The Broken Pipeline: Why No New Antibiotics Are Coming

One billion dollars. That is roughly what the biotech company Achaogen spent over fifteen years to develop plazomicin, a new antibiotic designed to fight some of the most dangerous drug-resistant bacteria in hospital intensive care units. The U.S. Food and Drug Administration approved the drug in June 2018. In its first months on the market, plazomicin generated $800,000 in revenue for the whole of 2018. Less than a year after approval, Achaogen filed for Chapter 11 bankruptcy. Its assets, including the rights to a drug that took fifteen years and a billion dollars to create, sold for approximately $16 million.

That number should unsettle anyone who takes antibiotics for granted. And it should terrify anyone who understands that climate change is accelerating the spread of resistant bacteria through drying soils worldwide.

A Drug That Costs a Billion and Earns Nothing

Achaogen's collapse was not a story of bad science or poor management. The drug worked. It received FDA approval for complicated urinary tract infections. Clinical trials showed efficacy against aminoglycoside-resistant Enterobacteriaceae, a class of pathogens that hospitals increasingly cannot treat. By the standards of pharmaceutical development, plazomicin was a success.

By the standards of pharmaceutical economics, it was dead on arrival. The problem is built into the product category. A good antibiotic must be held in reserve. Antibiotic stewardship programs, now standard at every major hospital system, deliberately restrict new antibiotics to cases where older drugs have failed. This is medically correct. It preserves the drug's effectiveness by limiting the bacterial exposure that drives resistance. It also means the drug sits on the shelf. An oncology drug prescribed to every eligible patient generates revenue from day one and continues for months, years, sometimes the rest of a patient's life. An antibiotic prescribed only when nothing else works, for a treatment course lasting seven to fourteen days, generates almost nothing.

The better a new antibiotic performs in its intended role, the less revenue it produces. This is not a fixable incentive problem. It is a structural contradiction between public health logic and commercial logic.

The Pipeline That Is Not Really a Pipeline

The World Health Organization's 2024 antibacterial pipeline analysis counted 97 agents in clinical development. That number sounds reassuring until you look at what is actually in it. Only 57 are traditional antibiotics, and of those, just 32 target WHO priority pathogens. The remaining 40 are biologics, bacteriophages, antibodies, and other non-traditional approaches, most still in early clinical phases with high attrition rates.

Of those 32 traditional antibiotics aimed at priority pathogens, only 12 qualify as genuinely innovative, and a mere four of those are active against WHO critical-priority pathogens, the carbapenem-resistant gram-negative bacteria that cause the most lethal hospital infections. Most pipeline candidates are modifications of existing antibiotic classes rather than novel mechanisms of action. Against bacteria that have already evolved resistance to a drug class, a modified version of that same class offers limited hope.

For context, more than 5,000 oncology drugs are currently in clinical development globally. The disparity is not a reflection of disease burden. The Lancet estimated 1.27 million deaths directly attributable to antimicrobial resistance in 2019, with 4.95 million deaths associated with resistant infections. That puts AMR in the same mortality range as HIV/AIDS and malaria combined. The pipeline disparity is a reflection of where the money goes.

Historical attrition rates make the picture worse. Published estimates suggest that fewer than one in thirty preclinical antibiotic candidates reaches the market. With the current pipeline, that arithmetic means roughly two to three new antibiotics will reach patients per year globally. Most of them will be incremental improvements, not the novel-mechanism drugs needed against pan-resistant organisms. The average cost to bring each one to market sits between $1.0 billion and $1.5 billion. The expected peak annual revenue for most hospital antibiotics is around $46 million, a fraction of what is needed to recoup development costs.

Why Big Pharma Left the Room

In 1990, eighteen major pharmaceutical companies maintained active antibiotic research programs. By 2024, that number had dropped to fewer than half a dozen. The exodus was neither sudden nor mysterious. It followed the money.

AstraZeneca sold its antibiotic portfolio to Pfizer in 2016. Novartis shuttered its antibiotic research division in 2018. Sanofi exited antibiotic R&D the same year. When AbbVie acquired Allergan in 2020, the antibiotic assets were a footnote in a $63 billion deal driven by the need to diversify beyond the immunology drug Humira.

The math behind these decisions is straightforward. A successful cancer drug can generate over a billion dollars in cumulative revenue within a few years of approval. A new antibiotic earns roughly $46 million per year. An oncology treatment may be prescribed for months or years. An antibiotic course lasts five to fourteen days. A cancer patient may take the same drug until disease progression, generating steady revenue. An antibiotic patient takes the drug once and, ideally, never needs it again.

These are not comparable investment propositions. Pharmaceutical companies allocate research capital to maximize returns for shareholders. Antibiotics cannot compete in that framework. The executives who shut down antibiotic programs were not villains. They were reading balance sheets.

The Subscription Model: Paying for Insurance, Not Pills

If the volume-based model is the problem, delinking revenue from unit sales is the logical fix. The United Kingdom's National Health Service launched a pilot in 2022 that does exactly this. The NHS agreed to pay a fixed annual fee of up to 10 million pounds per antibiotic for two critical drugs: ceftazidime-avibactam and cefiderocol. The payment is independent of how many doses hospitals actually use.

The logic mirrors insurance. A household pays for fire insurance not because the house is currently burning but because it might. The subscription model treats reserve antibiotics the same way. The health system pays for availability, for the assurance that the drug exists, is manufactured, and is accessible when a pan-resistant infection appears in an ICU. Whether that happens to fifty patients per year or five hundred does not change the payment.

Sweden pioneered a similar approach, guaranteeing minimum annual revenue for antibiotics deemed critical to its national formulary. Japan began developing a pull incentive framework in 2024, exploring guaranteed returns for antibiotics that meet specific medical need criteria.

