Cancer, Our Genes, and the Anxiety of Risk-Based Medicine
Advanced medical technologies make it easier to identify people at risk for cancer, but there are risks involved in oversurveillance too.
Siddhartha Mukherjee
In the summer of 2005 I met a woman, Laura M., whose life had been overturned by cancer. But she was disease free: It was the anxiety of cancer in the future that haunted her.
Laura had been diagnosed with a primary tumor in her breast that was small and localized. She had had surgery, radiation, and chemotherapy—a standard treatment protocol—and had then come to see me, an oncologist, to help manage her future care. I suggested doing nothing. Everything about her case suggested a good prognosis: We all agreed that she had likely been cured. But in the wake of her treatment, she became obsessed by the possibility of a relapse. She scoured her family history and discovered a distant aunt who had died of breast cancer at age 70. Her own mother had died at a young age from a car accident, but Laura became convinced that had her mother lived, she would have been diagnosed with breast cancer.
Laura’s visits to the clinic were punctuated by her sense of doom. She often came in with sheaves of papers that she had printed from the internet about “occult metastasis,” which had been found in patients who, like her, were thought to be at low risk. She repeatedly asked me to confirm that she “had been given the most aggressive chemo regimen that could be given.” (She had, in fact, been treated with the appropriate regimen recommended for her case.) We checked her for genetic susceptibilities, such as inherited mutations in the BRCA1 and BRCA2 genes that increase cancer risk, but found none. Nonetheless, she asked if she and her daughter could undergo “the most intensive form of cancer surveillance” to detect early cancers in her body.
Laura’s story highlights a new anxiety about illness that is permeating our culture. It is the anxiety of being under constant diagnostic surveillance, of inhabiting a state of vigilant watchfulness for illnesses before they can take root in one’s body. It is the state, as one patient described it, of “feeling under siege from the future.” Emblematic of this anxiety is the concept of a “previvor”—a strange new term invented to describe a person who is a survivor of an illness that she is predisposed to but has yet to have. For Laura, these states were contiguous: her survivorship from one breast cancer had turned overnight, it seemed, into “previvorship” for another breast cancer.
As I write this, two kinds of technologies are radically altering the landscape of cancer risk and screening. The first involves what we might call “genetic surveillance”—an attempt to quantify an individual’s inherited predisposition for cancer (that is, you should be surveyed for the disease because of the higher risk conferred by your genes). The second, in contrast, involves “chemical surveillance”—an attempt to detect chemical markers of incipient cancers in blood (that is, you should be surveyed because there’s a sign of early cancer circulating in your blood). The two technologies converge to increase the supply of men and women who are forced to enter the domain of surveillance and screening. Both, in short, encourage men and women without current cancer, but with the prospect of future cancer, to become citizens or permanent residents of what Susan Sontag once described as “the kingdom of the ill.”
For decades, perhaps centuries, we’ve known of families where some form of cancer (usually the same type: breast or pancreatic) is manifest in multiple individuals across multiple generations. Not every family member is affected, but the risk of cancer in such a family clearly lies beyond the average risk of cancer in the population.
Until recently, our capacity to identify the culprit genes in such families—or, more actionably, to identify the members of the family who carried the heightened risk—was limited to inherited single-gene mutations. These included mutations in genes such as BRCA1, BRCA2, and MLH1 that, if inherited from parents, increase the likelihood of breast, colon, and other cancers by severalfold over normal individuals. But many human traits, including cancer risk, might not track with single-gene mutations. Take human height as an example. Height is highly heritable—we know that tall parents tend to produce tall children, and shorter parents bear shorter children—yet early attempts to pin down the variation in human height to single-gene variations or mutations revealed only a smattering of candidate genes. (For this essay, I use the terms “variation” and “mutation” interchangeably, although there are subtle differences.) Geneticists described this conundrum, famously, as the “missing heritability” of height: we could infer from the pattern of inheritance that height-determining genes must exist in the human genome, but their precise identity and number remained unknown.
By similar logic, the inherited risk of cancer might be carried by mutations or variations not in one but in multiple genes, each of which acts together to increase an individual’s risk of the disease. In the 1990s, some breast cancer patients began to refer to the next as-yet-unidentified conglomerate of genes for breast cancer as “BRCA3.” That name carried both a sardonic and a hopeful edge. Unlike patients with definite BRCA1 and BRCA2 mutations, patients with potential BRCA3 mutations remained suspended in an anxious limbo. We could not diagnose a woman with this genetic syndrome yet because we had no idea what these genes might be (a few additional single-gene mutations that increased breast cancer risk were identified in the 2000s, but most patients with breast cancer continued to lack a single-gene explanation). A BRCA3 patient’s experience of her terrifying family history and dread of future disease were just as acute as those of a patient with known cancer-risk mutations, but the genes that precipitated the former patient’s fate were hidden from our view. As doctors, we’d acknowledge the risk that these patients carried—with their family histories scarred by breast cancer—but we were unable to offer a more tangible description of their susceptibility to the disease.
