Self & Other
12. Cancer and the Immune System
In 1891, a young surgeon at New York Hospital named William Coley took a case that had been bothering him for some time and tried something almost unhinged.
The case was a man named Fred Stein, who had a sarcoma in his neck — a cancer of the soft tissue — that had been operated on four times and kept coming back. By the fall of 1890 the tumor was the size of an egg, the surgeons had given up, and Stein had been sent home to die. But shortly after the last operation, Stein had developed a serious skin infection at the surgical site — erysipelas, caused by streptococcal bacteria — and over the weeks of fever and inflammation that followed, the tumor had shrunk. By the time the infection cleared, the cancer was gone. Coley searched for him afterward and found him alive in a tenement on the Lower East Side, seven years out, no recurrence.
Coley spent the next two decades trying to recreate this. He took bacteria he hoped were killed and others he hoped were attenuated, mixed them into preparations that became known as Coley’s toxins, and injected them into the tumors of cancer patients. Some patients had no response. Some got severely ill from the infections. Some, however, had striking, durable regressions of cancers that other doctors had pronounced incurable. Coley collected these cases for the rest of his life. He could never explain, to anyone’s satisfaction, why his treatment worked when it worked or failed when it failed. He was, by the time of his death in 1936, considered a sort of medical embarrassment.
What Coley could not have known is that he had stumbled into the central problem of an entire field that did not yet exist. He had recognized, however dimly, that the immune system can sometimes destroy cancer. He had no idea how. He had no tools to find out. The science available to him in 1891 could not have explained, even in principle, what he was seeing. The explanation would take another hundred and twenty years.
This chapter is about how the explanation was finally found, what it looks like, what it has changed, and what it has not. It is the most recent of the great immunological stories in this book — most of what we are about to describe was unknown when many of you reading this were born — and it is, in places, almost extravagantly hopeful. We will try to be honest about what works for whom. But the larger story is real, and it is one of the most remarkable transformations in modern medicine.
The Hardest Recognition Problem
To see why cancer is such a strange problem for the immune system, recall what the immune system is set up to do.
We have spent eleven chapters describing a system whose central question is: is this me, or is this something else? Things that are something else — viruses, bacteria, transplanted tissue, allergens — trigger responses. Things that are me are tolerated. The whole architecture, from the V(D)J shuffle to negative selection in the thymus to the MHC system on every cell, is built around making this distinction reliably.
Cancer cells violate the premise of the question. They are not something else. They are you, gone wrong. They started as ordinary cells in your body, accumulated a series of mutations in genes that control how cells grow and divide, and at some point crossed a line where they began to grow uncontrollably. They are, genetically, almost identical to the healthy cells around them. They display the same MHC molecules. They are part of the body. The immune system was, broadly, not built to attack them.
Almost. But there is a wrinkle. The mutations that turn a normal cell into a cancer cell are mutations in DNA, which means they produce, sometimes, mutant proteins. Those mutant proteins, when chopped up and displayed in the cell’s MHC molecules in the way we described in Chapter 6, look slightly different from the normal proteins the immune system was trained, in the thymus, to ignore. The cell holding up a card with a mutant fragment is, in principle, holding up something not-quite-self.
In some cases, this is enough. There is now reasonably good evidence — from a long arc of research that has come to be called cancer immunosurveillance — that your immune system is, all the time, identifying and quietly eliminating cells that have begun to turn cancerous. Most of the cancers your body starts to develop, you never know about. T cells with receptors that happen to fit the mutant fragment find the cell, recognize the abnormality, and kill it. The same goes for the NK cells we met in Chapter 2, which are looking for cells that have lost their MHC display — a trick that some early cancers try, and that NK cells respond to with the same cold logic they apply to any cell that has stopped wearing its badge.
The cancers that go on to become real diseases are, in this picture, the ones that have managed to slip past these early defenses. Some of them have so few mutations that there is nothing distinctive for the immune system to see. Some of them have actively evolved ways to suppress the immune cells around them. Some of them, despite the efforts of the immune system, simply grow faster than they can be eliminated. The cancer, when it eventually appears as a clinical disease, is what is left after a long and mostly invisible war that, for that particular patient, the immune system has lost.