The Emperor of All Maladies

Book notes for “The Emperor of All Maladies”, by Siddhartha Mukherjee

Last annotated on February 3, 2017

New drugs appeared at an astonishing rate: by 1950, more than half the medicines in common medical use had been unknown merely a decade earlier. location 461

Longevity, although certainly the most important contributor to the prevalence of cancer in the early twentieth century, is probably not the only contributor. Our capacity to detect cancer earlier and earlier, and to attribute deaths accurately to it, has also dramatically increased in the last century. The death of a child with leukemia in the 1850s would have been attributed to an abscess or infection (or, as Bennett would have it, to a “suppuration of blood”). And surgery, biopsy, and autopsy techniques have further sharpened our ability to diagnose cancer. The introduction of mammography to detect breast cancer early in its course sharply increased its incidence—a seemingly paradoxical result that makes perfect sense when we realize that the X-rays allow earlier tumors to be diagnosed. location 924

A model is a lie that helps you see the truth. —Howard Skipper location 2695

Skipper learned that he could halt this effusive cell division by administering chemotherapy to the leukemia-engrafted mouse. By charting the life and death of leukemia cells as they responded to drugs in these mice, Skipper emerged with two pivotal findings341. First, he found that chemotherapy typically killed a fixed percentage of cells at any given instance no matter what the total number of cancer cells was. This percentage was a unique, cardinal number particular to every drug. In other words, if you started off with 100,000 leukemia cells in a mouse and administered a drug that killed 99 percent of those cells in a single round, then every round would kill cells in a fractional manner, resulting in fewer and fewer cells after every round of chemotherapy: 100,000 . . . 1,000 . . . 10 . . . and so forth, until the number finally fell to zero after four rounds. Killing leukemia was an iterative process, like halving a monster’s body, then halving the half, and halving the remnant half. location 2735

Imagine now that in one of those villages, a new, highly specific test for cancer is introduced—say the level of a protein Preventin in the blood as a marker for cancer. Suppose Preventin is a perfect detection test. Preventin “positive” men and women are thus immediately counted among those who have cancer. Preventin, let us further suppose, is an exquisitely sensitive test and reveals very early cancer. Soon after its introduction, the average age of cancer diagnosis in village 1 thus shifts from seventy years to sixty years, because earlier and earlier cancer is being caught by this incredible new test. However, since no therapeutic intervention is available even after the introduction of Preventin tests, the average age of death remains identical in both villages. To a naive observer, the scenario might produce a strange effect. In village 1, where Preventin screening is active, cancer is now detected at age sixty and patients die at age eighty—i.e., there is a twenty-year survival. In village 2, without Preventin screening, cancer is detected at age seventy and patients die at age eighty—i.e., a ten-year survival. Yet the “increased” survival cannot be real. How can Preventin, by its mere existence, have increased survival without any therapeutic intervention? The answer is immediately obvious: the increase in survival is, of course, an artifact. Survival rates seem to increase, although what has really increased is the time from diagnosis to death because of a screening test. A simple way to avoid this bias is to not measure survival rates, but overall mortality. (In the example above, mortality remains unchanged, even after the introduction of the test for earlier diagnosis.) But here, too, there are profound methodological glitches. “Cancer-related death” is a raw number in a cancer registry, a statistic that arises from the diagnosis entered by a physician when pronouncing a patient dead. The problem with comparing that raw number over long stretches of time is that the American population (like any) is gradually aging overall, and the rate of cancer-related mortality naturally increases with it. Old age inevitably drags cancer with it, like flotsam on a tide. A nation with a larger fraction of older citizens will seem more cancer-ridden than a nation with younger citizens, even if actual cancer mortality has not changed. location 4461

To compare samples over time, some means is needed to normalize two populations to the same standard—in effect, by statistically “shrinking” one into another. This brings us to the crux of the innovation in Bailar’s analysis: to achieve this scaling, he used a particularly effective form of normalization called age-adjustment. To understand age-adjustment, imagine two very different populations. One population is markedly skewed toward young men and women. The second population is skewed toward older men and women. If one measures the “raw” cancer deaths, the older-skewed population obviously has more cancer deaths. Now imagine normalizing the second population such that this age skew is eliminated. The first population is kept as a reference. The second population is adjusted: the age-skew is eliminated and the death rate shrunk proportionally as well. Both populations now contain identical age-adjusted populations of older and younger men, and the death rate, adjusted accordingly, yields identical cancer-specific death rates. Bailar performed this exercise repeatedly over dozens of years: he divided the population for every year into age cohorts—20–29 years, 30–39 years, 40–49, and so forth—then used the population distribution from 1980 (chosen arbitrarily as a standard) to convert the population distributions for all other years into the same distribution. Cancer rates were adjusted accordingly. Once all the distributions were fitted into the same standard demographic, the populations could be studied and compared over time. location 4479

In 1870, the per capita consumption in America615 was less than one cigarette per year. A mere thirty years later, Americans616 were consuming 3.5 billion cigarettes and 6 billion cigars every year. By 1953, the average annual consumption of cigarettes had reached thirty-five hundred per person. On average, an adult American smoked ten cigarettes617 every day, an average Englishman twelve, and a Scotsman nearly twenty. location 4666

“By the early 1940s, asking about a connection619 between tobacco and cancer was like asking about an association between sitting and cancer.” If nearly all men smoked, and only some of them developed cancer, then how might one tease apart the statistical link between one and the other? location 4684

