Seeking the Path of Least Resistance

Rising antibiotic resistance is not just a medical crisis—it’s a scientific, economic, political, and moral problem. Can we act before it’s too late?

Story by Madeline Drexler
Photography by Kent Dayton
Spring 2019

In 1900, infectious disease was the leading cause of death in the United States and around the world. Staph in the bloodstream killed 90 percent of its victims. A man who nicked himself shaving could die from erysipelas, a strep infection. Children lost their playmates to scarlet fever, meningitis, osteomyelitis. Bacterial pneumonia, the leading cause of death, killed a third of its victims. Tuberculosis patients were advised to rest and seek clean air, because there was nothing medicine could offer. Gonorrhea had no cure.

In the mid-20th century, when antibiotics, also known as antimicrobials, hit the market, they had an astonishing impact. Previously fatal infections no longer struck terror. In the decades since penicillin made its debut, nearly 150 antibiotics in 31 classes have saved millions of lives and made possible the everyday interventions we take for granted, from open-heart surgery and cancer chemotherapy to hip replacements and cesarean sections. In 2016, the most recent year for which data has been analyzed, the leading infectious cause of death worldwide—lower-respiratory infections—ranked fourth on the list of global killers, behind ischemic heart disease, stroke, and chronic obstructive pulmonary disease. (Low-income nations, however, still saw respiratory infections and diarrheal diseases as their top two threats.) The World Health Organization (WHO) estimates that antibiotics add, on average, 20 years to human life expectancy.

Five years after Fleming received the Nobel Prize, the bacterial strains against which he first demonstrated the miraculous efficacy of penicillin—Staphylococcus aureus—were the first microbes to show widespread resistance to the medication. Today, nearly all the bacteria that afflict humankind have become, to varying degrees, resistant to the wonder drugs developed to stop them, imperiling the foundations of modern medicine. What were easily cured infections must now be treated with repeated rounds of powerful and sometimes toxic drugs, forcing a choice between safe medicines that might not work and dangerous ones that do.

Sometimes, nothing works. “There isn’t an ID doc in this country who hasn’t been at the bedside of a patient who’s desperately ill and for whom we did not have an antibiotic,” says Cynthia Sears, president of the Infectious Diseases Society of America and an attending physician at Johns Hopkins Hospital. “And it’s not always someone who’s older. The case I always harken back to was a woman in her mid-20s who had an intra-abdominal abscess. Things didn’t go well after surgery. And I literally had no drugs to offer her.”

“It’s not that most infections are untreatable—it’s that some infections are untreatable,” says Marc Lipsitch, professor of epidemiology and director of the Center for Communicable Disease Dynamics at the Harvard T.H. Chan School of Public Health. “But there are a lot of infections that are close to the precipice. Just barely, we seem to manage to pull out another class of antibiotics or a variant on an existing class. The concern is that new classes of drugs are slowing down, but resistance keeps growing.”

In a 2016 report, the Review on Antimicrobial Resistance, a comprehensive project supported by the United Kingdom and the Wellcome Trust, estimated that by 2050, drug resistance (the vast majority of which is bacterial) would claim 10 million lives a year—one person every three seconds. Even today, the 2016 report noted, 700,000 people succumb to resistant infections each year.

The report predicted that drug resistance would wipe out a cumulative $100 trillion of economic output by 2050. World Bank simulations published in 2017 found that as a result of shocks to the labor supply and livestock productivity, antimicrobial resistance during the same period would reduce annual global gross domestic product by 1.1 percent to 3.8 percent, compared with no effects from drug resistance—comparable to the losses inflicted by the 2007–2009 global financial meltdown at their most severe.

The federal Centers for Disease Control and Prevention (CDC) estimates that 2 million people annually are infected in the U.S. with antibiotic-resistant bacteria, and 23,000 die. The figures come from a 2013 study that drew on data from federal surveillance systems and hospital surveys, and CDC officials concede that the estimates are conservative.

Possibly, very conservative. In a letter published in November 2018 in Infection Control and Hospital Epidemiology, researchers at the Washington University School of Medicine arrived at a far higher number. Looking at the nearly 2.5 million inpatient and outpatient deaths in the country in 2010, they drew on three sets of data: estimates of deaths caused by sepsis (the body’s overwhelming reaction to infection, usually but not always among those with weakened immune systems); reported rates of multidrug resistance in American hospitals; and estimates of outpatient deaths caused by infections. They then calculated the number of patients who probably died from drug-resistant infections.