This state of suspension for polygenic (or multigene) diseases is finally being relieved: the combinations of gene mutations responsible for such genetically complex diseases are now being identified by powerful computational technologies. Computational algorithms (some involving deep learning or machine learning), in particular, have been unleashed on human genomes. By scanning millions of fully sequenced genomes, these algorithms dissect how variations in thousands of genes, each exerting a small effect, might ultimately add up to the heightened risk of an illness—a problem of such mind-boggling complexity that previous computational technologies had failed to capture it. One algorithm has learned to predict human height as the consequence of variations in a thousand-odd genes. (Take a moment to digest this startling fact: such an algorithm might soon predict your actual height, or the future height of your unborn child, based on your genetic sequence alone.) Another program is learning to predict the risk of cardiovascular disease—again, likely the consequence of hundreds of gene mutations or variations. With such advances, it is likely that an algorithm might identify those of us at highest genetic risk for future cancers. (Of course, for many cancers, even ones that run in families, there’s still a powerful influence of chance and the environment. A woman with a BRCA1 mutation might increase her risk for breast cancer through certain exposures, by virtue of inheriting other “modifier” genes or by chance alone. These additional variables are not yet part of the computational landscape but could become incorporated into algorithms in the future.) This technology, then, could serve as a portal of entry into the world of cancer for potentially millions of men and women who seek to be annotated for future cancer risk and potentially surveyed for cancer.
While computers seek out patients who have an inherited susceptibility to cancer, other machines are seeking to identify chemicals that might currently be in our blood or other organs that signal cancer risk. Termed “liquid biopsy” or “liquid surveillance,” these methods attempt to discover minuscule amounts of the products shed or spilled by cancer cells—DNA, proteins, and other substances—into the blood or other circulating tissues. Once such trace signs of an incipient cancer are found, the logic runs, cancer will be detected in its earliest stages and can be attacked with more effective therapies. We will scour the body to find ovarian, lung, and prostate cancers, for instance, before these become clinically manifest, thereby enabling better treatments.
These liquid biopsies run the risk of overdiagnosing patients, however. What if someone is found to carry a liquid marker for ovarian cancer, say, but that ovarian cancer never takes root in her body? Cancer cells, we now know, can exist in a body, or a site within the body, without becoming manifest as clinical disease or a detectable metastasis. (Most likely, this is because the “soil” of a particular organ does not allow the “seed” of a cancer to sprout.) Or what if some of the markers turn out to overlap with benign diseases (as was the case with earlier liquid surveillance markers, such as the prostate-specific antigen test for prostate cancer), thereby increasing the risk of false positive results?
Nonetheless, enthusiasm for the liquid surveillance of cancer seems to grow exponentially each day (one private company that hopes to advance this technology goes by the name Grail, emblematic of the near-religious fervor with which some advocates describe the power of liquid biopsies). Many patients in my cancer clinic now come to their appointments armed with brochures about liquid biopsies, wondering whether their tumors might have been detected earlier had such biopsies been performed. These technologies represent a second portal of entry into the world of cancer. By identifying men and women who might be bearing the first markers of cancer, these methods increase the pool of those who must be surveyed and further screened for the illness.
My aim is to neither exaggerate nor minimize the transformative potential of these technologies, although it’s worthwhile emphasizing this at the outset. The capacity to identify humans with an increased genetic risk for cancer, coupled with the possibility of detecting cancer at its earliest stages using a liquid biopsy, might radically change how we prevent, detect, and treat cancer. But my concern is the effect that such surveillance might have on our bodies and societies. In the 1950s the sociologist Erving Goffman wrote a remarkable article about the concept of a “total institution,” an idea that he expanded in subsequent work. “A total institution,” Goffman wrote in his 1961 book Asylums, is one “where a great number of similarly situated people, cut off from the wider community for a considerable time, together lead an enclosed, formally administered round of life.” Total institutions, such as mental hospitals, prisons, and even boarding schools, have rituals of entry and exit. They inculcate belonging. They invent their own vocabulary and codes of behavior; they have an internal logic, impenetrable to others. They encourage surveillance and create anxiety. Members are united by a common sense of purpose, by the feeling of being chosen or marked. Those who are expelled may feel a sense of betrayal, while those who remain can be consumed by the guilt of survivorship.
Cancer, too, runs the risk of becoming a “total institution.” A patient, once diagnosed, may be whisked away into a cancer ward, dressed in a patient’s smock—“a tragicomically cruel costume, no less blighting than a prisoner’s jumpsuit,” as I wrote in The Emperor of All Maladies—and stripped of his identity. When I once asked a woman with a rare sarcoma about her life outside the hospital, she observed, “I am in the hospital even when I am outside the hospital.”
In this new era of cancer treatment, I wonder if we are unwittingly, but insidiously, intensifying the totality of the “cancer institution” for patients. For people like Laura M., cancer has certainly become a total institution—or a “cancer world,” as some patients call it. They are in either treatment, remission, surveillance, maintenance, or resurveillance. Mavens of early detection are also working on deep-learning algorithms that will pick up cancerous lesions on patients’ imaging results and classify them as malignant, using criteria that seem to defy even the most acute human eye. In an April 2017 article in the New Yorker, I wrote about one of the pioneers of this idea, the German computer scientist Sebastian Thrun. Thrun imagines a world in which even the daily instruments of our normal lives are morphed into weapons of diagnostic surveillance—a bathtub that scans your body to detect abnormal masses that might require investigation; a mirror that could check your body for precancerous moles; a computer program that (with your consent) would scour your Instagram or Facebook page while you slept at night, evaluating changes in your photographs that might signal signs of cancer.
