After the meeting most of us went to a bar in Boston to be part of a science café—we each spoke for a couple of minutes, discussed the PGP with patrons, and imbibed (some of us anyway). I asked George what had gone wrong. As usual he was sanguine. “We have this mountain of bad data with people who have no family history for any genetic diseases. So … garbage in, garbage out. That was my expectation, but we had to do something … so we did what we could. Hopefully this will improve before our next installment. It’s not clear how many more events like today we have in us without making some major breakthroughs first. I think we have to cure a disease before we get the Washington Post, the New York Times, and the Boston Globe here again.”78
On the flight home I wondered if maybe this long, drawn-out process of returning our data was an opportunity to rethink the whole thing: Did I really want my genome to be available to everyone via a mouse click when I clearly didn’t feel that way about other aspects of my life? I was still not on Facebook or Twitter (though I would eventually succumb to the latter) and felt no burning desire to join the social networking crowd—fielding dozens of requests from other 23andMe customers wanting to “share” results and exult in our common Ashkenazic pedigree was more than enough.
My therapist speculated that maybe my whole foray into the land of public genomes was just another form of acting out, a cry for attention, a way to ingratiate myself with “the cool kids” like Esther Dyson and Steven Pinker and George Church. Maybe she was right. My problem was that, at the end of the day, I was not one of the cool kids. I was not as smart or unflappable or selfless or capable of seeing the big picture. I was constitutionally ill-equipped to roll with the punches; George, on the other hand, excelled at it. He told me that even as an infant his mother said he was an easy baby—"a stoic,” she called him.79 At the risk of invoking genetic determinism, as a baby I was a howler. So were my kids when they were infants. And as an adult, I was not about to put my Social Security number or the GPS coordinates of my house on the Web. Maybe this meant the PGP was not for me. Maybe negotiating the details of a public genome was more than I, to say nothing of my family, could handle. I resolved to not make any decisions until I learned my BRCA status for the most common mutations in Ashkenazi Jews. But with the Polonator still in the midst of a long, slow gestation and the Harvard sequencing facility having its own problems, the PGP was unlikely to provide that any time soon.
The only way around this was to look elsewhere: to the world of commercial genetic testing, which, presumably, would deliver garbage neither
in nor out. I called my local hereditary cancer clinic and explained the situation as best I could (glossing over the fact that I wanted to do this because I was about to, um, put my entire genome on the Internet). I asked if they would order a test for me for the Ashkenazi breast cancer mutations and send the bill ($535) to my insurance company. The genetic counselor on the other end of the phone was pleasant but officious: she told me the clinic staff would be happy to order the test for me as soon as my mother contacted them, answered the same set of questions I had, and then got tested herself. It was my mother who was the cancer patient after all, the counselor explained: if she didn’t carry a mutation, then the rest of the family was off the hook. It was simply the most efficient way of doing it. I understood this—it was the traditional, Bayesian* way of identifying mutation carriers in families.
But it wasn’t what I was after. I explained that my mother was seventy-five years old, had had both breasts removed, and as far as I knew was long done with her ovaries. Why did she have to be considered the proband—that is, the one who first sought consultation for a genetic disorder? I was the one who wanted the information, and if my insurance company wouldn’t pay, then I would. I loved my mother, but there was a whole bunch of reasons I didn’t want to make her an active part of my little journey of genomic self-discovery. The counselor politely but firmly refused to budge. “If your mother absolutely won’t do it, then we can talk,” was as far as she would go. I said thank you and hung up the phone. “Fuck you,” I thought. This was exactly the kind of situation that led to the rise of the personal genomics companies.
I thought about pulling “a Hugh Rienhoff”: extracting my own DNA, ordering the primers to amplify the BRCA genes, and borrowing some lab time, space, and equipment. I could have an answer in a week, assuming I could still recognize the business end of a pipette. That was a big assumption—I hadn’t done real lab work in years and I didn’t want to compromise my scientist friends by skirting the human subjects research rules (the same ones I was supposed to enforce) and infringing the Myriad patents on the breast cancer genes, as odious as I found them to be (more on those in a minute).
