wikimedia.org

wikimedia.org

Earlier this year, a research group revealed that they were able to identify 50 individuals whose genomes were sequenced as part of the 1000 Genomes Project. This wouldn’t be a problem, except the samples were supposed to be anonymous.

Led by Yaniv Erlich at the Whitehead Institute, this test of true anonymity was accomplished using publicly available sequences and genealogical information. The genetic sleuths did not actually reveal the individuals’ names, but the ease with which they identified “anonymous” samples makes this a cautionary tale.

Genomics promises huge societal benefits. Personalization of medicine would allow doctors to tailor therapies for each individual patient. During medical check-ups in the near future, a nurse might provide your doctor with your DNA sequence, along with your blood pressure and temperature, as another tool for diagnosis.

However, a great deal of genomic research must be done before personalized medicine becomes a reality. With this incredible new technology comes the need for a clear ethical framework for research. The current haphazard system, created on the fly alongside advancing genomic technology, fails to sufficiently help research participants understand what they might be getting themselves into – and to what hazards they may be exposing themselves.

The participants unmasked in the 1000 Genomes Project signed an informed consent form that notes uncertain privacy, given the unpredictable nature of technological advancement. However, the danger of identification is underplayed. The form stated that identification should be possible “only if somebody knew that you had given a sample to be studied for this project.” This is the standard language of the field. Indeed, this consent form is one of several provided by the National Human Genome Research Institute as templates for researchers. All of these templates frame the possibility of being identified as highly unlikely.

The Erlich study shows otherwise. And this means that the standard bar for informed consent is no longer sufficient. To understand why, we need to look at how the ease of genome sequencing has changed in the past decade and a half.

The first draft of the human genome was the result of the massive, publicly funded Human Genome Project. It cost roughly $3 billion from start to finish to get that one genome sequence. Today, a complete human genome can be sequenced for under $10,000. Sequencing of just the “exome,” the parts of the genome that actually encode for functional proteins, can be purchased from 23andMe for a mere $999. And it’s only going to get cheaper and, thus, more popular.

As the availability of genome sequencing expands, the number of curious individuals who have their genomes sequenced will increase exponentially. But if future employers, or life insurance companies, or health insurance providers can start using genetic information to assess risks, a simple act of curiosity today could have drastic consequences. Genetic discrimination is a possibility. It is therefore vital to inform these people about the extent to which they can expect their genetic privacy to be maintained.

Some prominent scientists have taken the lead on a “new” informed consent for genomics. George Church, a top geneticist at Harvard University, runs the Personal Genome Project (PGP). In introducing itself to potential research participants, the PGP uses dramatically different language than traditional consent forms. It is bluntly up-front about the difficulties of genetic anonymity, stating:

“While the Personal Genome Project will not intentionally associate your name with your genomic, trait or other data … the most ethical and practical solution at this time is for PGP participants to be enrolled with the expectation of full public data release and without any promises of privacy, confidentiality or anonymity.”

The PGP does not dodge their role in protecting privacy; elsewhere, the website notes that their researchers will take normal precautions to anonymize participants’ data. Where PGP differs from other programs is admitting one inescapable truth to potential participants: as genomic technology advances, it will be harder to truly ensure one’s genetic privacy.

Federal agencies that fund genomic research, such as the National Institutes of Health, and in particular, the National Human Genome Research Institute, should use the PGP consent system as the new model for genomic research. In the short term, this approach may scare off some potential research participants. However, it ultimately protects both subjects and researchers alike. It may also have the hidden benefit of encouraging “citizen science” as access to data can be broader with fewer restrictions on access to metadata.

Regardless, trust between the public and the scientific community is crucial for the long-term prospects of science. Being open and honest about the new world of limited genetic privacy is crucial to maintaining that trust.

Peter Yang is a PhD candidate studying genetics at Harvard Medical School.