Then there’s the question of treatment and cost. If an additional tumor—clinically undetectable, but discovered by these novel methods—were detected in Laura M.’s case and the primary lesion removed, what criteria would we use to determine whether we should use some form of adjuvant (or extra) medicine, such as cell-killing chemotherapy or targeted therapy, after the initial surgical removal, as is often done for most cancers? The costs of such surveillance and treatment—an astronomical amount if every human had to be genetically annotated, subjected to surveillance, and treated if a tumor was found—would overwhelm current projections of medical costs (although in the most optimistic scenario, the benefits would also be amplified in lives saved via early diagnosis). Thorny issues of overdiagnosis and overtreatment would have to be addressed. We would have to devise careful guidelines about when not to act and whom not to treat.
For Laura M., the answer to each of these questions carries immense consequences. She has entered a strange new world, one of constant diagnostic surveillance; of dealing with the anxiety of relapse and maintenance; of that peculiar desolation of the shuttle from clinical trial to clinical trial, and from hospital to hospital, as she tries to keep one step ahead in this chess game against cancer; and of watching doctors pit their will, wit, and imagination against a formidable enemy that keeps changing its shape. This world has created its own internal vocabulary. A “haircut party” is a celebration thrown in honor of a person about to enter the cancer world (as a sign of solidarity, even if the patient is spared hair-loss inducing chemo). “No Exit chemo,” as a patient of mine put it, describes the fact that a unique personalized chemo regimen for a patient produces unique toxicities, a phrase borrowed from the Jean-Paul Sartre novel in which every human being is assigned his or her own personal hell.
“A world in which cancer is normalized as a manageable chronic condition would be a wonderful thing,” the medical historian Steven Shapin wrote in a 2010 review of The Emperor of All Maladies. “But a risk-factor world in which we all think of ourselves as precancerous would not,” he continued. “It might decrease the incidence of some forms of malignancy while hugely increasing the numbers of healthy people under medical treatment. It would be a strange victory in which the price to be paid for checking the spread of cancer through the body is its uncontrolled spread through the culture.”
To date, Laura M. has not suffered from a relapse of breast cancer. Nor, fortunately, has she had a new cancer anywhere in her body. But the “strange victory” over her body has not spared her mind. She remains haunted by the future prospect of illness. When Susan Sontag wrote of a passport between the kingdom of the well and the kingdom of the ill, she imagined a bidirectional passage: men and women might pass into illness, but some would return to wellness. In inventing cancer’s new surveillance culture, I fear that we have closed the borders of the kingdoms. I fear that we now possess just one-way passports into the realm of illness.
Volume 37, Number 5. May 2018.
Editor’s Note: The patient’s name and certain identifying details were changed by the author to protect the patient’s privacy.
Beating a Cancer Death Sentence
A melanoma survivor describes his arduous—and costly—path to successful treatment and hope.
Jonathan Friedlaender
Several months ago, my wife, Françoise, and I attended something novel for melanoma patients: a survivors’ dinner. People said they wanted to make it an annual gathering. Planning anything that far in advance had been pointless for me. Two years ago, I was about to accept hospice care.
When I was first diagnosed in 1996, early surgery was the only reliably successful treatment. Anything more advanced was essentially a death sentence. Over the past five years, a series of revolutionary drugs have given me and many other people a surprisingly hopeful prospect. Nevertheless, the drugs’ development process has often been excruciating for clinical trial participants, and their remarkably high costs limit their value.
I have the most common form of melanoma, which occurs in fair-complexioned people who had blistering sunburns in their youth. I also spent a year in my 20s in the South Pacific, doing biological-anthropological fieldwork, which meant more episodes of particularly intense sun exposure.
My first melanoma lesion revealed itself three decades later—in 1996, when I was 56—as a small irregular raised blue-gray lump above my knee. When I showed it to a dermatologist, she unceremoniously told me to take off my trousers and lie down on her examination table. As she numbed the area and began to cut, I put my hands over my face. She said, “Am I hurting you?” No, I said. It was the shock. I knew nothing about melanoma except that it was particularly aggressive and lethal.
Follow-up surgery to remove more surrounding skin and to biopsy a sentinel lymph node in my upper thigh (the most likely node to which cells from my primary tumor would spread first) revealed no further evidence of disease. The tumor’s depth suggested that my prospects were alarming enough: there was a 20 percent chance of recurrence.
I was not yet ready to die, and I began to make decisions about things I had avoided or neglected. I resolved to get divorced, something my then wife and I had been considering for years. The stress of cancer is like a gale: it drives distant couples further apart and compels already close ones to cling more tightly together.
In the next few years, I had periodic skin examinations and chest X-rays. Everything looked normal. Of course, anxiety lingered.
By 2000 I had gotten divorced and met Françoise, a theoretical chemist, shortly after we had both entered the new world of online dating. We became a devoted couple over the next three years.
As I got up from bed one night, I felt a kumquat-size lump in the same lymph node bed that had been biopsied seven years before. I sat back down heavily and whispered a curse, waking Françoise. She wondered what was wrong—a nightmare, perhaps? I knew it was far worse.
The lump was confirmed as melanoma a few days later. We were stunned. Statistics suggested that my five-year survival chances had just plummeted to 20 percent.
Everything suddenly became uncertain and threatening. My focus changed. Concerns about world affairs, money, and social and professional status all receded. After a few days, Françoise said, “Jonathan, you have to take early retirement.” I replied, “And we have to get married.” Again, delay made no sense. We quickly married a few days before I had surgery to remove the entire lymph node bed.
