The Truth in Small Doses: How Cancer Progress Gets Measured, and Who Pays for It

The Truth in Small Doses: Why We're Losing the War on Cancer, and How to Win It began as a magazine article. Clifton Leaf was an editor at Fortune, and he had been treated for Hodgkin lymphoma as a teenager. In 2004 he wrote a cover story asking why, after decades of funding, cancer death rates had moved so little. The book is the long version.

Its subject is not cancer biology. It is the machinery around cancer biology. What gets funded, what gets published, what gets approved. And what any of that has to do with people living longer.

Leaf's answer is that the system aims at things it can count. Those things are not the outcome anyone wants. Grants reward safe, incremental proposals. Journals reward positive results. Trials are designed around endpoints that are easier to hit than survival. Careers are built on all three.

He is careful about one thing that matters. He is not arguing that researchers are dishonest, or that treatment does not work. He was cured by the treatment he is writing about. He is arguing that a well-intentioned system can be badly aimed.

The book runs to 499 pages in the Simon & Schuster hardcover, with extensive notes.

The opening section is history and framing. The National Cancer Act, the language of a war, and the numbers that made Leaf doubtful. He works through why age-adjusted death rates are the honest measure, and why five-year survival can move without them moving.

The middle is reporting, and it is the bulk of the book. He follows the peer review system and why it favours the predictable. He examines preclinical research and the poor track record of animal models. He looks at the drug development pipeline and at how trials get designed. He asks what an approval actually settles. He interviews researchers, who speak with some candour about the rewards they work under.

A separate thread is about prevention. He argues it is chronically underfunded next to treatment. Another is about screening, where he doubts more than the advocacy literature does.

The final section is prescriptive. He proposes changes to how research is funded, how trials are built, and how success is defined. That includes far more weight on preventing cancer rather than treating it late.

Leaf's own history appears throughout, in short passages rather than as a memoir.

The measurement argument is the durable part. Leaf shows, patiently and with examples, how a drug can be approved, marketed and prescribed. The endpoint behind it may never have been shown to mean a longer or better life. Once a reader has that distinction, a great deal of cancer news becomes readable.

The reporting on rewards is also strong, and it holds up because the structures have not really changed. Grant review still rewards feasibility. Publication still rewards novelty and positive findings.

He is good on prevention as well. Far more is spent on treating advanced cancer than on stopping cancer from starting. That gap is real, and he makes the case without turning it into a lifestyle lecture.

And his tone is right. He is a survivor of the system he criticises, and there is no trace of contempt for the people in it.

The timing is the main problem. The book was published in 2013 and reported in the years before. Checkpoint inhibitors were barely on the horizon, and several of the developments that most change the picture arrived after it was published. A reader who takes the book's pessimism as current will be reading a 2012 snapshot.

Some specific claims have also been overtaken. The list of drugs, prices, approval decisions and funding figures is a decade old and should not be quoted as current.

The framing of losing a war carries the same problem as the phrase always does. Death rates have fallen a lot for several cancers. The fall is uneven, and the reasons are mixed: less smoking, screening, and better treatment. The title flattens that.

And the prescriptions are the least tested part. Rebuilding how research is funded is easy to propose and hard to judge. The book cannot show that its alternatives would work better.

The federal material here is what turns the book's argument into something a reader can use in a clinic.

FDA issued guidance in December 2018 on endpoints for cancer clinical trials. It covers trials submitted to support effectiveness claims in new drug applications, biologics license applications or supplements. It applies to drugs for treating existing cancer. It explicitly does not cover endpoints for drugs to prevent or reduce the incidence of cancer.

The distinction at the centre of everything is between an endpoint that measures benefit and one that is thought to predict it. Overall survival measures how long people live, and it is not open to interpretation. Progression-free survival measures the time before the cancer grows or the person dies. That depends on when scans happen and on how growth is defined. Response rate measures whether tumors shrank.

FDA's own description of accelerated approval, a pathway established in 1992, makes the trade explicit. For serious conditions with unmet need, FDA can approve a drug based on a surrogate or intermediate clinical endpoint. A surrogate endpoint is a laboratory measurement, imaging finding or physical sign. It is thought to predict clinical benefit. It is not itself a measure of clinical benefit.

The safeguard is the second half. After accelerated approval, sponsors must run studies verifying that the surrogate endpoint predicts real benefit. Those trials may fail to confirm benefit, or show too little benefit relative to risk. FDA may then withdraw the approval or change the labelled indication.

That is exactly the mechanism Leaf's book describes, written down by the agency that operates it. It is neither a scandal nor a formality. It is a chosen trade: certainty for speed. There is a stated plan for settling the doubt later. For what these terms mean in a specific trial, see what are trial endpoints and clinical trial results, what they mean.

The other half of Leaf's argument is about funding, and NCI publishes the shape of it.

NCI received $7.35 billion for fiscal year 2026, an increase of $128 million over the FY25 enacted level. The largest single share goes to research project grants. That is the investigator-initiated mechanism that includes R01 and R37 awards. It pays for single projects at institutions around the country. That is the part of the system Leaf's critique of peer review is aimed at.

NCI funds more than grants. It funds its designated cancer centres. It funds Specialized Programs of Research Excellence aimed at particular cancer types. It funds training and career development awards, including National Research Service Awards, and clinical trials networks. Intramural research, conducted inside NCI itself, is approximately 18% of the budget. From fiscal 2017 through 2023, the 21st Century Cures Act provided $1.8 billion over seven years for the Cancer Moonshot.

Knowing that structure is what lets a reader evaluate the book's proposals. A criticism of how R01 grants are reviewed is a criticism of the biggest line item. A proposal to fund more prevention research is a proposal to move money between programmes that all have constituencies. Leaf's question is how progress gets counted, and who pays for the counting.

For the basics of how trials are run, see what are clinical trials.

It suits readers who want to know why cancer headlines and cancer outcomes drift apart. They have to be willing to read 499 pages to find out. It suits anyone facing a treatment decision. It shows them how to ask what a benefit figure was measured against.

It suits journalists, policy readers and students of how science funding works, and survivors who have wondered why progress feels slower from outside than the press releases suggest.

It does not suit someone looking for help with their own diagnosis. There is none in it. It does not suit a reader who wants current facts, since the examples stop around 2012. And it may be a hard read for someone in active treatment who needs to trust the people treating them. The book's subject is the system's failures, not its successes.

• Simon & Schuster — The Truth in Small Doses
• Open Library — The Truth in Small Doses (ISBN 9781476739984)
• FDA — Accelerated Approval
• FDA — Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics (Guidance, December 2018)
• NCI — Budget and Appropriations

This page discusses a book for education. It is not medical advice, and nothing here is a judgement of anyone's real medical care.