Pandora's DNA: Tracing the Breast Cancer Genes Through History, Science, and One Family Tree

by Lizzie Stark

  Lynch had the pedigrees but not the technical know-how to look inside their DNA, while King had the know-how but needed families; so they collaborated for a while. Later, in 1987, First Lady Nancy Reagan would inadvertently help King find families too. A local television reporter interviewed King for a spot on the birthday of the National Cancer Institute, and she explained her project and the need for cancer families. The next morning, Nancy Reagan announced her breast cancer diagnosis, and news media scrambled to find spots on the topic. King’s interview was rebroadcast across the country, and letters from families came pouring in.

  King used some of the sequencing techniques she’d honed during her dissertation, namely the use of polymorphic gene markers to investigate the families. Such markers help scientists locate genes by providing a landmark of sorts. Let’s say George is in New York City and I want to find him. Well, New York City is a huge place full of buildings and people—looking in every nook and cranny, assuming he stays still, would take several lifetimes. But if I know that George usually hangs out within five blocks of the Empire State Building, that narrows things down considerably. It’ll still take a long time to find him, but it’ll be much faster than ransacking the entire region. A polymorphic marker operates like the Empire State Building—it helps scientists know they’re at the right part of the genome.

  Physical traits can serve as gene markers. If everyone in my family who has asthma also has attached earlobes, then perhaps the genes that code for these two traits—asthma and earlobes—are inherited together. And if they are inherited together, perhaps it’s because they are located right next to each other on a strand of DNA. Scientists figured out how to find markers not linked to physical traits and used those to help study heredity.

  So King spent the next decade and a half searching for gene markers present in cancer patients from families with lots of cancer. If her lab could identify a good marker, then she’d be able to tell which gene might hold the key to familial breast cancer risk. Eventually, she and her team assembled nearly two dozen cancer families from the United States, Puerto Rico, Canada, the United Kingdom, and Colombia. They unearthed hospital reports and death certificates to confirm the presence of cancer in some people and drew blood from the 329 surviving family members so they could extract DNA. Her team was looking at 173 different markers in the twenty-three families, and as she told an interviewer, “of course everything was done by hand because none of the analysis was computer-based. We literally had the pedigrees rolled out across lab benches and floors. Then Beth Newman [a colleague] had the idea of arranging the pedigrees in the hallway in order by average age of breast cancer diagnosis in the family.” After all, they knew that women who developed breast cancer young were more likely to live in families with many cases of cancer. Focusing on these women, they hit pay dirt. At a genetics conference in October 1990, King announced the shocking findings: one part of chromosome 17 was altered in women from breast cancer families who had the disease. And as many as one in two hundred women carried such a mutation, making it one of the most common inherited syndromes. King still felt a bit unsure of herself. At the conference, “I presented our data as an interesting story. I knew it was statistically robust, but I was concerned it might be some elaborate fluke.” Several important research groups—including those of French scientist Gilbert Lenoir and Utah-based geneticist Mark Skolnick, who would put his own stamp on the breast cancer genes soon, were in the audience and asked King for her marker sequences, which she sent out. Two months later, the King lab published its results in Science. A few months after that, in early 1991, Gilbert Lenoir—then a collaborator of Henry Lynch—presented a talk on hereditary breast cancer at a conference in England. “He presented what I interpreted as a summary of my talk,” King told an interviewer, “with extremely familiar results, assuming that he would go on to describe his results next. But he stopped. I asked what were his results, and he said, ‘Those are mine!’ His results were virtually identical to ours! Same markers, same age effect, same lovely fit to the same model, exactly our results but on unrelated families.” Lenoir had confirmed her results and also improved slightly on King’s findings, showing that women with an alteration on this part of chromosome 17 had a high risk of ovarian cancer as well as breast cancer.

  Many years after snooping in my mother’s desk, I confessed and asked if I could see the letter. She dug it out, along with two other letters of note. We discovered that in 1991, the year King’s announcement was verified, my mother read about it in the paper and immediately sent letters to the King and Lynch labs offering the family DNA for study. She remembers answering questionnaires for both.