For the next five years, my disease continued to spread very slowly. One oncologist said, “Either your immune system is very smart or your tumor is just very stupid.” Nevertheless, I underwent two deeper surgeries in my lower right abdomen and joined two ineffective clinical trials. Another very promising drug trial targeting the common BRAF mutation was irrelevant in my case—tests showed that my tumors lacked that mutation.
By 2008 Françoise and I had retired to rural Connecticut, and I came under the care of Mario Sznol, leader of Yale University’s Melanoma Clinical Research Program. After carefully reviewing my earlier scans and history, he shocked us with the news that the disease had already spread to my lungs. My melanoma had reached stage IV, and I had a median expected survival of eight months.
Dr. Sznol said that chemotherapy or further surgery was pointless; immunotherapy was my best remaining option. However, the two forms available at the time (alpha interferon and interleukin-2) had extremely low success rates. I was treated with both, and they had no apparent impact on my disease. They did, however, cause nasty side effects. Alpha interferon gave me chills and a bad rash. I felt exhausted and spent most of the day in bed. Interleukin-2 disoriented me and caused me to gain 30 pounds within a week. I ended up in the cardiac ward.
Fortunately, the effects were all quickly reversible. And Dr. Sznol said that immunotherapy drugs then in development offered great hope, particularly something called anti-PD-1 (the “PD” stands for programmed death). It had just begun clinical trials a short time before.
Unlike chemotherapy, which simply kills susceptible cells, the goal of immunotherapy is to boost the production of certain white blood cells (the T cells) that can then detect and kill tumors. The old immunotherapy strategy of giving patients high doses of the molecules (such as interleukin-2 and alpha interferon) that normally stimulate T-cell production rarely worked because the immune system has its own brakes, or checkpoints, that prevent the accumulation of very high levels of these molecules. When the levels reach a certain point, the immune system shuts down its own production of the molecules.
In the 1980s, James P. Allison at the University of California, Berkeley, realized that a better strategy was to block or inhibit the checkpoints themselves, taking the normal brakes off the body’s T-cell production. Almost 30 years after his group identified such a major checkpoint in mice (the CTLA-4 protein receptor), its blocking antibody finally gained approval from the Food and Drug Administration (FDA) for use in humans with advanced melanoma in 2011 (Bristol-Myers Squibb’s anti-CTLA-4 antibody, ipilimumab).
Ipilimumab’s journey to the market wasn’t easy. Many melanoma patients anxiously followed the drug’s slow path through clinical trials, which began in 2000. Because the drug acted differently from chemotherapy, it confused researchers and regulators. In many cases, tumors might initially grow rapidly before slowly melting away, and some unexpected side effects such as colitis could be particularly severe—even causing some trial patients to die. Early trials (managed by Medarex) failed to meet expectations, mainly because traditional chemotherapy goals (for example, measurable tumor reduction within a short time) were used for assessment.
The delays were frustrating and demoralizing. For support, Françoise suggested that I join the online forum of the Melanoma International Foundation (MIF). The forum helped patients in their searches for promising clinical trials. Nevertheless, my tumors were spreading, and most of the patients I came to know and care about on the forum died.
In 2010, I and other patients with advanced melanoma were able to get ipilimumab a few months before its FDA approval through an expanded access program. My side effects were manageable, and twelve weeks after I began treatment, scans showed clear signs that my tumors had begun to shrink.
Optimism suddenly returned in a rush—perhaps I would survive after all. However, I was not one of the truly fortunate 10 percent of patients who became completely tumor-free following ipilimumab treatment. After a year and a half, new tumors began to appear in my brain, lungs, intestines, and abdomen. Although I had no painful symptoms yet, the cancer was metastasizing throughout my body.
My hope now centered on another immunotherapy checkpoint inhibitor, the anti-PD-1 drug that Dr. Sznol had mentioned to me years earlier. In 2008, Bristol-Myers Squibb had begun clinical trials for an anti-PD-1 drug that appeared to be more tolerable than ipilimumab and produce better survival rates. The initial clinical trial I tried to join was oversubscribed, and I was excluded from later ones either because I’d had a questionable biopsy for prostate cancer (like so many men my age) or because I’d had previous immunotherapy treatment.
My frustration and despair mounted. I and some other MIF members wrote and phoned Bristol-Myers Squibb executives to ask why the anti-PD-1 drug was not moving more quickly through trials. I also wrote top executives at the FDA and the National Cancer Institute, trying to speed up the drug’s approval process—all to no avail.
My health continued to decline. The growing tumors now began to cause discomfort in my intestines, abdominal muscles, and lungs. I required almost weekly blood transfusions because of internal bleeding. I tried a second and third round of ipilimumab with modest effects and another experimental anti-PD-1 drug from a small Israeli biotechnology company. Nothing worked.
I began to accept my imminent death. I had lived the life I had desired. I had found my life’s companion and confidante. I had repaired some frayed family relations. I had even lived to see a granddaughter born.
In April 2013, Catherine Poole of MIF suggested that I try a novel chemotherapy trial in Nashville, Tennessee. The trial drug was an antibody-drug conjugate, in which two molecules that bind preferentially to melanoma cells are linked to an especially toxic agent, auristatin. This reduced the drug’s toxicity to other cells. I flew to Nashville for treatments periodically during the next 10 months. However, after I’d had a good response to the drug for seven months, the cancer recommenced its relentless advance.