The low end of their estimate—153,113 inpatient and outpatient deaths a year from such infections—is nearly seven times higher than the CDC’s official figure. The upper end of 162,044 deaths would have made multidrug-resistant infections the third-leading cause of death that year in the U.S.—eclipsing Alzheimer’s disease, diabetes, flu and pneumonia, and suicide. Later this year, based on analysis of electronic medical records, the CDC will be coming out with a new, and assuredly higher, estimate of the toll of drug-resistant pathogens.

Complicating these calculations is the fact that hospitals and clinics don’t always report antibiotic-resistant infections; in many states, they are not legally compelled to do so. There are also no ICD-10 codes—the alphanumeric shorthand used by physicians, health insurers, and public health agencies—that specifically denote diagnoses of multidrug-resistant bacterial infections. “When a patient dies, even the families aren’t always aware that a drug-resistant infection was the ultimate cause,” says Kathy Talkington, who directs the Pew Charitable Trusts Antibiotic Resistance Project.

Death certificates further shroud the truth. One of the most damning indictments of end-of-life recordkeeping was written by Australian research microbiologist Christine Carson and published last December on the website of the Longitude Prize, a U.K.-based competition for innovative techniques to reduce drug resistance.

“In the reductionist realm of identifying and documenting causes of death, in which causation is reduced to a clutch of words on a certificate recorded for all posterity, the extinguishment of life due to the ravages of antibiotic-resistant bacteria doesn’t rate, doesn’t exist. It is, if you will, not a valid cause of death,” Carson wrote. “Instead, the focus is on the events that occur as a consequence of unmitigated bacterial infection—such as cardiovascular events or multiple irreversible organ failure—which culminate in death. This doesn’t mean death as a result of infection with antimicrobial resistant bacteria isn’t happening. It just means it is happening in an administrative vacuum, we are not measuring it, and we don’t know the scale of the problem, let alone how to address it.”

“An uncontrolled microbiological experiment conducted on an unprecedented scale”

>Antibiotics, first dispensed freely in the 1940s, were the world’s first “blockbuster” drugs. The term comes from military lingo, referring to powerful bombs. Antibiotics hav caused their own massive collateral damage. As a review article last year in Science put it, “What has followed in the subsequent 70 years or so has been an uncontrolled microbiological experiment conducted on an unprecedented scale.”

Bacteria can become resistant to single antibiotics or, in cases of multidrug resistance, to more than one class of drugs. In the extreme scenario of pan-resistance, bacteria foil all clinically available antibiotics, with no therapeutic alternatives.

Part of the problem is what scientists refer to as “bystander effect.” While resistance in targeted organisms was observed from the start, the collateral resistance among other organisms composing the human microbiome has been an unwelcome side effect of widespread antibiotic use. As Lipsitch explains, “If you treat a urinary tract infection [UTI], the normal staph and pneumococcal bacteria in your nose also get exposed to the antibiotic. All the organisms that are common parts of our normal flora—E. coli, Haemophilus influenzae, Strep pneumoniae, Klebsiella species, and many others—become drug-resistant. And the next time those bacteria cause an infection anywhere in the body, they will be resistant to the drug that was used to treat the UTI. And those resistant bacteria can be transmitted to other people.”

The elegant Latin and Greek names of bacterial organisms belie their viciousness. Many bacteria have become impervious to carbapenems and third-generation cephalosporins—the best available treatments for multidrug resistance. In scientific terms, these organisms are known as “Gram-negative,” a classification based on the color they turn after chemical staining; Gram-negative bacteria stain red or pink, Gram-positives, purple or violet, differences that arise in part from the structure of their cell walls. Gram-positive bacteria do not have an outer cell membrane, while Gram-negative bacteria do—a physical barrier that enables them to keep antibiotics out. “Gram-negatives are the biggest threat in terms of closest to the cliff, where they are resistant to everything and we have nothing to treat them,” says Lipsitch.

