genomeboy.com

A few weeks ago the Genetic Genealogist gave us a lively summary of the issues raised by this paper from Stanford’s Hank Greely. If you haven’t seen it, I think it’s probably as damning an indictment of current practice regarding human genetic research subject confidentiality and anonymity as you’ll ever read.

The current Nature Genetics features a commentary from the GAIN Collaborative Research Group (a public-private consortium of investigators carrying out genome-wide disease association studies) that offers a rather different perspective. Dig the disconnect:

GAIN:

“Investigators granted access to GAIN data should ensure confidentiality of study participants and follow any limitations specified by their informed consent.”

Greely:

“…plans for a Genome-Wide Association Studies database by the [NIH] illegally and unethically strip people of the legal status of human subjects…blanket consent, although extremely convenient for genomic biobanks, is both ethically and legally suspect.”

GAIN:

“Individual participant data will be kept secure and will not be distributed.”

Greely:

“…patient identity is not, and cannot be, effectively protected in large-scale genomic biobanks.”

GAIN:

“Data will not be used to identify or contact individual participants from any GAIN study.”

Greely:

“…the result of anonymizing data is not only nearly useless, but is itself unethical…”

“…biomedical research, including genomic biobanks, has a moral obligation, now almost always evaded, to inform research subjects of clinically significant information.”

You might say, “Yeah, well, Greely’s just one guy.” And I might say, “Yeah, well, read this article and the 15 commentaries that follow it and then talk to me about consensus.”

I don’t see a Kumbaya moment happening anytime soon.

A brief housekeeping note: genomeboy.com is now a member of the fabulous DNA Network, a one-stop shop for all of your molecular bloggerocity.

This afternoon I elbowed my way through a SRO crowd to hear Svante Pääbo talk about genomics as a tool to understand human origins. If you follow the paleogenomics literature, I suspect not much of what he said was new. I was familiar with a fair bit of it, but thought it was fascinating nonetheless.

His first topic was the study of positive selection in humans, that is, trying to find traits (and the genes that encode them) that put early humans at an evolutionary advantage. Obviously, it is those things that are likely to help distinguish us from our nearest primate cousins.

The second part of his talk was on the Neanderthal genome, which his group is using 454 sequencers to reconstruct. My God, what a painstaking job: throwing away 98% of what you’ve got because of contamination from modern human and/or microbial DNA and then standing on your head to make sure what’s left is truly Neanderthal. Brutal.

But what made me nearly fall out of my chair was when Pääbo talked about what he says is evolutionary genomics’ greatest unfilled need: rigorously defined human traits. That’s right, clean phenotypes! Sounds familiar, no? I imagine every PI doing genetic and genomic research on actual living human beings has more or less the same complaint. Think about your own medical records (assuming you have them at all and have seen them, which are two huge ifs): how valuable are they to your own health, let alone to science? From what I can tell, mine are worth bupkes (sorry, George!). Somehow, after schlepping out of Africa 100,000 years ago, you would think we’d've made it further than this…

However smitten I might be with technology, however high I’m prepared to fly the flag of the early adopters, I must admit I do have my blind spots. Exhibit A: the LongPen™.

When Margaret Atwood, brilliant writer and peerless creator of dystopian worlds (including a chilling one wrought by genomics), said she had developed a device that would let authors sign autographs remotely, lots of folks thought it was a joke.

It’s not.

But after watching the propaganda video, it strikes me as the kind of surreal technology that might spring up in one of Atwood’s own novels. Again, I’m hardly a Luddite (what, you thought Genome Boy wouldn’t give up his buggy whip?), but must every face-to-face human interaction now compete with a virtual one? “Yes yes,” they say, “…but you don’t understand: the LongPen™ is green!”

Whatever, I’m sure it is. But is this really about environmentalism? Is that why you talk about the “democratizing” powers of your new device, one that happens to spare you from sharing oxygen with the hoi polloi and exposing yourself to their nasty germs via a handshake or–shudder!–an embrace? Or is it just that, as you say in the video, sometimes you “don’t want to deal with room service?” Yeah, well, I’m told the Four Seasons can be just brutal…

God forbid you should get off your arse, venture out of your gilded garret and go some place to meet the occasional throng of adoring fans on someone else’s nickel.

