Published in MIT Technology Review by Maryn McKenna
We’ve known the dangers of antimicrobial resistance for years. What if we used what we learned from COVID to tackle it?
It was August 2017, and pleasant and breezy in the central mountains of Madagascar. The passengers loading their bags into the minibus leaving Ankazobe, a small town in the highlands, were grateful for the morning coolness. It would be warm and sticky on the trip they were taking to Antananarivo, the island’s million-person capital 100 kilometers to the south, and then to Toamasina on the coast, another 350 kilometers away. One of the passengers, a 31-year-old man, looked uncomfortable already. Four days before, he had arrived on a visit. Now he was headed home, but he was feverish, achy, and shaking with chills.
He never made it. The man died in the minibus after it drove through the capital; the panicked driver dropped his body off at a hospital and then continued toward the coast.
Within days, 31 people linked to the taxi trip and the hospital fell ill, and four died. Two weeks later, a woman with no known ties to the trip died in the densely packed capital. Shortly after, doctors discovered what was killing them: plague. By early October, there were 169 cases scattered across the island nation. By the end of the month, there were more than 1,500.
Small outbreaks of plague occur every year in Madagascar, transmitted by fleas that live on rats whose numbers boom after the rice harvest. This was not like those outbreaks, though. It arrived before the harvest was over. It spread primarily in cities, not the countryside. And, most important, it wasn’t bubonic plague, the historically dreaded but actually not very contagious form of the disease. Instead, it was pneumonic: highly contagious, transmitted by coughing and breathing, and lethal within 24 hours if not treated right away.
With $1.5 million in emergency assistance and 1.2 million doses of antibiotics from the WHO, Madagascar managed to contain the epidemic. But by the time it subsided, at the end of November, it had caused 2,348 cases and 202 deaths. Still, epidemiologists were conscious of having dodged a catastrophe—not just because the fast-moving, potentially fatal illness could have spread worldwide.
Twenty years earlier, in a small seasonal outbreak, Malagasy and French researchers had discovered a strain of plague that was resistant to almost all the antibiotics used against it. If that strain had been responsible for the 2017 outbreak, it would have been untreatable. The result could have been as grim as the plague epidemics of the past: the Manchurian Plague that killed 60,000 people in China in 1910; the Justinian Plague that destabilized the Byzantine Empire in 540; the Black Death, which killed an estimated 50 million and wiped out half the population of Europe.
Such a catastrophe would not have surprised the global circle of scientists who monitor the bacterial world’s ceaseless struggle against the antibiotics we use to contain it. While COVID-19 drew our attention to the threat of viruses, microbiologists have long worried that we have forgotten the threat of bacterial epidemics, and the looming danger that bacteria will become resistant to the drugs we rely upon.
“Antimicrobial resistance may not seem as urgent as a pandemic, but it is just as dangerous,” Tedros Adhanom Ghebreyesus, director-general of the World Health Organization, said in November, calling it “one of the greatest health threats of our time.”
In 2014 the Review on Antimicrobial Resistance, a research group put together by the British government, estimated that antibiotic resistance kills 700,000 people around the world each year, a number that was horrifying then but seems small in comparison to the spiraling losses of COVID-19. But the researchers also predicted that if nothing was done, the death rate by 2050 would reach 10 million per year— almost three times COVID-19’s toll so far.
In other words: COVID took us by surprise, but we already know another health crisis is coming, and now we know how to deal with it.
The response to COVID-19 shows what can be accomplished when focus, determination, and vast amounts of money are all directed at one target. The pandemic reorganized the everyday practice of science, the pace of clinical trials, and the willingness of governments to provide funds for that work. With a similar effort applied to antibiotic resistance, we might reorganize trial design, create new surveillance networks to detect resistant pathogens as they emerge, and devise new ways to fund drug development.
Or, to state this more simply: we need to treat antimicrobial resistance as an emergency too. Because it already is.