A few months later, Poole told me that an expanded access program for Merck’s competing anti-PD-1 drug (pembrolizumab) had just begun for advanced melanoma patients. I was able to join the trial and flew to the Mayo Clinic in Jacksonville, Florida, twice in May 2014 for infusions. By then I was too weak to walk through airports, and Françoise had to push me in a wheelchair.
I transferred my pembrolizumab treatment to Yale as soon as it was offered there, in June 2014. Tumors were intermittently blocking my small intestine, causing severe cramping and vomiting. An uncomfortable tube was inserted into my stomach via my nose. I was now in and out of emergency departments, taking opiates and hoping for a quick response to the drug. However, the blockages did not resolve themselves.
I was hospitalized, receiving nourishment via IV infusion only, and a second tube was inserted directly into my stomach. I had arrived at death’s door. I had conversations with my family in anticipation of the end. One way or another, I was prepared to move on.
I was offered abdominal surgery to remove the obstructing tumors in my intestines. However, the attending surgeon was hesitant because other tumors would remain and grow. Another attending physician was clearly pessimistic. But Dr. Sznol said that there was a real chance the drug could eliminate the remaining disease after the surgery. Although things might “go the other way,” he said, there was the potential for me to regain a normal life. That was a prospect we couldn’t refuse.
The operation was unexpectedly rough. When the surgical team opened my abdomen, my intestines popped out onto my belly from the accumulated pressure. My intestine had perforated en route to the operating room and was starting to leak stool. If the team had waited a few hours longer, I would have died from an infection. They removed almost a yard of intestine and created an ileostomy—a diversion of my small intestine through an opening in my abdominal wall, bypassing my colon.
I awoke in excruciating pain and was distraught when they told me how the operation had gone. I told the surgical team I didn’t want to live. Everyone suddenly became quiet.
The surgeon leaned over me and said, “Dr. Friedlaender, you will be out of most of the discomfort and off the opiates in a couple of days. The tubes to your stomach will come out, the IV nutrition will stop, and very soon you’ll be able to eat again. In two weeks you should be able to go home as you’ve wanted so badly.” It was also possible that the ileostomy would be reversible in time.
Françoise reminded me that it would take another month to know if the anti-PD-1 drug was working and said that I should not lose hope. This calmed me.
My recovery at home was difficult. I was very weak after almost six weeks in the hospital. Françoise became my constant nurse. I slowly regained my strength and began to take care of myself. My next scans showed that I was responding to the drug, and the prospect of imminent death immediately receded. The doctors were now all smiles. The surgery plus the immunotherapy had come in the very nick of time.
Two years after my crisis, I have one remaining pea-size tumor in my armpit. It will probably continue to shrivel away. I just stopped taking the anti-PD-1 drug last month. If the disease recurs, I could resume immunotherapy treatment or some combination of therapies. When I die, Dr. Sznol says, it will be from something besides this horrible disease that I have come to know so intimately.
During this 20-year journey, I necessarily learned a good deal about drug development and approvals. I also became an FDA patient representative, which taught me even more. The number of new drugs approved by the FDA has been increasing in the past three years, and 51 were approved in 2015—the highest number since 1950.
Immunotherapy drugs are among the greatest recent successes. The approved combination therapy of an anti-PD-1 drug and ipilimumab for advanced melanoma has had “stunning” effects, according to Dr. Harriet Kluger of Yale University. Immunotherapy drugs have been approved or are in trials for over 30 other solid tumor malignancies. Cancers of the kidney, bladder, and lung, as well as refractory Hodgkin’s lymphoma, have all been shown to respond to anti-PD-1 drugs at varying but clinically significant rates.
Although things are beginning to change, the drug approval process has been unnecessarily slow. During my long illness, I felt akin to the HIV/AIDS patients of 30 years ago, when exciting protease inhibitors were in development but not yet widely available. When an especially promising treatment is identified, we have to find acceptable ways to facilitate its approval and allow earlier access to it for people who are critically ill.
Faced with imminent death, informed patients have the right to risk their lives by taking a promising but unproven drug, just as they have the right to decide when to terminate further treatment. Withholding drugs in such situations is unethical and paternalistic, even if it may violate the physician’s Hippocratic oath to “first, do no harm.” There is certainly no excuse for trial designs that have placebo or standard treatment arms without a timely crossover provision for ultimate access to the trial drug.
After a great deal of criticism, the FDA does now recognize that when a drug shows early evidence of superiority, a small and relatively short trial with appropriately designed goals (such as surrogate endpoints instead of mortality rates) could support a provisional approval. This means waiving the randomized double-blind final trial that has been the “gold standard” since the thalidomide scandal over 50 years ago. Such trials have become a cumbersome multibillion-dollar enterprise.
According to the FDA, in fiscal year 2014 two-thirds of the newly approved drugs had been granted some kind of accelerated review. The “breakthrough therapy” designation, initiated in 2012, allowed Merck’s anti-PD-1 drug to go through its clinical trials and win FDA approval in three years (2011–14). This saved the lives of many critically ill patients, including me.
Medicines approved in this way are necessarily monitored closely afterward for undiscovered risks and interactions. This is really the way it should be for all drugs: first achieving provisional approval by satisfying a set of FDA-dictated goals in a restricted population, and then having close postapproval monitoring in a large clinical setting. Parallel efforts to speed up and humanize the drug approval process are also under way in the European Medicines Agency’s “adaptive licensing” pilot project.