Antibiotics that treat only Gram-positive or Gram-negative bacteria are known as narrow-spectrum. Those that treat both Gram-positive and Gram-negative infections are broad-spectrum—and are deployed when a doctor suspects that a patient has a bacterial infection but does not know the specific organism. Broad-spectrum drugs cause more damage to the normal, healthy, “innocent bystander” bacteria in and on our bodies, and exert more of the indiscriminate selective pressure that leads to resistance.

Many of the most devastating bacterial infections occur in health care facilities—not only because that is where patients with weak immune systems are crowded together, but also because of intensive antibiotic use. For many resistant bugs, hospitals become focal points for spread, with pathogens jumping from patient to patient, often on the gloves of health care workers or via tubes or invasive medical devices.

But outpatient care may turn out to be an even bigger factory of resistance genes. In 2016, the Pew Charitable Trusts published a study of 154 million annual outpatient visits at which a prescription was given. In around 47 million, or just under a third, the drugs were unnecessary. The organization found that inappropriate antibiotic prescribing was rampant in the cases of sinus infections, middle-ear infections, and viral upper-respiratory infections such as the common cold.

Why so much misguided prescribing? There are many reasons: Patients demand the drugs; doctors don’t have enough time to explain why the drugs are not needed; physicians suffer decision fatigue as their workdays wear on; and doctors may be uncertain about diagnoses because patients with viral and bacterial infections often display similar symptoms, such as congestion, cough, and sore throat.

Far more prescriptions for antibiotics are written for outpatient care than in hospitals. And that may make those cases a vast ground zero for resistance. In a 2018 article in eLife, the Harvard Chan School’s Marc Lipsitch and Yonatan Grad, the Melvin J. and Geraldine L. Glimcher Assistant Professor of Immunology and Infectious Diseases, examined 72 bug-drug combinations among 99.8 million outpatient pharmacy prescriptions for 62.4 million people in the U.S. from 2011 to 2014. They wanted to figure out how much each prescription affected antibiotic resistance at the population level.

“The answer was that, for most of the bacteria-antibiotic pairs, it was the occasional, one-time use that correlated better with population-level resistance than the intense, repeated use,” says Grad. The takeaway, he says, is that “while reducing inappropriate use is one of the most important interventions in preventing drug resistance, if the use-resistance association is causal, then we should focus on low-intensity use.”

“One should not try to second-guess microbes!”

>Bacteria rule the earth.

They have been around for nearly 4 billion years. They number around 1030: a thousand billion billion billion—more than the stars in the known universe. Most multiply every 20 minutes or so. As a report from the 2013 World Healthcare-Associated Infections Forum put it: “The infinitesimal generation time of a microbe will always confer it the advantage: It has infinitely more opportunities to gain resistance genes than we have to create new antimicrobials. In this race, humans are being outrun.”

Bacteria’s astounding variety and ingenuity often evoke a kind of perverse admiration from scientists. “I’m not in favor of infections. But I do love bacteria,” says Eric Rubin, chair of the Harvard Chan School’s Department of Immunology and Infectious Diseases. “Their evolution over the course of eons has enabled them to solve the same problems—including resisting antibiotics—but in many different ways.”

“Bacteria are a superkingdom of life,” adds Bill Hanage, associate professor of epidemiology at the Harvard Chan School. “Think about that. There is more variation in bacteria than there is in quite literally everything that you can see with your naked eye. If you look at the tree of life, there’s this huge expanse which is full of bacteria. Then there’s all the sorts of life forms which are like us—eukaryotes, which have cells that contain a nucleus within a membrane. Right at the tip of the tree where we find the eukaryotes is this tiny little thing called Homo sapiens. Their closest relative, just an inch away, is corn. Everything else—everything else—is microbes.”

Bacteria’s innumerable tricks include the ability to spurn antibiotics. One way they accomplish this is through genetic mutation. The more replications a cell undergoes, the higher the chances it will mutate. Bacteria are prokaryotic organisms—they lack a nucleus protecting the genome, which means that their genetic material is free-floating and itinerant.

Another way that bacteria gain the weaponry to resist antibiotics is by acquiring genes from elsewhere. Small necklaces of DNA called plasmids exist separate from the chromosome and are highly mobile, transferred even between bacteria of different species—a phenomenon known as horizontal gene transfer. Drug resistance, it turns out, can exponentially spread this way.

