The Laws of Medicine (5 page)

In retrospect, we now know that the behavior of the mothers of autistic children was not the cause of autism; it was the effect—an emotional response to a child who produces virtually no emotional response. There are, in short, no refrigerator moms. There are only neurodevelopmental pathways that, lacking appropriate signals and molecules, have gone cold.

....

The moral and medical lessons from this story are even more relevant today. Medicine is in the midst of a vast reorganization of fundamental principles. Most of our models of illness are hybrid models; past knowledge is mishmashed with present knowledge. These hybrid models produce the illusion of a systematic understanding of a disease—but the understanding is, in fact, incomplete. Everything seems to work spectacularly, until one planet begins to move backward on the horizon. We have invented many rules to understand normalcy—but we still lack a deeper, more unified understanding of physiology and pathology.

This is true for even for the most common and extensively studied diseases—cancer, heart disease, and diabetes. If cancer is a disease in which genes that control cell division are mutated, thus causing unbridled cellular growth, then why do the most exquisitely targeted inhibitors of cell division fail to cure most cancers? If type 2 diabetes results from the insensitivity of tissues to insulin signaling, then why does adding extra insulin reverse many, but not all, features of the disease? Why do certain autoimmune diseases cluster together in some people, while others have only one variant? Why do patients with some neurological diseases, such as Parkinson's disease, have a reduced risk of cancer? These “outlying” questions are the Mars problems of medicine: they point to systematic flaws in our understanding, and therefore to potentially new ways of organizing the cosmos.

Every outlier represents an opportunity to refine our understanding of illness. In 2009, a young cancer scientist named David Solit in New York set off on a research project that, at first
glance, might seem like a young scientist's folly. It is a long-established fact in the world of cancer pharmacology that nine out of ten drugs in clinical development are doomed to fail. In pharmaceutical lingo, this phenomenon is called the valley of death: a new drug moves smoothly along in its early phase of clinical development, seemingly achieving all its scientific milestones, yet it inevitably falters and dies during an actual clinical trial. In some cases, a trial has to be stopped because of unanticipated toxicities. In other cases, the medicine provokes no response, or the response is not durable. Occasionally, a trial shows a striking response, but it is unpredictable and fleetingly rare. Only 1 woman in a trial of 1,000 women might experience a near complete disappearance of all the metastatic lesions of breast cancer—while 999 women experience no response. One patient with widely spread melanoma might live for fifteen years, while the rest of the cohort has died by the seventh month of the trial.

The trouble with such “exceptional responders,” as Solit called them, was that they had traditionally been ignored, brushed off as random variations, attributed to errors in diagnosis or ascribed, simply, to extraordinary good fortune. The catchphrase attached to these case histories carried the stamp of ultimate scientific damnation:
single patient anecdotes
(of all words, scientists find the word
anecdote
particularly poisonous since it refers to a subjective memory). Medical journals have long refused to publish these reports. At scientific conferences when such cases were described, researchers generally rolled their eyes and avoided the topic. When the trials ended, these
responders were formally annotated as “outliers,” and the drug was quietly binned.

But Solit wanted to understand these rare responses. These “exceptional responders,” he reasoned, might have some peculiar combination of factors—genes, behaviors, risk factors, environmental exposures—that had made them respond so briskly and durably. He decided to use the latest medical tools to understand their responses as deeply and comprehensively as possible. He had inverted a paradigm: rather than spending an enormous effort trying to figure out why a drug had commonly failed, as most of his colleagues might have, he would try to understand why it had occasionally succeeded. He would try to map the landscape of the valley of death—not by querying all those who had fallen into it, but by asking the one or two patients who had clambered out.

In 2012, Solit's team published the first analysis of one such trial. Forty-four patients with advanced bladder cancer had been treated with a new drug called everolimus. The results had been uniformly disappointing. Some tumors may have shrunk a little, but none of the patients had showed a striking response. Then, in mid-April 2010, there was patient 45—a seventy-three-year-old woman with tumors filling her entire abdomen and invading her kidneys and lymph nodes. She started the medicine that month. Within weeks, her tumors had begun to involute. The mass invading the kidney necrosed and disappeared. Fifteen months later, when her CAT scans were checked again, her doctors had to squint hard to see any visible signs of tumor in her abdomen.

Solit focused on just that case. Reasoning that genes were likely involved, he pulled out patient 45's tumor sample from the freezer and sequenced every gene to find the ones that were mutated (in most human cancers, between 10 to 150 genes can be mutated). The woman's tumor had 140 mutations. Of all those, two stood out: one in a gene named TSC1 and another in a gene named NF2. Both these genes had been suspected to modulate the response to everolimus, but before Solit, no one had found formal proof of the link in human patients.

But this was still a “single patient anecdote”; scientists would still roll their eyes. Solit's team now returned to the original trial and sequenced the same genes in the larger cohort of patients. A pattern emerged immediately. Four other patients who had mutations in the TSC1 gene had shown modest responses, while none of the other patients, with mutations in other genes but not in TSC1, had shown even a sliver of a response. Via just one variable—the mutation in the TSC1 gene—you could segregate the trial into moderate or strong responders versus nonresponders. “Single patient anecdotes are often dismissed,” Solit wrote. But here, exactly such an anecdote had turned out to be a portal to a new scientific direction. In a future trial, a cohort of patients might be sequenced
up front
, and only those with mutations in the TSC1 gene might be treated with the drug. Perhaps more important, the relationship between the gene and the susceptibility of the tumor cells opened a new series of scientific investigations into the mechanism for this selective vulnerability, leading to yet new trials and novel drugs.