I was amazed by the prices of new drugs. The largest pharmaceutical companies are certainly benefiting handsomely. Their net profitability consistently ranks among the highest of the major industrial sectors, according to BBC News, and pharmaceutical research and development spending, while certainly large, is generally less than spending on marketing and advertising, as the World Health Organization has noted. Prices in the United States are particularly high: in a 2004 article in the New York Review of Books, Marcia Angell referred to the United States as the primary prescription drug “profit center.”
Beginning with their testimony in the 1959–62 hearings of the US Senate Subcommittee on Antitrust and Monopoly investigating drug costs, chaired by Sen. Estes Kefauver (D-TN), pharmaceutical companies have justified the high prices of prescription drugs because of their development costs, which are linked to especially long development times and high failure rates. The latest controversial estimate from the Tufts Center for the Study of Drug Development, an industry-supported source, for the cost of the successful development of a new drug is $2.6 billion. However, the CEO of GlaxoSmithKline has called this figure a myth, and Médecins Sans Frontières suggests that more realistic figures might be less than one-tenth of that number.
Newer value-based rationales for high drug prices also appear problematic, as it is very hard to calculate the worth of a drug—especially just after approval. For example, value-pricing a drug such as an anti-PD-1 drug should mean a comparatively high price for treating melanoma but a much lower one for treating lung cancer, where it is less effective. Drug prices clearly are not dictated by these circumstances. Instead, after reviewing the arguments surrounding drug pricing, a hundred experts on chronic myeloid leukemia concluded in a May 2013 article in Blood, “Of the many complex factors involved, price often seems to follow a simple formula: start with the price for the most recent similar drug on the market and price the new one within 10% to 20% of that price (usually higher).”
Drug companies simply charge what they think the market will bear.
The price of a drug can vary dramatically not only from one country to another but from patient to patient. The negotiated price to my insurer of each infusion of pembrolizumab I received was $9,000, or $153,000 for 17 infusions during a year (the list, or off-the-shelf, price for one infusion was $52,000). However, the price for a friend who has different health insurance but received identical treatment was four times as much—about $600,000 per year. The difference is apparently because my insurer’s negotiating group was bigger and had more leverage than that of my friend’s insurer.
I have been incredibly fortunate. My disease was especially indolent. I was able to retire early and look for promising clinical trials. I certainly could not have survived without the continuing support of my wife, Françoise. I was treated in an outstanding university teaching hospital that had expertise in immunotherapy, and I had comprehensive health insurance. My multiple surgeries, plus the expensive drug treatments, were enough to bankrupt almost anyone without coverage. My biggest out-of-pocket expense was commuting to Nashville for 10 months, which totaled $8,000.
During the long struggle, I experienced both remarkable and distressing aspects of the US health care system. I am trying as best I can to contribute to its correction, improvement, and humanization. Everything has changed in my life. There is much to appreciate and savor, and much more to do.
Volume 35, Number 7. July 2016.
A Black Alzheimer’s Patient Wants to Be Part of the Cure
A daughter gets her mother into a clinical trial for an Alzheimer’s drug, with few other black patients enrolled.
Katti Gray
At her thirty-third appearance as Subject 16019 in a clinical trial of an experimental drug she hoped would fix “this little problem with my memory,” Sandra Brannon sank into a medical exam room’s recliner and waited.
“What’s the date again?” Sandra asked me. I had escorted her to a wing at Bellevue Hospital Center in Manhattan, where NYU Langone Medical Center was conducting the trial—one of 210 institutions worldwide doing so. As a family friend, I was standing in for Sandra’s only child and chief caregiver, Monica Montgomery. Thirty-five-year-old Monica was a globetrotter and had business elsewhere that morning.
“November eighteenth,” I answered.
I’d responded to the same question from Sandra four times during our 27-minute ride from a Brooklyn Bridge subway station to Bellevue.
“Right. Got it.” Sandra scribbled my reply on her cheat sheet as a nurse bounded through the door to prep her for her monthly intravenous infusion of the experimental drug solanezumab, which Eli Lilly developed to target mild cognitive impairment caused by Alzheimer’s.
“How was your commute here?” the nurse began.
Sandra had been diagnosed with Alzheimer’s in 2011, at the age of 64. She was 69 on that day at Bellevue. Sandra’s mother had died of the incurable degenerative disease in March 2004 at the age of 83, about seven years after her diagnosis.
“Any changes in your health since the last time we saw you?” the nurse continued, probing Sandra. “You know what today is?”
Sandra cheerfully wiggled and snapped her fingers. She let out a blip of laughter and grinned.
“The third of May?” She hesitated, looked at the nurse’s raised eyebrows, and realized she’d gotten it wrong. “No. Ummm—oh, yes, it’s November eighteenth, Friday.”
“You looked at your pad, huh?” the nurse said, smiling and gently patting Sandra’s hand. She pushed two plastic water-cooler cups toward her: “You’re dehydrated, and we can’t get the needle into your vein easily when you’re dehydrated. Drink this.”
That cheat sheet of scribbled notes and details had become Sandra’s brace and comfort during visits to that special NYU Langone wing at Bellevue. It was ground zero in her quest for something—anything—to slow her dementia. As a black woman, she played a critical role in the trial, and not just because of her own plight: blacks and Latinos are diagnosed with Alzheimer’s more often than whites. Yet during those many visits to that wing, Sandra seemed to be the only black patient present. Ever.