  King’s news fired off the starting pistol in a heated race to find the gene, dubbed BRCA1. Although, of course, “race” isn’t the word King would use. As she told Waldholz, using that word “is just an awful way to describe what we’re doing, especially when you think about the women who are depending on us. I don’t mind this being called a race, if you mean a race against the disease. Every time we hear of another woman who died of breast cancer, we take it personally. We should have the gene by now.” In the twenty years it took to find the gene, King pointed out, more than one million women died of breast cancer. Later, in her 2014 piece recounting her search for the BRCA1 gene, she explained, “Until there are no more breast or ovarian cancers among women with BRCA1 or BRCA2 mutations, the real race is not over.”

  King’s 1990 results proved that some breast cancer was genetically linked—but there was still plenty of work to do. DNA has that familiar twisted-ladder structure, in which each rung represents a chemically joined base pair. Guanine binds with cytosine and adenine binds with thymine to make our DNA strands. The twisted ladder, which can be unzipped to make proteins that regulate cell function, is itself wrapped around proteins called histones, coiled up into larger structures called chromosomes. Humans have twenty-three pairs of chromosomes—one set from each parent—wound up inside the nucleus of almost every cell in the body. King’s results were astonishing because she’d narrowed down the gene’s location from three billion base pairs—the size of the whole human genome—to a particular region of DNA that was twenty million base pairs long.

  But twenty million base pairs is still pretty big. As geneticist Francis Collins put it to Waldholz,

  If you consider human DNA as being the size of Earth, King had just placed the gene somewhere in Texas. It was no small accomplishment because few people even believed it had existed at all. Yet, now the job ahead was to pinpoint the gene, to first map it to a particular county in the state, then to a town, then a street, then a house on the street, and finally, the exact room in the house. It was going to be a very difficult assignment.

  The prize—to be the first to clone the BRCA1 gene, earning fame and funding, and maybe saving some lives—enticed the world’s best genetic laboratories into the arena. According to King, more than one hundred researchers from at least twelve labs spent the next four years searching for BRCA1, a particularly extraordinary effort, given that the Human Genome Project had just begun, so the hunt started nearly from scratch.

  Mark Skolnick’s lab in Utah was already in the fray. Skolnick had begun his career looking at population genetics. In grad school he worked on a project tracing the family trees of the residents of Parma, Italy, to study the genetic shifts that occurred over time in a populace that had been mostly isolated. Thanks to his skill with computers and his previous background in demographics, his adviser tasked him with organizing centuries of family history. After the project ended in 1973, a professor at the University of Utah contacted him in hopes that he could apply the same know-how to help find cancer in the Mormon families of Utah.

  The Mormon family trees represented a huge advantage in the race to identify the breast cancer gene for several reasons. For starters, the Mormon Church encourages large families, and large families are good for cancer genetics research. The religion also believes that people can be baptized and r
eceive the gospel after death. As the church website puts it, “Discovering that you’re related to a renaissance nobleman could be fun. It could also mean giving him and his family an opportunity to receive the gospel of Jesus Christ.” In practice, this means that tracing family trees is a holy mission, and the church has assembled an incredible number of family pedigrees—over one billion people tracked down by Mormon sleuths—that would prove to be a treasure trove for researchers like Skolnick.

  In addition to the Mormon genealogical records, Utah had an extensive cancer registry that started compiling current cases in 1966 and had records on tumors that dated back to 1952. The potential juxtaposition of the two databases intrigued him—combined, they could be a powerful tool. The university didn’t have funds for such a large project, so the young Skolnick applied for funding from NIH, despite the fact that prevailing wisdom didn’t favor a link between cancer and genetics. In the meantime, he made a curious discovery in the genealogical records: the thirty-three cases of rare male breast cancer he found were all related to one another through relatives near and far. He also studied lip cancer, learning that it clustered in families too. He decided to investigate breast and colon cancer. After receiving the grant, his team spent the next five years on data entry, linking two hundred thousand family pedigrees that included records on about 1.6 million individuals from the church’s library with thousands of cancer cases from the state registry to form a database. As Skolnick identified possible cancer families, his lab collected blood and tissue samples from some twenty thousand people over a decade.

