The Breakthrough: What Immunotherapy Actually Does, and How Often

The Breakthrough: Immunotherapy and the Race to Cure Cancer is a popular history of one idea. The immune system can be persuaded to attack cancer. The reason it usually does not is that tumors switch it off.

Charles Graeber is a journalist, not a scientist, and the book is structured the way a magazine feature is structured. It moves between the laboratory and the clinic, and it uses a handful of named patients to carry the stakes.

The idea it traces is old. In the nineteenth century, surgeons noticed something odd. Tumors sometimes shrank after an infection. For most of the twentieth century, immunotherapy was where careers went to die. The book's real subject is the long stretch when the hypothesis was unfashionable and a small number of researchers kept working on it anyway.

Then, in the 2010s, checkpoint blockade arrived and the field changed shape. Graeber published in 2018. That was the year James Allison and Tasuku Honjo won the Nobel Prize in Medicine, for showing that cancer can be treated by releasing the brakes on the immune system. The book is written from inside that moment, and its mood belongs to it.

The book runs to 302 pages in the Twelve hardcover, plus a plate section, notes and an appendix on the science.

The opening chapters set up the immune system in plain terms: what T cells do, why the body needs brakes on them, and how a tumor exploits those brakes. Graeber is patient here, and this is the material most readers come for.

The middle of the book is history. William Coley and his bacterial toxins. The interleukin-2 era, with its dramatic and dangerous responses in a small number of people with melanoma and kidney cancer. The long argument over whether the immune system saw tumors at all. Then the discovery of CTLA-4 and PD-1, the first trials, and the fights over funding and credit.

Interleaved with that are patient chapters. Graeber follows people receiving early checkpoint drugs, including trial participants whose scans got dramatically better after appearing to get worse. These sections are the most emotionally forceful and the most easily misread.

The closing chapters look forward: combinations, CAR T-cell therapy, and the question of why most people do not respond. An appendix gives a more technical summary of mechanism for readers who want it.

Mechanism. Graeber explains checkpoint blockade well enough that a reader can afterwards understand why a drug that does nothing to the tumor directly can still make it shrink. That is genuinely hard to do and he does it without diagrams.

The history is also honest about how science behaves. Ideas were dismissed for decades, partly on evidence and partly on fashion. Grants went elsewhere. People who turned out to be right spent years being treated as cranks. The book is a good corrective to any account of research as a smooth accumulation.

The reporting on interleukin-2 deserves particular credit. It was a treatment that could produce durable remissions in a small minority and put people in intensive care in the process, and Graeber neither romanticises it nor dismisses it. That balance is the hardest thing to hold in writing about cancer drugs, and it shows the book is not simply cheerleading.

It is also good on the strangeness of the response pattern. Immunotherapy does not behave like chemotherapy, and tumors can appear to grow before they shrink. A reader who understands that will follow a scan conversation better.

The title is the first problem. It promises a race to cure cancer, and the honest picture is narrower. NCI's own page states that only a small portion of people who receive immunotherapy will respond to the treatment. The book knows this and says so, but the framing pulls the other way.

The patient stories are selected for drama, which means they are selected for success. Reading the book as a preview of what treatment feels like will set expectations too high. For every long-term responder in the narrative there are people not in the narrative.

Side effects are underweighted. The book does mention toxicity, particularly from interleukin-2. But a reader would not come away expecting inflammation of the thyroid, colon, liver or lungs as a routine possibility.

There is also very little in the book about cost, insurance, or the practical business of getting to an infusion centre every few weeks. That is a legitimate choice for a history of ideas, but it leaves out most of what actually determines whether a person receives one of these drugs.

It is also now dated in the specifics. Approvals, combinations and biomarker testing have all moved since 2018, and CAR T-cell therapy has changed considerably. Nothing about the underlying biology has been overturned.

NCI defines immunotherapy as a biological therapy that uses the immune system to fight cancer. It describes five main types. Immune checkpoint inhibitors. T-cell transfer therapy. Monoclonal antibodies. Treatment vaccines. And immune system modulators. NCI also notes that it is not yet as widely used as surgery, chemotherapy or radiation.