Antibiotic resistance is Darwinian evolution in action. Antibiotics, and especially their improper use, create selective pressure on bacterial colonies. The organisms most sensitive to the drugs die quickly, while the most resistant organisms survive and replicate. If even one bacterium becomes resistant to the antibiotic, it can multiply and replace all the bacteria that were killed off.

But while resistance is part and parcel of natural selection, “This is not a natural process, but a man-made situation superimposed on nature,” wrote the eminent microbiologist Julian Davies. “The moral of the story,” he added, is that “one should not try to second-guess microbes! If resistance is biochemically possible, it will occur.”

Most antibiotics work in one of three ways. They break down the bacterial cell wall or membrane. They prevent or slow the cell’s ability to synthesize proteins, which are used to build and repair the cell. And they halt the synthesis of the cell’s DNA (which contains its genetic material) or RNA (which directs protein synthesis).

With their canny genetic mutability, bacteria thwart these assaults through various stratagems. They crank out enzymes that directly inactivate the drug. They fashion pumps that remove the drug from the cell. They alter the structures of key proteins to prevent them from interacting with the antibiotic.

This evolutionarily entrenched tit for tat might seem to make drug resistance inevitable. But it isn’t. Lipsitch has a standard talk titled “Why Haven’t Antibiotic-Resistant Bacteria Taken Over the World?” Although the question hasn’t been studied well, one reason appears to be that there is a biological “fitness cost” for resistance. In the absence of antibiotics, drug-resistant bacteria are less likely to survive and cause disease, compared with their drug-susceptible counterparts.

The “Antibiotic Resistome” Complicates the Challenge

>Most antibiotics are derived from bacteria and fungi in the soil, making nature by far our most fecund apothecary. What is the role of these chemicals in the wild? “It’s classically talked about in terms of warfare, because that’s the easiest model to imagine: I’m going to kill you so I have more space,” says Sarah Fortune, the John LaPorte Given Professor of Immunology and Infectious Diseases. “But the emerging model is that antibiotics are used for communication. Bacteria exist in communities and they talk to each other. They use different molecules to communicate, depending on whether they live in a salt flat versus in a sea vent versus in your gut. They’re all making different things in different ways, and they need to clear out space for themselves. To do that, they send signals to other bacteria—exchanges that are sometimes mediated by small molecules: antibiotics. I might put a brake on your population, not because I want to obliterate you, but because the community itself is more metabolically stable if there’s a little bit more of me and a little bit less of you.”

Just as bacteria are ancient, so is the inborn process of antibiotic resistance. Resistance genes against even our newest magic bullets have been found in permafrost in the Canadian High North, in woolly mammoths, in the gut microbiome of Pre-Columbian Andean mummies—in countless creatures occupying pristine archaeological sites. In all these examples, selective pressure was hyperlocal and resistance genes formed naturally. “Until we started weaponizing antibiotics and using them in medicine, they were simply part of the environment—and so were resistance genes,” says Hanage.

The global collection of all resistance genes—in and on our bodies, in the clinic, in farm animals, in drug company effluent, and everywhere else—is known as the “antibiotic resistome.” In India, antibiotic concentrations in treated wastewater from pharmaceutical plants have exceeded concentrations one would find in the blood of patients taking the medication—turning the effluent into a prodigious reservoir of highly drug-resistant bacteria. So ubiquitous and borderless is the antibiotic resistome that if a person develops a urinary tract infection in the hospital, it is easy to determine what kind of bacteria are responsible (usually E. coli) but almost impossible to establish with certainty where those bacteria came from and when they developed resistance.

All of which makes tackling antibiotic resistance an almost insuperable challenge. “In resistance, the driver is life itself,” says Hanage. “It’s evolution, the same thing that gave birds their wings, fish their gills, and us our brains. In trying to curb resistance, we are literally presuming to manipulate or otherwise retard the most powerful creative process ever discovered.”

Fallout from the Farm

Globally, far more antibiotics are used in animals than in humans. In the U.S., more than 70 percent (by weight) of antibiotics that are medically important in humans are sold for use in farm animals—not only to treat infections but also to prevent them and to promote growth. Antibiotics applied in plant production, including as pesticides in orchards, are often identical or closely related to those consumed by humans. This spring, for example, farmers in Florida will be spraying orange trees with hundreds of thousands of pounds of two common human antibiotics— streptomycin and oxytetracycline—to halt the bacterial disease known as citrus greening.