It remains an astonishing, disturbing fact that in America—a nation where nearly every new drug is subjected to rigorous scrutiny as a potential carcinogen, and even the bare hint of a substance’s link to cancer ignites a firestorm of public hysteria and media anxiety—one of the most potent and common carcinogens known to humans can be freely bought and sold at every corner store for a few dollars. location 5455

Notably, the average age of diagnosis of women with such preinvasive lesions was about twenty years lower than the average age of women with invasive lesions—once again corroborating the long march of carcinogenesis. The Pap smear had, in effect, pushed the clock of cancer detection forward by nearly two decades, and changed the spectrum of cervical cancer from predominantly incurable to predominantly curable. location 5727

By the early 1990s, cancer biologists could begin to model the genesis of cancer in terms of molecular changes in genes. To understand that model, let us begin with a normal cell, say a lung cell that resides in the left lung of a forty-year-old fire-safety-equipment installer. One morning in 1968, a minute sliver of asbestos from his equipment wafts through the air and lodges in the vicinity of that cell. His body reacts to the sliver with an inflammation. The cells around the sliver begin to divide furiously, like a minuscule wound trying to heal, and a small clump of cells derived from the original cell arises at the site. In one cell in that clump an accidental mutation occurs in the ras gene. The mutation creates an activated version of ras. The cell containing the mutant gene is driven to grow more swiftly than its neighbors and creates a clump within the original clump of cells. It is not yet a cancer cell, but a cell in which uncontrolled cell division has partly been unleashed—cancer’s primordial ancestor. A decade passes. The small collection of ras-mutant cells continues to proliferate, unnoticed, in the far periphery of the lung. The man smokes cigarettes, and a carcinogenic chemical in tar reaches the periphery of the lung and collides with the clump of ras-mutated cells. A cell in this clump acquires a second mutation in its genes, activating a second oncogene. Another decade passes. Yet another cell in that secondary mass of cells is caught in the path of an errant X-ray and acquires yet another mutation, this time inactivating a tumor suppressor gene. This mutation has little effect since the cell possesses a second copy of that gene. But in the next year, another mutation inactivates the second copy of the tumor suppressor gene, creating a cell that possesses two activated oncogenes and an inactive tumor suppressor gene. Now a fatal march is on; an unraveling begins. The cells, now with four mutations, begin to outgrow their brethren. As the cells grow, they acquire additional mutations and they activate pathways, resulting in cells even further adapted for growth and survival. One mutation in the tumor allows it to incite blood vessels to grow; another mutation within this blood-nourished tumor allows the tumor to survive even in areas of the body with low oxygen. Mutant cells beget cells beget cells. A gene that increases the mobility of the cells is activated in a cell. This cell, having acquired motility, can migrate through the lung tissue and enter the bloodstream. A descendant of this mobile cancer cell acquires the capacity to survive in the bone. This cell, having migrated through the blood, reaches the outer edge of the pelvis, where it begins yet another cycle of survival, selection, and colonization. It represents the first metastasis of a tumor that originated in the lung. The man is occasionally short of breath. He feels a tingle of pain in the periphery of his lung. Occasionally, he senses something moving under his rib cage when he walks. Another year passes, and the sensations accelerate. The man visits a physician and a CT scan is performed, revealing a rindlike mass wrapped around a bronchus in the lung. A biopsy reveals lung cancer. A surgeon examines the man and the CT scan of the chest and deems the cancer inoperable. Three weeks after that visit, the man returns to the medical clinic complaining of pain in his ribs and his hips. A bone scan reveals metastasis to the pelvis and the ribs. Intravenous chemotherapy is initiated. The cells in the lung tumor respond. The man soldiers through a punishing regimen of multiple cell-killing drugs. But during the treatment, one cell in the tumor acquires yet another mutation that makes it resistant to the drug used to treat the cancer. Seven months after his initial diagnosis, the tumor relapses all over the body—in the lungs, the bones, the liver. On the morning of October, 17, 2004, deeply narcotized on opiates in a hospital bed in Boston and surrounded by his wife and his children, the man dies of metastatic lung cancer location 7597

In 2005, an avalanche of papers951 cascading through the scientific literature converged on a remarkably consistent message—the national physiognomy of cancer had subtly but fundamentally changed. The mortality for nearly every major952 form of cancer—lung, breast, colon, and prostate—had continuously dropped for fifteen straight years. There had been no single, drastic turn but rather a steady and powerful attrition: mortality had declined by about 1 percent953 every year. The rate might sound modest, but its cumulative effect was remarkable: between 1990 and 2005, the cancer-specific954 death rate had dropped nearly 15 percent, a decline unprecedented in the history of the disease. location 7801

The study, reported in 2006, appeared initially to confirm an increased risk of right-sided brain cancers in men and women who held their phone on their right ear. But when researchers evaluated the data meticulously, a puzzling pattern emerged: right-sided cell phone use reduced the risk of left-sided brain cancer. The simplest logical explanation for this phenomenon was “recall bias”: patients diagnosed with tumors unconsciously exaggerated the use of cell phones on the same side of their head, and selectively forgot the use on the other side. When the authors corrected for this bias, there was no detectable association between gliomas and cell phone use overall. location 8676

as the fraction of those affected by cancer creeps1059 inexorably in some nations from one in four to one in three to one in two, cancer will, indeed, be the new normal—an inevitability. The question then will not be if we will encounter this immortal illness in our lives, but when. Atossa location 8923