Monday, March 25, 2013

HeLa cells and cavalier scientists

Last week came the news that the HeLa cell genome had been sequenced and published.  HeLa are cells that were taken from the tumor that eventually killed Henrietta Lacks 60 years ago, without her knowledge or consent, immortalized and used in thousands of labs around the world ever since.  They've been used for everything from significant medical research to teaching students to do lab work.  The story of Henrietta Lacks and HeLa cells was told by Rebecca Skloot in her best-selling book, The Immortal Life of Henrietta Lacks.

The sequence is, according to a story in Nature, "a mess." While this had been known to some extent from previous work, the research team
... confirmed that HeLa cells contain one extra version of most chromosomes, with up to five copies of some. Many genes were duplicated even more extensively, with four, five or six copies sometimes present, instead of the usual two. Furthermore, large segments of chromosome 11 and several other chromosomes were reshuffled like a deck of cards, drastically altering the arrangement of the genes.
Some of the chromosomal rearrangement was presumably because the cells were taken from a tumor and other aberrations were probably introduced by 60 years of large-scale cell division in labs around the world.  Chromosomal duplications, losses, and rearrangements along with other forms of mutation are expected--even may have been in the cells originally obtained from Ms Lacks.  One caution this leads to is that no two labs using "HeLa" cells are using cells with exactly the same genome.  Is this important to know, and does it justify this sequencing effort?  Those are two separate questions.

The variation the sequencing documented doesn't mean that personal genetic information is not discernible from the data.  Even duplicated bits, as well as bits not duplicated, will have the detailed sequence (plus some additional mutations) that Ms Lacks carried, and alignment of these parts, and they would be extensive, would make it possible to say many things about her.  She would carry sequence details known to be found in Africans but not Europeans (and, since she probably had some European admixture, a subset of clearly European-derived ancestral sequence as well -- African Americans have about 15% European ancestry, due to the rape of slaves, voluntary mixed mating and marriages, etc.).

Genetic variation related to some personal traits could easily be found if she carried them.  Variation related to skin or eye color, or to susceptibility to malaria, and others could easily be found if she had it.  Enough individual sequence would be easily available that, were her sequence (even before her cancer arose or the cells were distributed and grown in laboratories; or the cells at her original biopsy if still available) would make it easy to align the sequence obtained in a lab and find the match.  From a technical point of view, there is no privacy with this kind of data.

In yesterday's New York Times, Skloot tells this new story, of the sequencing of the HeLa cells, again done without the consent of anyone involved.  And the story is as disturbing as the first violation of Henrietta Lacks' privacy.  Not in the least because so much of the major media that covered the story, Nature, Scientific American (which republished the Nature piece), The Atlantic, Science 2.0, and so forth, ignored the ethical issues.  

The largest issue, of course, is that not only does the genome of these cells contain information about Henrietta Lacks, but she had five children, and they each have a 50% chance of having inherited any risk alleles she might have carried (for cancer, say).  They weren't asked whether they wanted to know their risk, and they weren't asked whether it was all right with them if that information was made public.  The data have now been taken off-line, but not before it was seen, and downloaded, by many.  

As Skloot points out in her piece in the NYT, nothing that was done was illegal--either in the sequencing of the cells or the publishing of the data, or when her cells were originally taken.   It's just that it wasn't right to publish the sequence, as the original use of the tumor wasn't right.  It's an example of the cavalier attitude of science.  The researchers had absolutely no business publishing the sequence, and it's astonishing that they did, given the widespread airing of the ethical issues surrounding the history of the HeLa cells.

And in a way to compound the felony, while we might on the surface think it honorable and exemplary that the investigators took the data off line, they are reported to have been seeking to get family permission to post it again.  But even just keeping the data to themselves is dishonorable from a human subjects point of view: neither Lacks, of course, as she died long ago, nor her family, as far as we know, consented for the lab even to do the sequencing in the first place.  The honorable thing, really, was not to have done it, but having now seen the issue raised, the data should be erased from all computers and entirely discarded.

Of course, life is complex and scientists have power (unlike the general public) and we'll get our way as a rule.  We'll find a post hoc way to justify, rectify, bully, bribe, cajole, or whatever to get to do what we want to do.  This doesn't make the specific Lacks sequence story special or in any way specially culpable, as similar kinds activity are occurring widely.  Indeed, things are more subtle and complex than that....

We blogged back in 2010 about the HeLa story, not long after the Skloot book first came out.  We were a distinct minority (of maybe two) who believed that the book, while telling an important story, was too much a continuation of the invasion of privacy that began when Henrietta Lacks' tumor was taken from her by scientists.  Skloot has, we understand, set up a foundation to help the family and donates some proceeds from her book to it.  But she is making her career on the back of the Lacks family, who she readily admits to having had to essentially coerce to let her tell their story, no matter that she also is trying in her way to help the family.