Like humans, all animals carry bacteria in their intestines. And as in people, when antibiotics are ingested, they will kill susceptible bacteria and leave resistant organisms to survive and multiply. When food animals are slaughtered and processed, these resistant bacteria can contaminate meat or other animal products. The bacteria can also enter the environment through animal stool and spread to fruits and vegetables irrigated with contaminated water. People become exposed to the resistant bacteria from handling or eating raw or undercooked food from animals, or eating contaminated produce.

Although it is not yet possible to accurately measure what proportion of total resistance springs from antibiotic use on farms, the animal-human resistance link is not theoretical. Take colistin, a drug in the carbapenem class of antibiotics and one of the last-line therapies for Enterobacteriaceae infections caused by E. coli, Klebsiella, and Salmonella. Although colistin, which was introduced in 1959, can cause kidney failure or nerve damage, it is increasingly the only option left for these dangerous Gram-negative infections.

In some countries, colistin is also deployed extensively in livestock. In 2016, a plasmid with a gene known as mcr-1, which translates into resistance to colistin, was discovered in China. That was no surprise; Chinese livestock producers were using colistin heavily as a feed additive, and the mcr-1 gene was found in E. coli colonizing food animals and, to a lesser extent, humans through contaminated meat. Researchers fear that if a bacterium with genes that make it resistant to every other drug then picks up mcr-1, it will become pan-resistant.

The clock is ticking. Mcr-1 genes have already been reported in 47 countries across six continents. In 2016, a Pennsylvania woman was diagnosed with a colistin-resistant UTI—the first appearance of colistin resistance tied to a plasmid that jumps easily between organisms. Some researchers worry that we’ve breached the last line of defense against deadly carbapenem-resistant enterobacteria—or CRE—infections that the relatively old drug colistin has been taken off the shelves to treat. As Lance Price, professor at the George Washington University Milken Institute School of Public Health, described the prospect of mcr-1 horizontal gene transfer, “[T]he last card in the CRE royal flush is in play.”

The Drug Pipeline Runs Dry

>Between 1940 and 1980, researchers developed 20 new classes of antibiotics, each based on a different mechanism of action. But since then, the antibiotic pipeline has dried up. The last new major class of antibiotics was discovered in 1987. The last class of drugs targeting Gram-negative bacteria was discovered in 1962.

Most antibiotics are derived from natural compounds. About 75 percent of known antibiotics are produced by a group of thready soil bacteria called actinomycetes, and about 75 percent of these originate in a single genus, Streptomyces. Almost all of today’s 100 or so antibiotics have been chemically modified from nature’s original stock.

The environment’s vast reservoir of antibiotic-producing bacteria and fungi should theoretically keep the antibiotic pipeline flowing. Existing bacterial species in nature are estimated to number somewhere between 10 million and 1 billion. Yet about 99 percent of the microorganisms in the wild that compose a potential source of new antibiotics cannot be grown in a lab dish and therefore remain uncultured. These include bacteria found in oceans, tropical rain forests, and in the earth’s niches of extreme high temperature or pressure—remote sites that remain largely unfathomed for their clinical potential.

“The field of natural product chemistry has died off,” laments Eric Rubin. To address that, some scientists are trying to find new chemical structures elsewhere—such as in pharmaceutical company “libraries,” where compounds are catalogued, stored, and often screened for potential new drugs.

But such libraries may not be helpful in the search for new antibiotics. “The problem is that those libraries come from campaigns where the companies were trying to make anti-cancer drugs or anti-diabetes drugs or anti-hypertension drugs,” says Rubin. “And those compounds don’t look like antibiotics.”

That said, he remains almost defiantly optimistic that scientists will get around these practical obstacles. “There is no reason to have a sky-is-falling attitude. We’ve made better drugs for lots of diseases, and we can make better antibiotics.” Sarah Fortune shares that improbable confidence: “I don’t think antimicrobial discovery is such a huge biologic challenge that it’s not solvable.”

The magic ingredient isn’t scientific ingenuity. It’s money.

