardry of modern drug design—small molecules, high throughput screening, and the like—might reinvigorate our war against bacteria, the same way it reinvigorated our war against cancer. No such luck. "There are a set of rules that chemists follow when looking for new drugs," says AstraZeneca medical director John Rex. "To make an antibiotic, you have to break those rules. They are different from anything else we make because they are designed to kill a living organism inside another living organism."

That is not to say that there are no more wonder drugs left to find. Indeed, oceanographers have turned up hundreds of thousands of previously unknown organisms in a single cubic centimeter of deep-sea mud. Bacteriophages—viruses that infect and kill bacteria—remain a vast untapped source of potential anti-microbial medications, and the same soils that gave us some of the first and most successful antibiotics are just waiting to be reexamined with newer technology.

But the researchers best equipped to do that examining have long since abandoned antibiotics. According to the Infectious Disease Society of America, only five of the 13 biggest pharmaceutical companies are still looking at all. Medications that treat cancer or chronic conditions like diabetes, heart disease, or even baldness are simply much more lucrative. That's because a single course of the most expensive antibiotics (linezolid and daptomycin) runs just seven days and sells for $1,000 to $2,000, while a single course of, say, any given cancer treatment runs for weeks to months and costs from 10 to 20 times as much. Chronic-disease medications, which a patient typically takes for the rest of his or her life, yield even higher returns. It doesn't matter that bacterial infections are infinitely more common than cancer or even heart disease; antibiotics are still less profitable, in part because we undervalue them. "We expect antibiotics to be not only safe and highly effective, but cheap too," says Princeton University health economist Ramanan Laxminarayan. "We see them as like a God-given right." Case in point: in 2009 several large chain pharmacies began giving away free generic antibiotics in an effort to lure more customers.

Even when an antibiotic is highly prized, the twin forces of drug resistance and patent law still conspire against it in the marketplace. To prevent resistance from developing, doctors are encouraged to conserve the newest, most powerful antibiotics, using them only in the direst of circumstances. While this preserves the drug's value as a treatment option, it also means that sales will be slowest when a compound is new, and will pick up only toward the end of its patent life, when it can easily be supplanted by cheaper generic brands. "It's a huge disincentive to develop anything new," says Rex, whose company is one of the few still developing new antibiotics.

Of course, those economic conundrums pale in comparison to the unique regulatory uncertainty that antibiotics face. To test any compound against any infection, one first needs to find a group of patients. This is easy enough for cancer or chronic diseases, or any fatal condition where prospective study participants have some time to consider their options. It's a bit trickier with infectious diseases, especially those that are drug-resistant. "You pretty much need to wait for an outbreak to occur," Rex says. "And there's no way of knowing when or where that will happen." To test a compound against vancomycin-resistant Enterococcus (VRE), a super-bug that infects the digestive system and urinary tract, one pharmaceutical company opened 54 different testing sites for two full years. Only three patients enrolled in the study. A second attempt secured just 45 patients in 18 months, thanks largely to an outbreak at one testing site. Both studies were closed due to insufficient enrollment. Five years later, the drug remains stuck in clinical-trial limbo. That is not to say that VRE isn't a significant problem. According to the Centers for Disease Control and Prevention, this particular super-bug infects some 26,000 hospital patients each year. But so far, no one has a good way to match those patients to trials.

When drugmakers do manage to cobble together a decent-size group of study participants, they need to test their prospective drug against a proven medication. If the experimental drug is as good as or better than the existing treatment, it gets a green light. Sounds simple, but it isn't. The problem is that many antibiotics were tested and approved before there was any such thing as the FDA or double-blind placebo-controlled clinical trials. And that has left clinicians and statisticians at odds over how to measure success. "The clinical evidence that has guided antibiotic therapy for decades means nothing to the statisticians," says Paul Ambrose, president of the Institute for Clinical Pharmacodynamics in New York. "This is part of a larger FDA-wide crisis over how to measure drug effectiveness. But antibiotics have it the worst because they have no placebo-controlled data to begin with."

For just one example of how sticky this can get, take skin infections. Last year the FDA issued guidelines for companies seeking to develop new antibiotics to treat this condition. Its definition of a successful drug? One that stopped skin lesions from spreading, even if it didn't make the lesions disappear. "So if you come to me with skin lesions, and I give you a drug, and in three days the lesions are exactly the same, that counts as a success," Ambrose explains. "How did they come up with this?" It turns out that the FDA based its proposed rules on just two studies, both published in 1937, which compared an antibiotic to UV-lamp therapy. The UV therapy was so ineffective—and skin infections were so fatal at the time—that investigators were thrilled just to have patients not get worse. So they counted that as a success.

The FDA has spent the past year working with would-be antibiotic manufacturers to develop a clear path around such issues, but critics say that in that time, the level of uncertainty has only increased. "There are viable solutions, but we need fresh voices," Ambrose says. "It's one or two statisticians, and a handful of clinicians arguing back and forth and back again over the same few points." In the meantime, without clear guidance on how to get new antibiotics approved, pharmaceutical companies aren't going to invest the money to find and develop them.

Neither, it seems, is anyone else. In 2009 the National Institute of Allergy and Infectious Diseases spent $94 million on research into treatments for potential, albeit exceedingly rare, bioterrorism agents like anthrax and plague, and only $16 million on developing new medications to treat drug-resistant pathogens. Likewise, Project Bioshield, enacted after the 2001 anthrax attacks, spurred volumes of research on the most worrisome bioterrorism agents. By promising to purchase any successful products, the federal government ensured that drugs to treat anthrax, plague, and Ebola would be profitable, despite the rarity of those conditions. But that legislation included no such guarantee for antibiotics to treat much more common drug-resistant infections. An analogous plan to combat antimicrobial resistance was launched around the same time, but critics say that it was grossly underfunded and thus never quite put into action.

Policymakers are, at last, working to change this. One bill floating through Congress, the Generating Antibiotic Incentives Now Act, would extend the patents on certain types of antibiotics by five years to protect them from generic competition. Other proposals include tax breaks, liability protection, and the same guaranteed government purchases that have been so successful in spurring bioterrorism research.

Of course, as critics point out, market-based incentives are no substitute for more judicious use of existing antibiotics. "Its like saying, 'We're running out of oil because we are doing a terrible job at conserving it, and the only solution is to find more oil,'?" says Laxminarayan. "If we just keep developing new drugs, and then overusing them, we aren't going to solve anything." The trick, he says, is to start talking about antibiotics in the same terms that we use to discuss other nonrenewable natural resources—namely, conservation and sustainable use.

That would certainly mark a paradigm shift. The CDC estimates that as much as 50 percent of all antibiotic use is unnecessary. Doctors routinely prescribe antibiotics as a precautionary measure, to ward off infections that have not yet occurred, or to appease patients worried about falling ill while on the road. And for patients who are infected, the approach is hardly more sophisticated. Most diagnostic methods pre-date penicillin, which is to say they are slow, cumbersome, and unreliable. This means doctors often have no way of knowing which bug a patient is infected with, or which drugs that bug might be susceptible to. Rather than waste time and money trying to answer those basic questions, they often prescribe a couple of antibiotics at once and then just cross their fingers. This approach is cheaper, and in the short term it works, but it also contributes to overuse and thus allows resistance to grow and spread.

The future will almost certainly be different. Antibiotics will cost more and do less. They will also be less readily available than we are accustomed to. No more fast cures, free giveaways, or "just in case" prescriptions. But if we act quickly—and if we're lucky—it's still possible that we won't have to know a world without them.