For checkpoint inhibitors specifically, NCI explains that these drugs block checkpoint proteins from sending an off signal to T cells, which allows the T cells to kill cancer cells. It names PD-1, PD-L1 and CTLA-4, and notes that some tumors turn down the T-cell response by producing a great deal of PD-L1.

The approvals NCI lists are wide. They cover breast, bladder, cervical, colon, head and neck, liver, lung, kidney, stomach and rectal cancers. They cover skin cancers, melanoma among them, and Hodgkin lymphoma. They also cover any solid tumor that cannot repair errors in its DNA. That last one is keyed to a molecular feature, not to an organ.

Two things in that list are worth pausing on. Hodgkin lymphoma sits alongside solid tumors, which is a reminder that checkpoint blockade is not a solid-tumor phenomenon. And the DNA repair approval is the first of a family of approvals defined by what a tumor's molecular testing shows rather than where in the body it started.

Treatment goes in by drip, by mouth, on the skin, or into the bladder. It is usually an outpatient visit. Response is judged by examination, blood tests and imaging.

The honest summary has three parts. Immunotherapy is now standard in a meaningful number of situations. It is absent in many others. And where it does apply, it is unpredictable. For the comparison most people actually want, see immunotherapy versus chemotherapy. For the overview, see immunotherapy.

This is where a reader of the book needs supplementing.

NCI explains the mechanism directly: side effects occur because an immune system revved up to act against the cancer also acts against healthy cells. That is a different category from chemotherapy toxicity, and it is managed differently.

NCI lists the common effects. Pain, swelling, rash and redness where the drug goes in. More generally: fever, fatigue, nausea, muscle aches, diarrhea, fluid retention and a blocked nose. Less often, the immune system inflames an organ. That can mean the lungs, the colon, the liver, the pancreas, the thyroid, the pituitary, the heart muscle, the kidneys or the nervous system. NCI notes that severe or fatal reactions are rare but possible. It also notes that any of this can appear at any point, during treatment or after it.

That last point is the one most worth carrying. Unlike chemotherapy side effects, which tend to follow the cycle, an immune-related event can arrive weeks or months later. What counts as a same-day call is a question for the team giving the drug, and it is worth having the answer written down before it is needed. Immunotherapy does not work like chemotherapy, and its side effects can arrive months late. That is the part worth carrying out of the book.

For what these events look like in practice, see autoimmune side effects from immunotherapy.

A book that follows individual patients gives the reader anecdotes, and an approval decision rests on rates. Both are real, and they answer different questions.

A response rate is the proportion of people in a trial whose tumors shrank by a defined amount. It says nothing on its own about how long the shrinkage lasted, and nothing about whether the people who responded lived longer than they otherwise would have. NCI describes response being assessed by physical examination, blood tests and scans that measure whether a tumor has changed size.

What makes checkpoint inhibitors unusual is the shape of the tail. A minority of responses last years, which is why the language of cure appears at all in a field where it rarely does. It is also why an average survival figure describes almost nobody: the group contains people who did not benefit at all and people still in remission long after the trial closed.

None of that is a reason to discount the book. So treat its patient chapters as illustrations of what is possible, not as an estimate of what is likely. For the likely figure in a specific situation, ask the team holding the pathology report.

It suits a reader who wants to understand how checkpoint blockade works and why it took so long, and who enjoys the sociology of science as much as the science.

It suits caregivers who want a mental model before a treatment conversation. It also suits anyone tired of news coverage that treats every result as a cure.

It does not suit someone looking for a practical guide to receiving immunotherapy, because the side-effect material is thin. It also does not suit a reader in a fragile moment who would be hurt by triumphant case histories, since the people in the book are largely the ones for whom it worked.

• Open Library — The Breakthrough (ISBN 9781455568505)
• Internet Archive — The Breakthrough, catalogue metadata
• NCI — Immunotherapy to Treat Cancer
• NCI — Immune Checkpoint Inhibitors
• NCI — Side Effects of Immunotherapy

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