Instead I called a friend who worked for DNA Direct, a San Francisco–based company that was an early proponent of providing Web- and phone-based access to genetic testing for people who wanted to avoid wading into the morass of paperwork or otherwise involving their health-care providers. Or their mothers. DNA Direct would submit my test for coverage to my insurer; my out-of-pocket cost would be two hundred dollars for the counseling session (again, I noticed that insurance companies didn’t like to pay for medical care that only involved talking).
For me it was money well spent. I didn’t have to bug my mom or my colleagues. And my genetic counseling session with DNA Direct was like a warm bath. I sat in my office and talked on the phone with Lisa Kessler, an experienced counselor who had just had a baby and worked from her New Jersey home. Listening to her reminded me of why I was such a hopeless genetic counselor and never even bothered to take the board exam. I was nervous and awkward with patients; Lisa’s voice and telephone manner, on the other hand, were quiet and soothing. She was comfortable in her own skin. Her empathy felt genuine. We discussed the PGP, genetic counseling training programs, her newborn, and all of the various breast cancer risk models. She explained what we could and couldn’t say about my risk with and without a genetic test for the three Ashkenazi mutations in BRCA1 and BRCA2. Given my family history (which was incomplete) and ethnic background, my mother’s risk for carrying a BRCA mutation was somewhere between 12 percent and 38 percent. My risk for carrying one was therefore half of that. Myriad told her that the three mutations I was getting tested for accounted for 90–95 percent of the mutations in the two genes that were found in Ashkenazim.80
A few weeks later she called with my results. “I have good news,” she said.81
“I know,” I said. Eight weeks before her call, I had churned two milliliters of my saliva into a plastic tube and sent it to 23andMe’s lab for processing. The company had typed 600,000 markers on my DNA and I could now log on and see what those markers had to say about various traits of mine. To my surprise, those traits now included the three major Ashkenazi breast cancer mutations! Why was I surprised? First, my impression from an early conversation with company cofounder Linda Avey was that 23andMe was not interested in typing customers for SNPs that conferred high risks for single-gene disorders. It seemed to me that in the beginning they were skittish about the prospects of having to deliver bad news over the Web. But when I looked at my report online in 2009, I saw my carrier status for mutations not only for the Ashkenazic breast cancer mutations, but for a handful of other single-gene diseases, too, cystic fibrosis and sickle-cell anemia among them (I carry neither). In fact, for months George had been agitating for direct-to-consumer genomics companies to start testing for rare, single-gene disorders. Hugh Rienhoff also had said that by causing such serious phenotypes, nature was telling us that these genes were the important ones and the ones the consumer genomics companies should be paying attention to.
The other reason I was surprised that 23andMe had dipped its toe into breast cancer genetics was the fact that both the BRCA1 and BRCA2 genes were patent-protected and controlled by a company called Myriad Genetics.82 23andMe was almost certainly infringing upon Myriad’s patents.