In a more pure world, she would give all proceeds to the family or, better, would have realized during the process that the book should not be written.  Due to the arrogance of science, and journalism, this family's privacy has been violated over and over again.  Whether Skloot does, in the net, help the family as well as herself, as she may, the ethical dilemmas persist: is it OK to make a book or movie, for profit, of any story so long as we try to compensate for the ethical violations in some way--even in cases where no such compensation would otherwise be available?  This is a question that anthropologists should ask themselves all the time, though too few do.  Here is where the ethical rubber meets the road--where there are no easy answers.

This new chapter of the story is more troublesome to many than was the book, it seems, presumably because it's much less ambiguous.  Everyone now seems to agree that this personal genetic information, that the family hadn't even asked to know themselves, shouldn't have been made public without their consent.  And this along with other stories shows how non-private DNA sequence may be, even if an actual name is not attached to it on a web page somewhere.  Apparently this hadn't occurred to the scientists who did the HeLa sequencing.  Nor had they thought of 'private' meaning not even known to investigators who should refrain from even peeking at it without explicit permission.  Another example of the cavalier attitude of too many scientists.  


Anonymous said...

Why not give free access to decent health care to all of Henrietta Lacks offspring - say for the next sixty years ?
Sponsored by companies having sold/selling HeLa cells, by major national and international cancer research funding agencies, by journals publishing HeLa-cell-generated results ...

Adelaide Ungerer

Ken Weiss said...

It's obviously the right idea. But sadly it should not be necessary. We should have national health care for everyone. It's shameful, absolutely, that we don't.

But even that won't accommodate the way her family, and countless others past and to come, for science's liberties--or, better, for science to be realistically forthcoming, and self-restrained, about what they are doing, and accept that we can't do all that we might want to do and/or sincerely believe is perfectly innocuous.

Markus Perola said...

I agree with the general idea of this being non-ethical, but then again, is this a real issue? The relevant health information that one can extract from raw sequence data is very limited...

Anne Buchanan said...

Absolutely, I agree. It's the ethical issues of invasion of privacy and arrogance of science that are at play here, not what could actually be predicted from the data.

Ken Weiss said...

There is a historical aspect to this because many different kinds of human cells could be used in research these days, anonymously. That wasn't true back when HeLa were first available. Perhaps calling them 'HeLa' seemed anonymous enough at the time.

So there are several issues: the ethics of exploitation for publicity or gain by authors, even if they partly 'compensate' the family; the scientific knowledge gained; and whether if the cells had been called XY371A, totally depersonalized and with Lacks' name erased from any identifiability, it would have been ethical to study the cells at all without explicit permission.

Permission in those days was much less regulated than it is now. But scientists and others do often take advantage, when they can

Anne Buchanan said...

Jonathan Eisen does an excellent job with this whole issue over on his blog, The Tree of Life. He includes a Storify including tweets by Rebecca Skloot about the ethics of the issue, and a link from Skloot to a presidential commission report on bioethics, including discussion of the ethics of old samples for which consent for modern analyses was never given.

Holly Dunsworth said...

Great piece, Anne.

Here's where my fear of fear comes in though...

What are we keeping private? Aside from a few smoking guns, the genome is entirely probabilistic, not to mention only probabilistically similar in relatives.

And if Henrietta had said okay, should she have had to get consent from all relatives to truly consent her participation? And which relatives? How far distantly does a person have to go to "own" the privacy rights, to share publicly, their biological information. I didn't ask my parents before sharing my SNPs with 60 and counting people on 23andMe.

Ken Weiss said...

One way would be to have strict policy by which your doctor (or any relevant person you delegate) can look at your DNA, in parts or for reasons you agree on, and nobody else or no other parts. Period.

And insist that it be destroyed afterwards.

These things are in place in this or that way, here or there, but nothing systematic. However, as you say, once public, it extends its reach, intercalating into others' privacy.

Or, of course, we could just agree that, like your photo, or fingerprint, or whatever, that we are used to stuff being more or less public and this isn't anything really different.

From your photo we can guess about other parts of your anatomy not visible. We can look at your body size and shape, skin color, and so on, and tell things about your ancestry, or about your disease risks that (as, also, we can tell, probabilistically (as with DNA), by knowing semi-public things like what you eat, whether you exercise, drink, or smoke, etc.

Maybe the main idea is truly informed permission. Certainly, however, DNA introduces more serious prying than many other things (such as photos).

Of course, in our evolutionary ancestry, in a small tribal deme, nothing was secret.

Anne Buchanan said...

I agree, all great questions and points, which need to be hashed out by the genetics community, but also by consumers of the technology at large. I guess the presidential commission report on bioethics is a start, but as the well-kept secret it has been, not a great one.

I think part of the problem in all this has to do with the widespread assumption that genes are determinative. Thus, our genomes harbor secrets that perhaps even we ourselves don't want to know. But, as Markus Perola points out above, and as you point out, that's a misplaced assumption.

So, the invasion of privacy issue with respect to what we can learn about Ms Lacks' relatives from the HeLa genome is more a concern in principle than reality. Indeed, as has been pointed out elsewhere, the cells have been sequenced in bits all over the place already. The bigger issue is the scientific hubris with which these tumor cells have been treated, from start to finish.