But is it a
law
of medicine that such outliers will provide
the most informative pieces of data in our attempt to revamp the core of our discipline? In Lewis Thomas's time, such a law would have made no sense: there was nothing to “outlie.” The range of medical and surgical interventions was so severely limited that any assessment of variations in response was useless; if every patient with heart failure was destined to die, then it made little sense discriminating one from another (and even if some survived long term, no tools existed to investigate them). But this is precisely what has changed: pieces of data that do not fit our current models of illness have become especially important not only because we are reassessing the nature of our knowledge, but also because we are generating more such pieces of data every day. Think about the vast range of medicines and surgical procedures not as therapeutic interventions but as investigational probes. Think of every drug as a chemical tool—a molecular scalpel—that perturbs human physiology. Aspirin flicks off a switch in the inflammatory system. Lipitor tightens a screw on cholesterol metabolism. The more such investigational probes we use, the more likely we are to alter physiology. And the more we alter physiology, the more we will find variations in response, and thereby discover its hidden, inner logic.

....

One morning in the spring of 2015, I led a group of medical students at Columbia University on what I called “outlier rounds.” We were hunting for variant responses to wound healing. Most patients with surgical incisions heal their wounds in a week. But what about the few patients whose wounds don't heal? We moved from room to room across the hospital, trying to find cases where postsurgical wounds had failed to heal. Most of these were predictable—elderly patients with complex surgical incisions, or diabetics, who are known to heal poorly. But after about nine such cases, we entered the room of a young woman recovering from an abdominal procedure whose incision was still raw and unhealed. The students looked puzzled. Nothing about this woman, or her incision, seemed any different from the hundreds of others that had healed perfectly. After a long pause they began to ask questions. One of them asked about her family history: Had anyone else in her family had a similar experience? Another wondered if he might swab the tissue to check for unusual, indolent infections. The orthodox models of wound healing were coming apart at the seams, I suspected, and a novel way of thinking about an old problem was being born.

We have spent much of our time in medicine dissecting and understanding what we might call the “inlier” problem. By “inliers,” I am referring to the range of normalcy; we have compiled a vast catalog of normal physiological parameters: blood pressure, height, body mass, metabolic rate. Even pathological states are described in terms that have been borrowed from normalcy: there is an average diabetic, a typical case of heart failure, and a standard responder to cancer chemotherapy.

But we have little understanding of what makes an individual lie outside the normal range. “Inliers” allow us to create rules—but “outliers” act as portals to understand deeper laws. The standard formula—height (in cms) − 100 = average weight plus 10 percent (in kgs)—is a rule that works for most of the human population. But it takes a single encounter with a person with genetic dwarfism to know that there are genes that control this relationship and that mutations can disrupt it quite acutely.

In his 1934 book,
The Logic of Scientific Discovery
, the philosopher Karl Popper proposed a crucial criterion for distinguishing a scientific system from an unscientific one. The fundamental feature of a scientific system, Popper argued, is not that its propositions are verifiable, but that its propositions are
falsifiable
—i.e., every theory carries an inherent possibility of proving it false. A theory or proposition can only be judged “scientific” if it carries within it a prediction or observation that will prove it false. Theories that fail to generate such “falsifiable” conjectures are not scientific. If medicine is to become a bona fide science, then we will have to take up every opportunity to falsify its models, so that they can be replaced by new ones.

....

For every perfect medical experiment, there is a perfect human bias.

I
n the summer of 2003, I finished my three-year residency in internal medicine and began a fellowship in oncology. It was an exhilarating time. The Human Genome Project had laid the foundation for the new science of genomics—the study of the entire genome. Although frequent criticism of the project appeared in the media—it had not lived up to its promises, some complained—it was nothing short of a windfall for cancer biology. Cancer is a genetic disease, an illness caused by mutations in genes. Until that time, most scientists had examined cancer cells one gene at a time. With the advent of new technologies to examine thousands of genes in parallel, the true complexity of cancers was becoming evident. The human genome has about twenty-four thousand genes in total. In some cancers, up to a hundred and twenty genes were altered—one in every two hundred genes—while in others, only two or three genes were mutated. (Why do some cancers carry such complexity, while others are genetically simpler? Even the questions—not just the answers—thrown up by the genome-sequencing project were unexpected.)

More important, the capacity to examine thousands of genes in parallel, without making any presuppositions about the mutant genes, allowed researchers to find novel, previously unknown genetic associations with cancer. Some of the newly discovered mutations in cancer were truly unexpected: the genes did not control growth directly, but affected the metabolism of nutrients or chemical modifications of DNA. The transformation has been likened to the difference between measuring one point in space versus looking at an entire
landscape—but it was more. Looking at cancer before genome sequencing was looking at the known unknown. With genome sequencing at hand, it was like encountering the unknown unknown.

Much of the excitement around the discovery of these genes was driven by the idea that these could open new vistas for cancer treatment. If cancer cells were dependent on the mutant genes for their survival or growth—“addicted” to the mutations, as biologists liked to describe it—then targeting these addictions with specific molecules might force cancer cells to die. The battle-ax chemical poisons of cellular growth would become obsolete at last. The most spectacular example of one such drug, Gleevec, for a variant of leukemia, had galvanized the entire field. I still recall the first patient whom I treated with Gleevec, a fifty-six-year-old man whose bone marrow had been so eaten by leukemia that he had virtually no platelets left and would bleed profusely from every biopsy that we performed. A fellow had to meet Mr. K with a brick-size pack of sterile gauze pads in the exam room and press on his biopsy site for half an hour to prevent bleeding. About four weeks after he started treatment with Gleevec, it was my turn to perform his biopsy. I came prepared with the requisite armfuls of gauze, dreading the half-hour ordeal—except when I withdrew the needle, the wound stopped bleeding by itself. Through that nick of the skin, its edges furling with a normal-looking clot, I could see the birth of a revolution in cancer treatment.