Whenever I told someone outside the genetics world that as
part of my research I was studying the effects of gene patents on access to genetic testing, he or she usually did a double take. “You can patent a gene? How is that possible? Isn’t that like patenting a tree? Does that mean some company owns part of me?” I was and am sympathetic to those intuitive suspicions. But holders of gene patents made several counterarguments. One was that they have not patented a gene found in nature. They have instead patented a gene that’s been cloned, amplified, and/or otherwise manipulated by humans:
gene patents, their holders said, cover inventions, not discoveries—just like all valid patents do … or should. And, they argued, the isolation of a gene was a true invention. Furthermore, proponents would say that they have patented the method (or methods) of looking for medically important mutations in those genes. Finally, companies with exclusive licenses to particular genes like to say that they need their particular monopolies to expand the market for testing of those genes.83 In their view, exclusive patent licenses are necessary to spur innovation: without the promise of a nice payoff at the end, investors would not sink money into commercial genetic tests and patients would be forced to rely on the vagaries and slow turnaround times of research-based testing.84 In general, Bob Cook-Deegan, postdoc Shubha Chandrasekharan, and I (along with many colleagues at Duke and elsewhere) had found that the last argument did not hold water. Rights to the cystic fibrosis gene, for example, were licensed broadly and the test had always been cheap, reliable, and easily available.85
Indeed, genetic screening itself seemed headed toward the “close to free” model. In 2009 a company called Counsyl had sprung up whose product was a “Universal Carrier Screen” for more than one hundred genetic diseases, $349 out of pocket or free to customers with insurance. The company included PGPers/famous-Harvard-guys Steven Pinker and Henry Louis Gates, Jr., among its advisers.86 The Counsyl tag line: “Thinking about starting a family?”87 Its website described a “campaign” to end preventable genetic disease,88 presumably by prospective parents who were carriers of mutations in the same genes not having kids or by selecting against affected embryos.
Preconception genetic screening has improved people’s lives immeasurably. That said, for Counsyl to call this a “campaign” akin to the campaign to fight AIDS was a bit over-the-top, I reckoned—not everyone thought of hereditary deafness as a disease, for example.89 And just as being a sickle-cell carrier is protective against malaria, being a carrier of—and perhaps even afflicted with—other genetic diseases or “disabilities” probably confers evolutionary advantages in certain cases.90 But Counsyl made no bones about its mission:
It is something new, born of the realization that cutting-edge science and market forces can actually increase equality and promote social justice. It is a cause, a campaign to finally end the needless suffering of preventable genetic disease. And most of all, it is you. Call us idealistic, but we believe that everyone loves their children and will do the right thing when it comes to safeguarding their future.91
For Myriad, neither Counsyl’s launch (although Counsyl didn’t test breast cancer genes) nor the 23andMe move into BRCA1/2 could be construed as good omens. Although Myriad’s profits from BRCA testing had been robust, its image had not. Because it had a monopoly and refused to sublicense its tests, and because it charged more than three thousand dollars, Myriad had been called “probably the most hated diagnostics company.” (Given Myriad’s many years of experience with BRCA testing and its high rate of insurance coverage,92 whether it was the “most hated” was debatable.)
But the company was under assault on two fronts. First, by returning results on BRCA1 and BRCA2, even on just three mutations out of the more than 2,500 known, 23andMe appeared ready to flout Myriad’s licenses, although Linda Avey downplayed the intellectual property aspect. “Because this subset of BRCA markers doesn’t fully replicate the Myriad test (and it’s not positioned to be an alternative for in-depth BRCA testing),” she wrote to me in 2009, “we’re not certain how Myriad will react.”93 But if 23andMe could get away with it, how long before other labs started screening for hereditary breast cancer susceptibility genes?
The year 2009 also brought a full-blown legal challenge: in May the American Civil Liberties Union filed suit against Myriad, claiming that the company’s monopoly on the BRCA genes: 1) made it impossible for women to access other breast cancer genetic tests; 2) prevented them from getting second opinions about their results; and 3) allowed Myriad to charge exorbitant prices.94 Myriad responded by filing a motion to have the lawsuit dismissed, saying that the ACLU lacked the standing to bring the suit and that “a mere policy disagreement” did not warrant a declaratory judgment.95 Judge Robert Sweet said in so many words that the plaintiffs did indeed have the standing to bring the suit and that he would like to hear the case.96
At the time I thought the ACLU suit could be important, but I also thought it might be too little too late: even its critics conceded that Myriad did an excellent job of testing BRCA1 and BRCA2,97 although the company had been criticized for overselling its test to women who didn’t need it.98 I believed this cash cow would dry up of its own accord: most of the relevant BRCA patents were set to expire within the next few years, as were most other gene patents related to single-gene disorders. But Judge Sweet opted not to wait. In March 2010 he handed down a shocking summary judgment that made the court’s view perfectly clear:
In light of DNA’s unique qualities as a physical embodiment of information, none of the structural and functional differences cited by Myriad between native BRCA1/2 DNA and the isolated BRCA1/2 DNA claimed in the patents-in-suit render the claimed DNA “markedly different.” This conclusion is driven by the overriding importance of DNA’s nucleotide sequence to both its natural biological function as well as the utility associated with DNA in its isolated form. The preservation of this defining characteristic of DNA in its native and isolated forms mandates the conclusion that the challenged composition claims are directed to unpatentable products of nature.99 [emphasis added]
In other words, DNA in one’s body does the same thing it does outside one’s body: it carries information. Thus, a gene in a test tube or a sequencing machine is not much different from a gene in a cell in a human body. It cannot be considered a true invention and therefore cannot be patented. Obviously, the defendants were not happy about this ruling and vowed to appeal.