Early results from the UK pilot are cautiously encouraging. The participating pharmaceutical companies maintained supply and continued to invest in their manufacturing capacity, outcomes that the old volume-based model actively discouraged. The drugs remained available without any incentive to promote wider usage.

But subscriptions require governments to pay for drugs that may sit on shelves for years. That is a difficult budget line to defend in any political system, however rational the economics. A finance minister who authorizes ten million pounds for a drug that treats fifty patients per year will face questions that a finance minister who spends ten million on cancer drugs used by thousands never will. The subscription model is economically sound and politically fragile.

PASTEUR Act: The Bill That Keeps Failing

The largest pharmaceutical market in the world has no pull incentive for antibiotics. The United States PASTEUR Act, the Pioneering Antimicrobial Subscriptions to End Upsurging Resistance Act, would change that. It would authorize up to $6 billion over ten years for subscription-style contracts with antibiotic developers, creating a guaranteed revenue floor for drugs that meet critical medical need criteria.

Senators Michael Bennet and Todd Young first introduced the bill in 2020. It was reintroduced in 2021 and again in 2023. Each time it attracted bipartisan cosponsors. Each time it failed to reach a floor vote.

The obstacle is not opposition. Few legislators openly argue against antibiotic development. The obstacle is budget scoring. The Congressional Budget Office scores subscription payments as government spending, a cost line in the federal budget. The economic benefit of having effective antibiotics available, reduced ICU stays, shorter hospital admissions, lower mortality, shows up in the healthcare system but not in the federal ledger in a way that offsets the scored cost.

Meanwhile, the pipeline continues to depend on a patchwork of push incentives. CARB-X, the Combating Antibiotic-Resistant Bacteria Biopharmaceutical Accelerator, has invested approximately $500 million in early-stage antibiotic research since its launch in 2016. The U.S. Biomedical Advanced Research and Development Authority, BARDA, funds late-stage development. The Global Antibiotic Research and Development Partnership, GARDP, a nonprofit backed by the WHO and the Drugs for Neglected Diseases initiative, has partnered with Venatorx Pharmaceuticals to secure access to cefepime-taniborbactam, which completed Phase 3 trials.

These organizations do valuable work. CARB-X has funded more than 100 projects across 15 countries. GARDP provides a development pathway that does not require commercial returns. But none of them solve the market-entry problem. A drug that reaches FDA approval still faces the same economic void that destroyed Achaogen. Push funding gets drugs to the finish line. Without pull incentives, there is nobody waiting at the finish line with a check.

Why Climate Makes All of This Worse

Research published in 2026 by Dianne Newman's laboratory at Caltech, in Nature Microbiology, demonstrated that drought increases the concentration of natural antibiotics in soil. As soil dries, the chemical compounds that soil bacteria use to compete against each other become more concentrated. This creates selection pressure that favors resistant organisms. Clinical data from multiple countries confirmed the correlation: drier regions consistently show higher rates of antibiotic-resistant pathogens in hospitals.

This finding connects the antibiotic pipeline crisis to the climate crisis in a way that compresses the timeline for action. The bacteria that thrive in drought-stressed soil are predominantly gram-negatives, the same category of pathogens that the antibiotic pipeline most conspicuously fails to address. Carbapenem-resistant Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa: these are the organisms that drying soil selects for, and these are the organisms that almost no drug in the current pipeline can reliably treat.

Climate projections show arid and semi-arid zones expanding by two to six percent globally by 2050 under moderate warming scenarios. That expansion translates directly into a larger area of soil exerting selection pressure for resistant bacteria. Rising temperatures also increase the rate of horizontal gene transfer between bacteria, accelerating the movement of resistance genes from environmental organisms into human pathogens.

The O'Neill Review on Antimicrobial Resistance projected 10 million deaths per year by 2050 if resistance trends continued unchecked. That projection, published in 2016, assumed relatively stable environmental conditions and focused on the clinical drivers of resistance: overprescription, poor infection control, agricultural overuse. It did not account for climate-driven acceleration of environmental resistance. If the Newman findings hold across ecosystems, and the early epidemiological data suggests they do, the actual trajectory may be steeper than any model currently projects.

The Math That Does Not Work

Developing a new antibiotic takes ten to fifteen years from preclinical research to market approval. Once approved, a hospital antibiotic may maintain clinical effectiveness for one to two decades before resistance erodes its utility, though the timeline varies widely by drug and pathogen. At current pipeline throughput, roughly two to three new antibiotics reach the market each year, and many of those are incremental modifications rather than novel mechanisms effective against pan-resistant bacteria.

The replacement rate cannot keep pace with the resistance rate. Not under current conditions, and certainly not under a climate scenario that accelerates resistance selection in the environment.

Global public funding for AMR research and development stands at roughly $1.4 billion per year from G7 countries and the EU alone, the vast majority of the global total. The direct healthcare cost of AMR is estimated at $66 billion annually, and the full economic burden including productivity losses is likely far higher. The ratio of investment to damage remains staggering. No other public health challenge of comparable scale operates with that level of underinvestment.

The arithmetic leads to a conclusion that requires no editorial commentary. If development takes fifteen years, if resistance can develop within one to two decades, if the pipeline delivers two to three drugs per year, if most of those are not novel, and if climate change is accelerating the rate at which resistance emerges in the environment, then the system is drawing down a finite stock of effective antibiotics faster than it can replenish them.

Achaogen's assets sold for $16 million. A single round of ICU treatment for a patient with a pan-resistant gram-negative infection costs $50,000 to $100,000. The drug that took a billion dollars and fifteen years to build was worth less at auction than the hospital bills of two hundred patients who might need it. The market knew the price. It never learned the value.