“From the beginning of this,” Monica told me, “I’d see lots of little white ladies lovingly leading their girlfriends into the office and asking questions: ‘I hear you have a clinical trial for Alzheimer’s? We want to get in that.’ But hardly ever—if ever—did I see others of us, black people, there signing up for the same thing.”
When Eli Lilly reported preliminary results of the trial in December 2016, the data told a similar story: 90.8 percent of the trial participants who reported their race were white, 1.7 percent were black.
Sandra grew up in Washington, DC, the child of a schoolteacher and a preacher, and she moved to New York straight out of high school to study art at the Pratt Institute. I’d met her and Monica more than 20 years earlier at Emmanuel Baptist Church in Brooklyn. At the time, Monica was active in the church’s Teen Canteen group, Sandra was a trustee, and I did double duty as a choir member and newsletter editor.
At the start of our surrogate kinship, Sandra was assistant principal at a high school. She’d gone into education after being a graphic designer at the New York Times and CBS News. She switched careers as computer-made illustrations were supplanting the pen-and-pencil renderings that she preferred. Plus, an educator’s work hours were more suited to raising a child during a rocky marriage. She and her husband, a college professor, divorced when Monica was a teenager. That was back when Sandra, voluble and vibrant, commanded a room, weighing in on any discussion and perhaps, to liven things up, peppering the conversation with cuss words.
Monica was living in DC when she first noticed Sandra’s lapses. Sandra had driven there to visit her, but on her way back to New York, she called Monica to ask what highway she was supposed to take.
Over time, Sandra started repeating herself and misplacing things. “She was having these small accidents,” Monica says. “Bumping a street sign with her car. Losing her keys, wallet. It was scary, nerve-racking … and I instantly knew what it was.”
Years earlier, Monica had been involved, hands-on, in her grandmother’s 24-hour care at the end of her life. Eventually, home health aides were hired as well.
“I knew this thing ran in families,” Monica says. “I felt this disappointment and dread, and rugged resignation.”
Monica is an arts activist and museum curator who has lectured internationally and been an adjunct professor at Harvard University. Like her mom, she is charismatic, whip-smart, and a life-of-the-party type. Like her mom, she can be no-nonsense and resolute. Her resolve would serve her well after her mom was diagnosed.
An internist in a private practice, who was also black, had diagnosed Sandra but offered little in the way of treatment options. Instead, the physician recommended that Sandra do crossword puzzles and that the family hope for the best. After Monica’s repeated requests, the doctor finally prescribed Aricept, a treatment for mild-to-moderate Alzheimer’s symptoms.
Monica was living in Philadelphia then. She emptied her rental and moved back to her mom’s Brooklyn apartment. As she researched Alzheimer’s and searched for physicians lauded for their work on the disease, she also conferred with a dear friend, a geriatric social worker, about how to move forward. In the fall of 2013, Monica chose a neurologist at NYU Langone Health’s Center for Cognitive Neurology to treat her mother and began discussing how to enroll her in a clinical trial.
“Actually getting her into the trial was an uphill battle,” Monica says.
They needed a letter from Sandra’s diagnosing physician, but she ignored Monica’s pleas for assistance.
“At a certain point I just rolled in there and, without causing a scene, said ‘My mom is deteriorating,’ ” Monica remembers.
She also wrote a pointed letter to the Center for Cognitive Neurology. She was frustrated after too many delays and we’ll see’s.
In the fall of 2014, Sandra joined what ultimately were the 2,129 patients in Eli Lilly’s trial of solanezumab at those 210 sites in the United States, Canada, Australia, Japan, and Europe. Sandra was notified in fall 2015 that she was being infused with solanezumab and not the trial’s placebo drug.
Sandra’s optimism spiked.
In 1994 the National Institutes of Health (NIH) mandated that participants’ enrollment in NIH-approved clinical trials reflect the nation’s racial makeup and gender breakdown.
Whites accounted for 61.6 percent of the US population in July 2015, according to the most recent census data. Census analysts project that share to dip to 44 percent by 2060 if current trends hold. And in 2020 more than half of the nation’s children up to age 18 are projected to be people of color, the Census Bureau says.
According to a 2013 NIH report, minorities accounted for 36.5 percent of the 17.6 million participants in NIH-registered clinical trials of drugs and other medical interventions in fiscal year 2012.
But researchers in the EMPaCT Consortium, which provides training to medical professionals and community organizations on the mechanics of clinical trials in a bid to increase minority participation in trials, estimate the share of minority participants in NIH-registered clinical trials to be less than 10 percent. Depending on the disease targeted by a trial, that rate could be even lower. A study published in Cancer in April 2014 concluded that fewer than 2 percent of National Cancer Institute clinical trials focused primarily on any minority population.
The percentage of minorities in clinical trials conducted by drug manufacturers—which now carry out the vast majority of drug trials—is more difficult to pin down. According to Nathaniel Stinson, a medical doctor who heads the Division of Scientific Programs at the National Institute on Minority Health and Health Disparities, companies often don’t disclose much—if any—information about the race of participants or other aspects of their trials until the trials are over.
“At that point, everything, regrettably, is after the fact,” he says.
Given the nation’s changing demographics, it’s neither good health policy nor good business to be developing drugs and possible cures that are tested in only a subset of the population, says Willie Deese, who retired in June 2016 as executive vice president at Merck. Months after retiring, he earmarked part of a $1 million gift to the North Carolina Agricultural and Technical State University in Greensboro—a historically black college and Deese’s alma mater—for its groundbreaking Center for Outreach in Alzheimer’s, Aging and Community Health. In addition to collecting and studying the DNA of black patients with Alzheimer’s, the center provides support services for such patients and their caregivers and educates blacks about scientific research.