(hat tip: Rake)

“I’m not doing it just to satisfy myself. I’m doing it because I know that, if the audience only gets this story and the way that it is written, their lives are going to be changed.”

- Sir Ian McKellen on performing Shakespeare

I think most if not all of us PGPers are motivated by a desire to begin to desensitize people to their own genomic information, chip away at notions of genetic exceptionalism and convince folks that, even in a climate where the 14th Amendment might have become a hazy memory, public release of individual DNA information need not inspire fear.

I also think that, even when we disagree with him, we all have real affection for and trust in George.

Not long ago I gave someone something I’d written about the PGP in which I went on at length about him. She didn’t get it. “What is it with this guy? Why the man crush?” I’m not sure I’d call it that, but I suspect even his fiercest detractors (he has a few) would admit that he is a charismatic dude. This is a guy who was adopted twice, flunked out of Duke, and somehow landed on his feet at Harvard, in Walter Freaking Gilbert’s lab no less. Yet to meet him is to meet someone warm, outgoing, at peace, and who moves with a kind of Zen deliberation–not your stereotypical high-strung gene jockey.

And he’s unique in other ways. Not only has he distinguished himself in sequencing technology development, he is making a real effort to figure out how that technology will actually be used by average people en masse. It seems to me that just about every PGP communique is marked by variations on the same refrain: “When we get to a million” or “How do we scale this up?” And it’s that failure to recognize the PGP as a template rather than as an end in itself, I think, that may be the biggest misunderstanding surrounding George, or at least, the one that makes me the batshit craziest. Those who insist that George has assembled a little band of “celebrities” (I wish) pursuing “vanity genomics” just like Craig and Jim, and that this whole enterprise is a tragic misuse of the noble cause of human genomics (summarized beautifully here), don’t really understand either George or the project.

Could I have gulped down the PGP Kool-Aid in too much haste? Could my man crush be a manifestation of Stockholm Syndrome? Perhaps. But to suggest that the nine of us are somehow Paris Hilton to George Church’s Steve Wynn seems to me a bit of a stretch…alas…(I mean, I could do without the video and the jail time, but let’s not diminish Ms. Hilton’s achievements as a recording artist and author. Heh.)

Ultimately history will judge the PGP by what happens when people get their data and begin to make sense of it. Or not: How will the project deal with bad news lurking in subjects’ genomes (and not just the first ten, but the next thousand)? I’d be lying if I said I am feeling completely sanguine about that prospect and what it might mean for my family.

But for now I take no small comfort in the knowledge that the buck stops with George.

Or, as my Texan father would say, INsurance…

I will get back to George, but this article in The Economist captures some of both the heat and the light surrounding personal genomics and deserves comment. This passage in particular struck me:

The problem arises when one looks to the medium term. If genetic information is kept secret from insurers, but individuals have the freedom to add or drop insurance coverage, then problems of adverse selection may arise. People will be tempted to “game” the system. Those who test negative for a serious and costly disease may drop coverage, while those who test positive may add or increase coverage. Insurers worry this will lead to a collapse of their risk pools, and ultimately to financial ruin.

Well, they should be worried. As my colleague Bob Cook-Deegan notes in the article, folks with Alzheimer’s susceptibility alleles are five times more likely to alter their insurance coverage than those without them.

And some would argue they shouldn’t even wait to get tested: Not long ago, I interviewed my own physician. I asked her if she had any wisdom for me given that I would soon have my genome sequence made public. I expected generic advice about looking for susceptibility alleles and modifying my lifestyle accordingly. Instead she said, “I would load up on disability insurance if I were you. Buy as much as you can afford.” I was gobsmacked because 1) it seemed to me a deeply cynical response to the Genome Revolution from my doctor and 2) it implied that insurance companies might actually pay attention to personal genomics in the near term, which I still find hard to believe (although that, I suppose, is a deeply cynical response from me). But the Economist is right: of course we will game the system. That’s capitalism, baby. And if Jerome Bettis can fake an injury, anything is possible…

To me, the answer seems obvious (which is why I’m not a health care economist): a single-payer system coupled with a deCODE-style pool where, unless you opt out, your genome (and perhaps some fraction of your phenotype) becomes part of a national database that is made available for biomedical research. The benefits of that research would then be returned to citizens at no cost. Of course, as George Church is learning, genetic exceptionalism is not easily overcome. DNA is still the bogeyman and most people don’t want to show you theirs. As another colleague, philosopher Alex Rosenberg, once said to me, it will be tough to get people to trade autonomy for beneficence. But that doesn’t mean we shouldn’t try.