The math of antibiotics
It is dizzying to look back 18 months, to before the pandemic began, and remember that COVID-19 had never been seen before—and therefore there were, of course, no vaccines against it. What we’ve achieved by now—with eight vaccines approved, almost 100 more in trials, and more than 2.7 billion doses administered worldwide—was possible only because extraordinary amounts of funding were allocated and rules were changed to make it easier to produce drugs.
The US government gave $18 billion to Operation Warp Speed to fund vaccine and treatment research and production. It streamlined clinical trials, allowing vaccines to enter the market without full approval from the Food and Drug Administration. And it agreed to purchase up to 900 million doses of vaccine from six companies if their formulas passed FDA scrutiny.
“Antimicrobial resistance may not seem as urgent as a pandemic, but it is just as dangerous.”
Tedros Adhanom Ghebreyesus, director-general of the World Health Organization
Those grants and promises guaranteed the vaccine manufacturers an income, while relieving them of almost all the financial risks of drug development. Drug makers often talk about navigating the “valley of death,” the difficult-to-fund gap between making a promising discovery and concluding clinical trials. Operation Warp Speed took the valley and laid a six-lane suspension bridge over it.
Antibiotics makers look at these guarantees wistfully. It’s hard to turn a profit on new antibiotics—even ones that could deal with a bacterial pandemic. Antibiotics are cheaper than other drugs sold in the US, but hospitals and physicians feel pressure to use them conservatively to keep resistance from emerging.
Those two influences combine to keep revenues so low that almost all the firms that created antibiotics in the 20th century have left the sector. The last new family of antibiotics was a product of those big-company research programs; it debuted in 2003.
The gap they left has been filled by small biotech companies, with small staffs and a small number of products. Sometimes they have no approved drugs in production at all, leaving them exposed to a second valley of death: the one between achieving licensure and earning enough revenue to be sustainable. Most don’t make it. Since 2018, multiple small companies making new antibiotics—including Achaogen, Aradigm, Melinta Therapeutics, and Tetraphase Pharmaceuticals—have gone bankrupt or sold off their assets.
The math that explains why is uncomplicated. It takes up to $1.5 billion to shepherd an antibiotic all the way through approval, but the average income from a new drug is just $46 million a year. The Review on Antimicrobial Resistance has estimated that a new antibiotic doesn’t reach profitability until 23 years after its development. That’s 13 years after going on sale, and just two years before generic versions can compete against it. Most small companies simply can’t afford to wait that long.
“Investors look at this and say: ‘Why should I put money in a company that is not going to be able to see a return on investment?’” says Ramani Varanasi, who was president and CEO at X-Biotix Therapeutics until it shut down its research programs in April.
Operation Warp Speed solved that problem for COVID by throwing money at research teams that had survived on crumbs. The question is whether a Warp Speed for novel antibiotics could find support to do the same.
“You can always put off investing in tunnel maintenance, until the day the tunnel fails,” says Kevin Outterson, a Boston University law professor who founded and leads CARB-X, a nonprofit that has gathered almost $500 million in philanthropic and government funds to support early-stage antibiotics research. “Antibiotic effectiveness is like that: It’s something that is valuable to all of society, and if we don’t make these investments to keep it up, we’ll regret it.”
Antibiotics date to Sir Alexander Fleming’s serendipitous discovery in 1928 that a substance excreted by mold on his laboratory plates was killing the bacteria he had cultured there. The mold was producing the raw version of penicillin, which after a decade of further research was turned into the first modern antibiotic.
Antibiotics are complex molecules that interfere with cellular reproduction in a range of ways—compounds that are made by organisms to compete with other organisms. By adopting them for human use, medicine stepped into the middle of an endless evolutionary battle in which bacteria both produced weapons against each other and developed defenses against those weapons. Fleming understood this. In 1945, three years after penicillin was first distributed to troops in World War II, he predicted that bacterial evolution—antibiotic resistance—would eventually undermine the new drugs. He said at the time that the only remedy was to use them conservatively, so that the bacterial world would be slow to adapt.