“A funny kind of economics”

>Research and development costs for a new antibiotic can exceed $1 billion. Shepherding a drug from the lab to the clinic can take 10 to 15 years. But for all those exertions, the profit is slim compared with new treatments for chronic conditions, such as diabetes, hypertension, psychiatric disorders, asthma, or even heartburn.

While most new and needed medications have a decent return on investment over the drug’s patented life, antibiotics are different. Both prices and sales volumes are low. And when a breakthrough drug does reach the market, it is the new last-line drug, prescribed only when first-line treatments fail. To preserve the new antibiotic’s effectiveness, it must be used sparingly. But patents expire after 10 to 15 years, so by the time the debut drug becomes the first- or second-line therapy—because bacteria have gained resistance against current front-line drugs—its developer has lost profits to generics.

Currently, there are 42 new antibiotics in the pipeline. Of those, only 11 have the potential to treat the WHO’s Gram-negative priority pathogens. Historical data suggest that about one out of five infectious-disease drugs that reach the initial phase of human testing will receive approval from the Food and Drug Administration (FDA)—meaning that perhaps only two of the 11 potential priority-pathogen fighters now in the pipeline will make it to market. By comparison, there are more than 1,000 cancer drugs in the pipeline.

While 25 large pharmaceutical companies fielded active antibiotic discovery programs as recently as 1980, today only three are still in the game. “Antibiotics is a dying field,” says Rubin. “Why? The major problem with infectious diseases is that they’re curable, and that is very bad for drug development. For drug companies, the money is being made in two areas. One is cancer, which, while it is curable, takes a long time to treat and does not have limits on what you can charge for a drug. The other is chronic diseases, where people take their medications forever.”

Rubin adds: “There’s a funny kind of economics at play. You can charge what you want for a cancer drug just because the cancer world and insurers and governments all accept that we’re willing to pay a lot of money for these drugs. That doesn’t apply to antibiotics—for no particularly good reason, except that traditionally antibiotics have been cheap.”

One of the rare resounding success stories of late in antibiotic discovery is the Johnson & Johnson (J&J) tuberculosis drug bedaquiline, which was approved by the FDA in 2012. Together with the drug delamanid, made by the Japanese company Otsuka Pharmaceutical and approved in 2014, it has transformed the lives of people infected with drug-resistant TB.

“We are in this amazing moment,” says Fortune. “Five years ago, drug-resistant TB was treatable only with complex cocktails of very toxic drugs—think toxic on the order of cancer chemotherapy, where people’s lives grind to a halt. But investment over the last 20 years—by the National Institutes of Health, the Gates Foundation, and pharma—has resulted in these two new agents, which have taken that 18-month, intravenous, highly toxic regimen down to a nine-month, all-oral, well-tolerated regimen.”

How did it happen? Time and money—and, in this case, the outsized and even obsessive commitment of one individual, a Belgian microbiologist named Koen Andries. He had begun his career researching viral diseases in animals. After reading a WHO report in 2001 that described the rising prevalence and growing drug resistance of tuberculosis, he switched his focus to human medicine and began searching for a cure.

Andries and his team pored over the “compound library” of his employer, Janssen Pharmaceutica, which is part of J&J. They tested thousands of chemical substances, repeatedly recombining and retesting them, until they homed in on the quinoline molecule that became the basis for their new medication. At times when it looked like J&J would kill the bedaquiline program because the drug was costly to synthesize, Andries personally kept it going, arguing that the drug would save lives, have a far bigger impact than the company’s other ongoing research, and would burnish the corporate image.

“This is not a unique story,” says Rubin. “If you look at the history of any new drug, it’s come about because of strong advocacy within the organization. But all credit to this guy. He was advocating for a drug that didn’t make money, and he pushed it over a lot of hurdles. His moral suasion was successful.”

“Push” and “Pull”: Incentives to Spur Drug Development

>Solving the broader problem of antibiotic resistance, however, can’t depend on a handful of dogged individuals. In a February 2019 letter to congressional leaders, the Pew Charitable Trusts, Infectious Diseases Society of America, and Trust for America’s Health, together with U.S. antibiotic developers large and small, called on Congress to move swiftly to enact a package of economic incentives to reinvigorate the stagnant pipeline of antibiotics. “Few major pharmaceutical companies remain engaged in antibiotic discovery and development, and small biotech firms, even those that have launched or are close to launching products, struggle to sustain a viable commercial enterprise,” the letter warned.