Update: Esther offers her take.

I have yet to articulate the appeal of the Personal Genome Project in a way that really captures how I feel. I think my fellow subjects have done a better job of it than I, especially here (subscription only–sorry!), where I think I managed the neat trick of coming across as both flaky and mercenary:

To Misha Angrist, a science editor at the Duke Institute for Genome Sciences and Policy, it was less practical considerations than making his own genome more tangible. Angrist, who has a PhD in genetics, said he pipetted many DNA samples over the years, “and it was still always kind of an abstraction. Maybe I am naïve, but I’d like to think that having a look at my own genome might make it less of an abstraction.”

Angrist said that he plans to write about the project, which also factored into his decision to participate. “I think it’s much more interesting to write about something from the inside, as a participant observer.”

I stand by both of those statements and I don’t fault the reporter at all, but they are only a small part of what I was trying to say. As someone who presumes to call himself a writer, I should have known better the peculiar corollary of Murphy’s Law–the best parts of a 45-minute conversation will rarely make it into print; the dorkiest parts almost always will. So let me clarify: the PGP is, in my biased view, at the vanguard of efforts to understand personal genomics and its impact on health care, insurance, personal privacy, genealogy and all the rest. In my professional life I think about these things every day. But rarely are they any more than a thought experiment, that is, an abstraction (there has been one notable exception I hope to address in a future post).

This is not to say that when my genome goes up on the internet I will suddenly be overwhelmed by a sense of impending doom. Hardly. But the PGP is not without risk to subjects (and to their family members, another topic of a future post). A variety of bad things could happen that would be anything but abstract. When the ten of us signed the PGP consent form, those things (or at least some of them) were outlined in fairly stark terms:

…anyone with sufficient knowledge could take your genome and/or posted medical information and use them to (1) infer paternity or other features of your genealogy, (2) claim statistical evidence that could affect your employment or insurance, (3) claim your relatedness to infamous villains, (4) make synthetic DNA and plant it at a crime scene, (5) reveal the possibility of a disease or unknown propensity for a disease.

Holy Sally Hemings, Batman. Can you imagine five more ready-made plots for a Lifetime Movie? At first glance, they seem remote, perhaps even comical (”Hey, uh, Mom, did you know you’re descended from the bastard son of Blackbeard?”). But remote is not the same as implausible. And so here is another reason I think the PGP is so appealing: truth in advertising. I’ve read a fair number of consent forms; all use the Latin alphabet, but how many are actually written in English? They often portray risks as vanishingly small or couch them in scientific jargon. Be honest, Dr. Principal Investigator: Is Billy Bob Study Subject really gonna parse your 12 single-spaced pages and understand all of the things that might go wrong? As George Church told me in an interview last year, “Most informed consent forms tell you everything you need to know except when to quit.”

George, I should say, is another appealing aspect of the PGP (so far, at least). I will try to introduce you to him through my eyes in the next post.

And the award for the 2007 Genomics Article Title Most Likely to Appeal to Twelve-Year-Old Boys goes to…

I should back up for a minute; I don’t want to assume that everyone is already hip to the PGP and the broader concept of personal genomics or why they should give a transgenic rat’s patootie. Fortunately Dana Waring and Jack Bateman from Ting Wu’s lab at Harvard have created a website that does much of the heavy lifting by taking up questions like these: What is personal genomics? What will it mean for medicine? What are the risks and benefits? What are the ethical issues? What are the technologies behind it and when will all of this become reality? The site is already a useful primer on personal genomics for the newbie; I suspect it will soon become much more.