Case Study 25.2 — The Drug You Get Punished for Selling

The antibiotic market failure: why the commercially worst class in medicine is the one we most need, and what has been proposed about it

The situation

Almost every argument you will read about why we do not have new antibiotics is a scientific argument. Discovery is hard. The Gram-negative outer membrane is a formidable barrier. The easy targets were taken in the 1950s. All of that is true.

None of it is the binding constraint.

Suppose the science were solved tomorrow — suppose a laboratory produced an antimicrobial peptide with a wide selectivity margin, no renal toxicity, and reliable activity against the organisms in §25.1. That compound would still face a market in which succeeding is financially indistinguishable from failing, and it would face it after the eight-to-twelve-year, several-hundred-million-dollar process that every drug must survive.

This case study asks you to reason carefully about a structural problem, because structural problems have a particular property that makes them easy to misdiagnose: they produce outcomes that look like somebody's fault, and they are not.

Part 1 — The arithmetic

A commercially viable drug is one taken by many people, for a long time, at a price the market will bear. Consider two drugs approved in the same year.

Drug 1 treats a chronic metabolic condition. A patient who starts it may take it daily for twenty years. The prescriber base grows as the indication broadens. Revenue accumulates across a patent lifetime and continues after it.

Drug 2 is a novel antibiotic with reliable activity against a resistant organism nothing else touches. Consider what happens to it:

It is used briefly. A course lasts days.

It is used rarely. The great majority of infections are treated empirically with older, cheaper, well-characterized drugs, and should be.

It is deliberately withheld. This is the part that has no analogue anywhere else in medicine. The better Drug 2 is, the harder infectious disease specialists and antimicrobial stewardship programs will work to prevent its use — restricting it to cases where nothing else will do, requiring approval for each prescription, and monitoring consumption. This is correct medicine. Widespread use would select for resistance and destroy the drug's value for everyone. From the manufacturer's point of view, it is a formal, well-resourced, ethically mandatory program to suppress sales of the product.

And success erodes the asset. Every course administered applies selective pressure. The drug depreciates through use in a way no other class does.

Notice what this means. For Drug 2, the clinical system functioning correctly produces the commercial outcome. There is no villain. Every actor — the stewardship pharmacist restricting it, the physician who reaches for an older agent first, the formulary committee that puts it behind an approval requirement — is doing exactly the right thing.

Part 2 — What this has produced

The consequences are visible in the structure of the industry rather than in any single decision.

Most large pharmaceutical companies have exited antibacterial discovery. What remains is concentrated in small firms, academic groups, and public-private partnerships dependent on grant funding.

And approval has repeatedly failed to rescue the companies that achieved it. A firm that won approval in 2018 for a novel antibacterial agent against multidrug-resistant Gram-negative infections filed for bankruptcy less than a year later, unable to generate meaningful revenue from a drug that did what it said it did. Another antibiotic-focused company entered bankruptcy in the same period.

In this class, approval is not the finish line, and several firms have discovered that after crossing it. For anyone deciding where to allocate research capital, that observation does more work than any scientific argument about difficulty.

Now layer the peptide-specific problem on top. Peptides cost more to manufacture than small molecules — solid-phase synthesis builds a chain one residue at a time, and yield losses compound across the sequence. Recombinant production is awkward for antimicrobial peptides specifically, because they kill the host organism being used to produce them. A high cost of goods is absorbable for a drug taken daily for decades. For a drug used for a week, rarely, under restriction, it is not.

Antimicrobial peptides therefore sit at the intersection of the two problems: an expensive molecule class entering the one market that cannot pay for expensive molecules.

Part 3 — What has been proposed

Because the problem is structural, the proposed remedies are structural. All of them share one insight: if the trouble is that revenue is tied to volume, break the link between revenue and volume.

Subscription-style procurement. A health system pays a fixed annual sum for access to a drug regardless of how much is used — sometimes described by analogy to a streaming subscription. The manufacturer is paid for availability rather than units, which removes any incentive to push volume and prices the drug as what it actually is: insurance against a future emergency. Pilot programs along these lines have been implemented in at least one national health system.

Market entry rewards. A large lump-sum payment on approval of a qualifying novel antibiotic, decoupled from sales entirely. A prominent international review of antimicrobial resistance published in 2016 recommended rewards on the order of a billion dollars per qualifying drug.

Extended or transferable exclusivity. Additional market exclusivity for the developer, or a voucher applicable to a different product. Repeatedly proposed and repeatedly criticized, because the cost falls on payers of an unrelated medicine — potentially a very expensive one.

Push funding. Direct subsidy of early research and development, which already exists through public and philanthropic mechanisms and is why much of the current pipeline exists at all.

Chapter 25 deliberately takes no position among these. The trade-offs involve public spending, intellectual property, and international coordination, and reasonable people disagree. What is not in dispute is the diagnosis.

Part 4 — Why this belongs in a book about peptides

Because it changes what you should conclude from an absence.

The most common inference about antimicrobial peptides is: forty years of research, no systemic drug, therefore the science does not work. That inference requires the assumption that a working compound would necessarily have reached market. In most therapeutic areas that assumption is roughly sound. In antibacterials it is not, and a reader who does not know that will systematically misread the field's record.

This does not mean the science is fine and only the market is broken — §25.5 describes a genuine, specific, physical obstacle that has defeated well-funded programs on its own merits. It means the absence of an approved systemic AMP is weaker evidence against the science than it looks, because a second filter sits downstream of the first.

Discussion questions

1. Explain, without using the words "greed" or "profit," why a pharmaceutical company might rationally decline to develop a drug it agrees the world urgently needs. Then identify the specific feature of good clinical practice that makes this outcome unavoidable under the current model.

2. The chapter says that for a novel antibiotic, "the clinical system functioning correctly produces the commercial outcome." Is there any other drug class where this is true? If you can name one, say what it shares with antibiotics; if you cannot, say what makes antibiotics unique.

3. Choose one of the four proposed remedies and argue for it in a paragraph. Then argue against your own choice in a second paragraph, using an objection you find genuinely troubling rather than one you can easily dismiss. Finally, say what evidence would move you.

4. Subscription-style procurement pays for availability rather than units. Identify two practical problems with implementing it that the summary in this case study does not mention — for example, concerning who sets the price, what happens across borders, or how a drug qualifies.

5. A commentator argues: "If a company can't make money selling a drug that saves lives, that is an argument against leaving drug development to companies." Steelman this position, then steelman the opposing view that public development would face its own version of the same problem. Which parts of the disagreement are empirical and which are about values?

6. Return to the inference in Part 4. Write a short paragraph — for a reader who has not read this chapter — explaining why "no approved systemic AMP after forty years" is weaker evidence against the science than it appears, without implying that the science is therefore fine. This is the harder half of the exercise: your paragraph must leave the selectivity problem of §25.5 fully intact.