  In the meantime, because cancer risk is incredibly complicated, Skolnick went after a simpler hereditary disease in hopes of developing a method that could be used to attack the problem of finding a cancer gene. He focused on hemochromatosis, a rare disorder that prevents an individual’s body from processing iron properly, leading to buildups that can cause cirrhosis of the liver, diabetes, and heart damage. Researchers already knew it was hereditary but weren’t sure whether it was a dominant trait, requiring one copy of the gene, or a recessive trait that required two copies of the gene. Building on French research that found a blood protein associated with hemochromatosis, Skolnick figured out that the gene coding for that protein must be a marker for the hemochromatosis gene, which helped locate the gene responsible, proving that it was recessive and leading to treatments such as blood transfusions for people who had not yet developed the full-blown disease.

  And then, at a ski retreat in 1978, Skolnick, MIT biologist David Botstein, and geneticist Ronald Davis of Stanford University had a brainstorming session that would change the landscape of genetic research. Essentially, they figured out how to make markers—like the DNA coding for the blood protein had been for hemochromatosis—synthetically using enzymes. Different enzymes snip DNA strands at specific places, say between the A and the G (adenine and guanine, two bases that make up the rungs of the DNA ladder, along with thymine and cytosine) in the sequence TCTCAG. But human beings have a lot of variation, and some of us have longer genes than others. The enzyme still snips our DNA strands at the same place, but in me, the resulting fragment might be longer than yours, with an extra string of letters in there. Those letters, known as restriction fragment length polymorphisms, or RFLPs, can serve as gene markers and are passed down through generations. If I have my mother’s RFLP, perhaps I also have her breast cancer gene. If everyone with cancer in my family has that RFLP, then perhaps it could be the genetic Empire State Building, guiding researchers ever closer to the location of the BRCA1 gene. The discovery of RFLPs revolutionized genetic research.

  Months after Mary-Claire King announced that she’d narrowed down the location of the gene, Mark Skolnick joined forces with businessman Peter Meldrum to form Myriad Genetics “specifically with the goal of isolating the breast cancer susceptibility gene,” as Skolnick told the DNA Learning Center. Skolnick needed funding in order to compete with other gene hunters. As he put it in a deposition for the later Supreme Court case around the BRCA1 gene, “I was also keenly aware that NIH had awarded [renowned geneticist] Francis Collins a massive genome center grant which would allow him to pursue cloning this gene, and that my group would most likely not be given adequate funds to compete. This in fact turned out to be true. My collaborators and I submitted a small grant proposal to pursue BRCA1, but we were turned down. We were told that we didn’t have the family material to be competitive, when in fact it was common knowledge that we had spent years collecting the most extraordinary breast cancer families in the world.” If Skolnick couldn’t get the money through grants, then he’d get it through private funding. By August 1992, Myriad had received a $4 million investment from a pharmaceutical company and added Dr. Walter Gilbert, a Nobel laureate in biology, to its team. By March 1993 they had $8.8 million more in investments, according to Skolnick’s deposition.

  With considerable funding at his disposal, Skolnick was hell bent on success. He stated in his deposition, “We were acutely aware that if we were to fail to find BRCA1 we would have had great difficulty in surviving as a company and that our jobs would be lost.” By the end of 1994, after many long hours in the lab, they had identified and cloned the BRCA1 gene, beating out some of the world’s best laboratories in the process, including King’s.

  “People ask me how I felt,” Skolnick told Waldholz. “You know, we didn’t pop champagne, we didn’t go out to dinner. We just felt this tremendous relief. After all these years, I could finally take off my racing shoes.” Skolnick and King had had a brief collaboration around the breast cancer gene a decade earlier but had gone through an ugly split. A 1994 New York Times article about the discovery mentioned that King, “whose personal less-than-tender feelings about Dr. Skolnick are well known to her colleagues, nonetheless described the discovery as ‘beautiful’ and ‘lovely’ and deserving of all the praise it might win.” Later, she would tell the Lancet, “While the race was in progress I certainly thought that I would be disappointed if we didn’t win. But actually, much to my surprise, I was not.”