What might those incentives look like?

Economists talk about “push” and “pull” incentives in the drug market. Push incentives support basic research, as well as the expensive preclinical and clinical trial phases, through grant funding, public–private partnerships, and tax credits. Pull incentives support the approval process and post-marketing period, and include market entry rewards, prizes, and extended market exclusivity.

“The most difficult problem here is not science, it’s economics. Full stop,” says Kevin Outterson, director of the Social Innovation on Drug Resistance Program at Boston University, where he is a professor of law and executive director of the Combating Antibiotic Resistant Bacteria Biopharmaceutical Accelerator—CARB-X—a global partnership started in 2016 by the Biomedical Advanced Research and Development Authority (part of the U.S. Department of Health and Human Services) and the Wellcome Trust, and hosted at the Boston University School of Law.

CARB-X is an example of an enlightened push incentive, putting its money on innovative research to combat the most serious drug-resistant bacteria. The program focuses on projects in early development, often referred to as the “Valley of Death,” because it is in the early phases of research that many projects are abandoned for lack of funding and support. The idea for CARB-X came about in 2014 when U.K. Prime Minister David Cameron talked to U.S. President Barack Obama about the global health crisis caused by antibiotic resistance. The following year, the U.S. government released a national action plan that called for a biopharmaceutical accelerator that would advance research and development and look for higher-risk research on new antibiotics.

CARB-X became that engine of exploration. It offers 70 percent funding for the projects it chooses, provides technical, business, and scientific support, and in effect stamps its imprimatur on the highest-quality research for private funders and venture capitalists to consider investing in.

“We have met many scientific teams that have built companies on inventions discovered through National Institutes of Health funding or the equivalent from other countries. These small companies work on very tight budgets and they’re very efficient,” says Outterson. “If you look at oncology, you see incredibly high valuations chasing hypotheses without much data. In antibiotics, it’s the opposite: We have high-quality science with dirt-cheap valuations because of economic problems in the antibiotic market.”

By the end of this year, CARB-X will be supporting more than 55 projects—“and at least 15 of them could be a first-in-class drug against Gram-negative bacteria,” Outterson says. “As with any preclinical research, most of the drugs in our pipeline will fail. But we have many shots on goal, and if one of them makes it in, it would be the most dramatic achievement against Gram-negative bacteria in the last six decades.”

If CARB-X is a visionary push incentive, what would be a similarly visionary pull? In March, Lord Jim O’Neill, a former Goldman Sachs chief economist who chaired the U.K.’s influential Review on Antimicrobial Resistance, excoriated big pharma for dragging its feet on antibiotic drug discovery. “If pharmaceutical companies delivered just a tenth of the commitment that comes from their words, we might actually get somewhere,” he told the BBC. “It leads me to think that some of the more radical ways of changing the risk/reward incentive and social circumstances of it now need to be explored more.”

O’Neill floated the idea of nationalizing antibiotics production, perhaps through a taxpayer-supported utility company that would focus solely on the costly enterprise of drug manufacture and distribution. He had previously advocated a “pay or play” mechanism, in which an international body such as WHO would levy a small charge, perhaps 2 percent, on all drug sales by pharma companies that do not have antibiotic-development programs. The proceeds would help fund a market-entry prize of $1 billion to $1.5 billion for each novel antibiotic that meets a defined medical need.

Outterson offers what he calls a “less dramatic” alternative to nationalization: create a for-profit company for which the core investors are governments and charities, with the rest owned by the public. “The company’s purpose would always be locked to antibiotics. They won’t be allowed to go off and chase more lucrative classes like cancer. And as a result, the expectations on profit must be reasonable. Not 15 to 20 percent, like biotechs, but more like a public utility, with a 4 percent or 5 percent rate of return.”

For that to work, there must be a more profound change in the antibiotics marketplace. Unlike other drugs, these treatments must be paid for not based on sales volume but on their value to society. The concept is known as “delinkage,” and while visionaries like Outterson embrace it, “we don’t see consensus yet among the governments that write the checks,” he says.

Recasting Antibiotics as Public Goods

>With impending disaster and promising innovation, why isn’t the antibiotic-resistance crisis near the top of the political agenda?