Deese, whose mother has Alzheimer’s, says there’s a growing recognition within the black community that its members can’t be absent from clinical trials and expect medicine to work as well for them as it does for other populations.
“We have to be included,” he says. “Enlightened companies are ensuring that that’s taking place today.”
Raegan Durant, an internist and professor at the University of Alabama at Birmingham, believes that scientists and policy makers are far more aware today of the critical need for a diversity of clinical trial participants than they were when the NIH issued its mandate. But the challenge is figuring out how best to achieve a representative mix and to keep the recruitment of minority volunteers from being an afterthought.
“There must be a shared power,” he says, “between science and laypeople.”
The almost-four-year-old Mississippi State Department of Health’s Community Research Fellows Training Program is considered one leading model of change. It schools community members in how trials work and partners with grassroots organizations that directors of the fellows project view as deeply invested in public health and wellness. It’s creating a community-clinician conversation about trials in the hope, among others, of keeping doctors from mentioning clinical trial enrollment opportunities at the moment when they deliver a dire medical diagnosis. That might be the least opportune time to raise the possibility of joining a trial.
Thus far, some of the more than 50 graduates of the intensive four-months-long training program have gone on to sit on clinical trial institutional review boards in Mississippi and to win grants for public health initiatives.
Still, as more minorities express interest in enrolling in clinical trials, there are some looming questions and challenges: How can medical science and the culture surrounding trials be demystified? Should clinical trials continue to exclude, as they generally do, patients with comorbidities—especially given the disproportionate percentage of minorities with more than one illness?
“That’s the million-dollar question,” says Michelle Martin, a preventive medicine professor and researcher who was hired by the University of Tennessee Health Science Center in 2016 to help shape its research programs.
“It probably does matter in some cases,” Martin tells me. “But there has been a movement toward more pragmatic trials where the inclusion-exclusion criteria are a little broader.”
Whether trials accommodate the everyday circumstances of participants is another key concern. What of the added costs of taking time away from a job to keep a spate of clinical trial appointments? What of a participant’s childcare or eldercare needs? Who’ll put gas in the tank or cover public transit costs so participants can get to trial sites?
Too often patients and caregivers are solving those problems on their own and doing all the legwork needed to get into clinical trials they hope will benefit their health, says Jennifer Wenzel, a professor at Johns Hopkins University’s School of Nursing and School of Medicine and an EMPaCT researcher and advocate. It’s an unfair burden, she says.
Some patient navigators specializing in clinical trials have helped trial participants of color stick with their experimental treatments. The navigators function as patient advocates. They serve as liaisons between patients and their physicians and trial researchers. They whisk away tears. They explain things. They cheer on participants and have helped ensure that patients stay on board until the trial ends. Yet navigators are not spread across the entire clinical trial system, and some health systems that do have them struggle with how to compensate them for their time and effort.
Meanwhile, experts say that private physicians, no matter the many demands on their time at work, must do more to change the racial makeup of trials. Referrals to trials may be more common at university-run medical centers, where staff members know about on-campus research and help funnel patients to those researchers. Unfortunately, not every doctor has this access or will make the effort.
In November 2016, Eli Lilly reported that solanezumab had no effect on people with mild Alzheimer’s symptoms and began winding down the trial. Monica forwarded the emails from the NYU researchers to me. She and her mother were crushed by the drug’s failings. Monica tried not to show her mother the fullness of her disappointment and anger, afraid they would rub off. They decided that Sandra’s thirty-fourth appointment at Bellevue would be her last, though she had been given the option to make more visits to the clinic.
Monica requested all of her mother’s clinical trial files, the stuff in the black vinyl binder that nurses annotated during Sandra’s monthly visits. For being a trial volunteer, Sandra had gotten a $40 monthly stipend and extravigilant checks of her weight loss, blood pressure, cholesterol level, and assorted physiological markers. Monica wanted those notations.
“I want a written report of their findings and her progress, if she made any,” she tells me. “I just want more insight into my mom’s health.”
Today Sandra is no longer avidly reading the New York Times. She does head to a senior citizen center several afternoons a week, but only if Monica lists the landmarks along the way for her, a new kind of crutch.
For 14 consecutive days this March, Sandra phoned the doorman of her high-rise co-op apartment building for instructions on using the elevator that had ferried her for four decades.
“She just stands in front of it at times, totally bewildered,” Monica says.
Monica and the home health aide decided that Sandra should no longer light the stove to cook anything. She has grudgingly surrendered her driver’s license and the keys to her Toyota. Except for sporadic engagement with a handful of friends, she is more isolated than ever.
Yet when I talk to her after the clinical trial ended, Sandra sounds pretty chipper, all things considered. She once told me that she veers toward joy, no matter what life throws at her.
“Thank you,” she tells me, “for being my balm in Gilead.”
An emotion I cannot name rears up in me. My eyes tear.
“You’re my jewel and joy,” I say. “Let’s get our nails done and go to lunch soon.”
“Absolutely,” she says. “And you never know. The drug they were giving me might kick in.”
Monica shares her mother’s cautious optimism.
“We still hope, somehow, that all of this will lead to a cure,” she says. “We can feel like my mom was a part of that.”
Volume 36, Number 6. June 2017.