  With its commercial motives, Myriad went on to patent the BRCA1 gene in a move that caused immediate controversy. But, as an October 29, 1994, article put it, “the gene’s co-discoverers, the University of Utah and Myriad Genetics Inc. of Salt Lake City, left their Government collaborators off the patent for BRCA1. That meant the National Institutes of Health would lose control over diagnostic tests or cancer therapies developed from BRCA1 and their prices.” Myriad president Peter Meldrum told the Associated Press that the company didn’t include NIH on the patent because it hadn’t made a significant contribution to the discovery. The lead scientist for NIH begged to disagree. And while Myriad said it had spent $14 million discovering the gene, NIH contended it had contributed $4.6 million.

  Though the patent disputes over BRCA were just beginning, the search for hereditary breast cancer was not yet over, because BRCA1 did not account for all the many cancer families scientists across the world had been digging into. This problem fascinated Dr. Mike Stratton, a scientist with the Institute of Cancer Research and the Wellcome Trust’s Sanger Centre in the UK. Unlike King and Skolnick, Stratton had a strong medical background, beginning his career as a histopathologist, someone who looks through the microscope at tissues—in Stratton’s case, cancer cells—to decide whether they are healthy or harmful. “I knew the DNA was abnormal, but I couldn’t see what the abnormalities were,” Stratton tells me. “So it seemed to me a fascinating and all-engrossing question to find out what those abnormalities were.” At age thirty he left medicine, earned a degree in the molecular biology of cancer, and started developing the Breast Cancer Genetics Programme at the Institute of Cancer Research in London.

  After the discovery of the BRCA1 gene, he began collecting cancer families without BRCA1 mutations in hopes of discovering another breast cancer gene. At first, he gathered families in southwest England, people who had kids in the 1980s and 1990s, but it proved a nonstarter—the family groups we
re too small, and people often lost contact with the relatives who were so essential to track in a study like this. He needed bigger families, so like Skolnick, he worked the religious angle, deciding to focus on the nearby Catholic country of Ireland. He sent letters to all the oncologists in the country, and through these connections he located one particularly large, particularly cancer-prone family. Over a year and a half, his team tracked down several hundred members of this family stretching over four generations; the family had about thirty cases of breast cancer not linked to BRCA1. And then they learned about a case of male breast cancer, something that is rare even among BRCA1 families. Stratton’s team added their discovery—male breast cancer in a non-BRCA1 cancer family—to the evidence that there had to be another breast cancer gene. As Stratton puts it, “This family became the centerpiece of our research for BRCA2. We now had evidence, strong evidence, that there was another gene.”

  Stratton’s lab in the UK formed a collaboration with Skolnick’s lab in Utah to find the general location of the BRCA2 gene, which was now a relatively fast process because maps of markers throughout the genome had improved since King and Skolnick had begun their work more than a decade earlier. Stratton and Skolnick located BRCA2 on the long arm of chromosome 13 in record time—about a year. The next step was to locate the BRCA2 gene itself inside the arm and to clone it.

  But Stratton worked for a nonprofit institute, the Sanger Centre, and Skolnick worked for a for-profit company, so they parted ways. “We’d had a cordial collaboration,” Stratton says, “but if we continued that collaboration to the identification of the gene itself, that would be with Myriad, and we were not comfortable with the patenting and monopolizing policy of Myriad.” The race was on again, with Myriad, Stratton’s lab, and many others of the world’s best plunging into the fray. “I have to say that we were not—certainly, I was not—optimistic as to the likelihood of our chances in this race,” Stratton says. “Myriad had just identified BRCA1. And they were very experienced in the art of finding genes by this process known as positional cloning. They were very well resourced and well financed.” Still, Stratton’s lab forged ahead. “If nothing else, we would learn from the experience of doing it,” he says.