But in George’s view, even if the ruling had gone the other way, Myriad’s business model was still doomed. “I think that protecting individual SNPs is legally unsustainable,” said George of the Myriad patents. “It’s so easy to get people that data without infringing. We’re just going to sequence whole genomes and give them the raw data. If the companies think that that’s infringing their patents, then we’ll just give people a machine to do it themselves. And if that’s infringing, then we’ll just give them the parts to make the machine. It’s reductio ad absurdum. There’s not a lot of sympathy for patenting of DNA sequences. It’s something that’s at the intersection of the open source and human rights movements: ‘This is mine and you can’t have it.’ ”100
On a personal level, knowing that I was “clean” for the Ashkenazi mutations via two independent sources was a relief. There was still a chance that I carried another BRCA mutation. Ann said she could live with that risk if I could; I thought I could. In the two years since I’d been consented, the world had made small but significant steps in the direction of less genetic privacy: dozens of people were in the process of getting sequenced and even some research participants were receiving their own genetic information. The first six whole genomes had all made their data public. By 2010, the wholesale cost for a full sequence at high coverage was less than $10,000, which meant a lot more people would soon be in the pipeline. And judging by the twelve thousand prospects who had already signed up to join the Personal Genome Project,101 the destigmatization process appeared to be well under way.
Of course, in the final analysis none of that really mattered: this
was my genome to do with what I wanted. I had already made my phenotypic information public: if the world cared, it could read my profile and know that I was prone to anxiety and depression. Was it likely that I carried anything in my genome more stigmatizing than that? Of course, any of tens of thousands of researchers could now test that hypothesis by ordering my cells ($85) or my DNA ($55),102 sequencing me eight ways to Sunday, and publishing any damning allele I might carry. As I’ve said, this was not something I was terribly concerned about, but until now no one had ever had to give it much thought one way or the other.
When I moved to Durham, I had no idea how to find a doctor. I’m still not sure I do other than by asking my white-coated colleagues. To find a general practitioner, I wound up making a list based on somewhat arbitrary criteria: she had to be reasonably close to my house, an internist, and accept my insurance. I preferred female physicians: I don’t know why, perhaps because of some deep-seated mommy issue. Other than a Club Med in Grenada, I didn’t much care where she went to school or did her residency or even if she came highly recommended—I assumed she would be competent, what with Durham being the “City of Medicine” and all. I had never really thought of my doctor as “someone to talk to.” She was someone who made me stick out my tongue, whacked me on the knee, and told me to turn my head and cough, which I did with great awkwardness. I found one who was affiliated with Duke and she was terrific—pleasant and sharp, and with a Ph.D. in biochemistry even. We chatted about drug development as she looked in my ears. But not long after my second visit she moved to Singapore. Argh. I had to start over. I got my insurer’s provider directory and started googling. This process yielded not much more than addresses and phone numbers—it seemed like it led mostly to commercial sites that, for a fee, would tell me my prospective doctor’s “grade” or “score.” Based on what, exactly? Choosing a physician was not like choosing a restaurant. Or maybe it was.
Here Is a Human Being Page 24