For the first few decades after penicillin’s introduction, bacterial adaptation and drug discovery leapfrogged each other, keeping antibiotics’ ability to treat infections in front of pathogens’ skill at evading them. But by the 1970s, that midcentury burst of innovation had faded. Making antibiotics is hard: the drugs have to be nontoxic to humans but lethal to bacteria, and they must use mechanisms that dangerous bacteria haven’t yet evolved defenses against. But moving from antibiotics produced in nature to synthesizing compounds in a lab was even harder.
Resistance, meanwhile, leaped ahead. Overuse in medicine, agriculture, and aquaculture spread antibiotics through the environment and allowed microbes to adapt. Between 2000 and 2015, use of the antibiotics that have been reserved for the most lethal infections almost doubled worldwide. Levels of resistance differ by organism, drug, and location, but the most comprehensive report done to date, published in June 2021 by the WHO, shows how fast the situation has changed. Among the strains of bacteria that cause urinary tract infections, one of the most common health problems on the planet, some were resistant to a common antibiotic up to 90% of the time in certain countries; more than 65% of the bacteria causing bloodstream infections and more than 30% of the bacteria causing pneumonia resist one or more treatments as well. Gonorrhea, once an easily cured infection that causes infertility if left untreated, is rapidly developing resistance to all the drugs used against it.
At the same time, resistance factors—the genes that control bacteria’s ability to protect themselves—are traveling the globe. In 2008, a man of Indian origin was diagnosed in a hospital in Sweden with a strain of bacteria carrying a gene cluster that allowed it to resist almost all existing antibiotics. In 2015, British and Chinese researchers identified a genetic element in pigs, pork in markets, and hospital patients in China that allowed bacteria to defuse a drug called colistin, known as an antibiotic of last resort for its ability to tackle the worst superbugs. Both those genetic elements, hitchhiking from one bacterium to another, have since spread worldwide.
In the face of drug development’s difficult economics, antibiotic research has not kept up. In March, the Pew Charitable Trusts assessed the global pipeline of new antibiotic compounds. Though the group found 43 somewhere in preclinical or clinical research stages, it determined that only 13 were in phase 3, only two-thirds of those would be likely to make it through to licensure—and none possessed the molecular architecture to work against pathogens that are already the most difficult to treat.
Lessons from Warp Speed
So what would an Operation Warp Speed for antibiotic resistance look like?
The antibiotic pipeline needs a boost in several key areas: basic research, trial design, and post-approval incentives. Fortunately, the global response to COVID created precedents for all three.
The first step would be supporting basic research in the long term. The Moderna and Pfizer-BioNTech vaccines were ready to go less than a year from the first recognition of human infections. But that readiness came from 10 years of basic research with no specific disease in mind. Once COVID appeared, Warp Speed brought the Moderna vaccine to the finish line with extra research funding. (Pfizer didn’t receive research support from Warp Speed, but both companies got funds for manufacturing and production.)
Most early research funding for antibiotics currently comes from a patchwork of investment and philanthropy. So the first lesson of the COVID response may be that basic research into antibiotic compounds needs more support, more broadly distributed—because no one knows which research team will be the next Moderna or BioNTech.
The COVID response demonstrated regulators’ willingness to talk with companies and modify trial procedures to get a faster result. Changes included allowing clinical trials to drop placebo components, for example, or letting participants know which compounds they received. Antibiotic trials can struggle to recruit enough patients, so the prospect of simplified or smaller trials—the kind authorized for rare-disease drugs, for instance—could make a difference in keeping a research program funded.
Antibiotic developers talk about “push” and “pull” incentives. Pushes provide enough funding to propel an antibiotic research program up to the point of approval; pulls contribute a second tranche of cash that carries a new drug through post-approval marketing, surveillance costs, and shortfalls in earnings until they reach profitability. Most of the funding sent toward antibiotic research now constitutes push incentives, designed to kick-start research.
But Warp Speed was both push and pull: it included not just research support but funds for scaling up manufacturing and guarantees that the vaccines would be bought. That two-tiered funding structure could set a pattern for a way of supporting new antibiotics long enough to let them find their footing.
“These are commercial products, but they are also public health goods that we need to remain viable,” says Phyllis Arthur, vice president of infectious diseases and diagnostics policy for the industry organization BIO. “They’re supposed to be kind of behind glass. But being behind glass means that there’s no ROI that makes sense, so you have to do something that captures their value without putting the onus on the commercial market to provide it.”