Keiji Fukuda, a former high-ranking WHO official, now a clinical professor at the University of Hong Kong School of Public Health, asks the same question. When he spoke at a 2017 National Academy of Medicine workshop, he called for the “socialization” of the drug-resistance issue, urging that it be vaulted into a high-profile social concern—like cancer, tobacco control, or HIV/AIDS. He also urged that the problem be couched as personal, urgent, and potentially reversible with the right action. “The concrete challenge is how do you take something that in general is seen as abstract, technical, and distant—almost like science fiction—and … make that personal?” he asked. “How do you humanize it? How do you make it seem like something that has to be dealt with now?”

An idea from economics is relevant: the notion of a public good. These are products or services that we all rely on and use, but for which it is difficult to charge people individually. Examples include the navigational aid of a lighthouse, law enforcement—and, of course, public health itself. The conception of antibiotics as a public good underscores the responsibility of public authorities, especially national governments working together, to protect this precious resource.

“We need to think about antibiotics as public goods, like airline safety or highway infrastructure,” Yonatan Grad says. “In the same way, isn’t there a moral argument to be made that we should be investing in our antibiotics infrastructure—to curtail, as much as possible, antimicrobial resistance? How many people have to die? How many bad outcomes do there have to be before we say this is a severe enough problem to warrant an investment? A sense of injustice comes to the fore.”

So far, the moral argument has not prevailed. “In some cases, industry spending on promoting products is greater than government investment in promoting rational use of antimicrobial medicines or providing objective information,” noted a 2013 WHO global plan. According to Lipsitch, “The countries that have a stronger sense of public good—the Nordic countries, the Netherlands—have much less overuse of antibiotics. To me, that is not a coincidence.”

“Antibiotic resistance is a test for humanity’s ability to work together,” says Outterson. “This is a global problem that is solvable with science—and it’s easier than many of the other global problems we have. My goal—and the goal of lots of other people working in this field—is for humanity to pass the test.”

There are obvious resonances between today’s threat of antibiotic resistance and climate change. As with climate change, drug resistance seems a distant and abstract risk, even if the signs are all around us. As with climate change, people have trouble visualizing the consequences of an invisible, slow-moving phenomenon. As with climate change, every individual contributes through his or her “footprint.”

But unlike with climate change, scientists agree on what we need to do—there are no “antibiotic-resistance deniers.” And unlike with climate change, there is no entrenched industry, such as fossil fuel companies, that can be demonized—although pharma is getting pushed to do more to solve the problem than it has of late.

Climate change and antibiotic resistance are ecosystem catastrophes. The intuitive link between a warming planet and antibiotic resistance may find its most eloquent avatar in the French microbiologist René Dubos. Although Alexander Fleming is sometimes called the father of modern antibiotics, as are a handful of other groundbreaking scientists of that era, the claim to paternity arguably goes to Dubos. In 1939, he launched the antibiotic era by reporting the discovery of tyrothricin and gramicidin after the first systemic search for antibiotic agents in soil; these drugs were the first natural antimicrobials produced commercially on a large scale and used clinically.

But soon after this signal discovery, Dubos switched gears. He withdrew from antibiotic research, convinced that such agents would only foster the growth of bacterial resistance. In 1942, he cautioned that bacterial resistance to antibiotics was inevitable. The next year, he advised premed students not to follow the example of their mentors, who practiced “the wasteful and inconsiderate use of antibiotics.” By the late 1950s, Dubos explicitly warned that “at some unpredictable time and in some unforeseeable manner, nature will strike back.”

Now nature has struck back, some antibiotics have become futile treatments, and others are poised to fall. Tellingly, Dubos went on to become a passionate voice in the environmental movement, in 1972 coining the iconic phrase “Think globally, act locally.” He championed the idea of a radical interrelatedness: that no organism—whether a microbe, a human being, society, or the earth itself—exists in isolation, and that context is everything.

The “antibiotic apocalypse” that some predict may not happen. But solving the problem, however vast, will mean pulling on biology, economics, politics, culture, psychology, and moral choices. Could there be a better vision than Dubos’ holistic model for figuring out the answer? Will we rise to the challenge before it’s too late?

Madeline Drexler is editor of Harvard Public Health and author of Emerging Epidemics: The Menace of New Infections (Penguin, 2010).