There are existing proposals that would funnel more cash to antibiotics makers, but without the urgency of an event as apocalyptic as the COVID-19 pandemic, they have not yet won enough public or political support to launch.
In the US, several pieces of legislation that could help are awaiting scrutiny in Congress. One, called the DISARM Act, aims to improve the market for newly produced antibiotics by creating financial incentives that encourage hospitals to purchase and use them. Right now, government reimbursement for hospital care encourages health-care institutions to use less-expensive drugs first, and more-expensive, newer drugs if the first round doesn’t work—a situation that fosters resistance without getting manufacturers the sales revenue they need.
“We need to treat antimicrobial resistance as an emergency too. Because it already is.”
But in the same way that Operation Warp Speed opened the door for more appropriations—the Biden administration committed $500 million in March to a new national center for forecasting possible epidemics, for example—the realization that we are increasingly vulnerable to bacterial infections might inspire even bolder actions. Governments could plan for new antibiotics the way militaries plan for new planes and tanks, providing the weaponry for imagined battlefields with contracts that extend years into the future.The creators of the second proposal, known as the PASTEUR Act, have called it a “Netflix for antibiotics.” It proposes federal payments to companies that bring out novel antibiotics, as a way of guaranteeing the drugs’ availability in the future. (The act is based in part on an antibiotic subscription model introduced by the government of the United Kingdom last summer, which would pay lump sums to companies at the start of antibiotic research programs in exchange for guaranteed access to the drugs once they are developed.)
Brad Spellberg, the chief medical officer of Los Angeles County + University of Southern California Medical Center, has proposed a different model for antibiotic development: endowing nonprofits that would continuously develop new compounds but not go through the expense of clinical trials.
The point, he says is that companies seeking profit must focus on getting one drug at a time through approval—but to defeat resistance, society needs multiple drugs and a predictable supply of new ones. “You want to have a steady, slow drip every few years of new needed molecules,” he says, “so that when there is a new, emergent pathogen, you can pull a drug out of the bullpen and do rapid clinical trials, the way that has been done with COVID.”
The boldest idea inspired by the COVID response might not be about investing in making drugs, however. Instead, it could be about investing in the people who make them. As big antibiotics makers left the field and small companies crashed, the teams that did the work were broken up and lost; almost all the antibiotics we consume today were developed by people who have since retired, and few researchers are vying to replace them.
“If you’re an up-and-coming young scientist and you’re looking at the big problems you can tackle, but you understand that they have to be financed in some kind of way, picking antimicrobial resistance as the lane you’re going to go down is almost career suicide,” says Gerry Wright, director of the Michael G. DeGroote Institute for Infectious Disease Research at McMaster University.
If the first lesson of the COVID response was the value of funding basic research over time, maybe the last should be the value of finding researchers—for this pandemic, and for the next one too.
“If I were going to make a big play, I would invest in people,” Wright says. “Graduate students, postdocs, assistant professors, associate professors. Pay their salaries. Give them money to take risks, because solving this problem will mean taking enormous risks. There’s no shortage of brains. It’s just a shortage of opportunity.”
A quiet warning
Last week, deaths from COVID-19 in the US topped 600,000. Worldwide, the toll of death from the disease crept above 3.8 million. At this moment, cases have topped 178 million.
Among those enormous numbers, it would have been easy to miss a small bulletin that was also published last week. In the province of Ituri in the northeast corner of the Democratic Republic of the Congo, health officials announced that 19 people had fallen ill, and 11 people had died. They had pneumonic plague, the same disease that had killed hundreds in Madagascar four years ago. Samples taken from the victims had been shipped to a regional lab, the announcement said, but there was no immediate notification of what they might show.
In the avalanche of terror and grief caused by COVID-19, the news was barely the fall of a pebble. But it ought to be a reminder that pebbles can trigger avalanches too. COVID was the pandemic that took us by surprise; It will be on us if we allow antimicrobial resistance to do the same.