Research subjects should have access to their own data

My contribution to Genomics Law Report's superb "What ELSI is New" series is up now.

The gist of my argument: as we move into an era of large-scale whole-genome sequencing studies and the utility of genomic information grows, researchers will increasingly frequently be faced with the discovery of highly medically relevant information within their subjects' genomes. Yet under the consent procedures established for most modern genetic research projects the anonymised subjects would never have a chance to learn about this information.

As such, people will die from breast cancer and other diseases even though someone out there possesses genetic information they would have needed to avoid this that could have been used to guide additional screening to detect such cancers early.

Throwing medically actionable data away without the subjects ever having a chance to learn about it is unconscionable - we need to start changing the way we think about the rights of research subjects to their own data.

It's important not to understate the challenges posed by returning research data to research subjects - such as balancing against the need for privacy and data security, or unduly alarming people with inaccurate results - but these are not reasons to avoid data return. They are simply obstacles we need to overcome.

rss-icon-16x16.jpg Subscribe to Genetic Future.
i-1e8735341225e739a7862450baf40589-twitter-icon-16x16.jpg Follow Daniel on Twitter.


More like this

I'm slowly catching up on genomics news from the last week - this story in particular has been getting a lot of press. The executive summary: Jay Flatley, CEO of genomic technology manufacturer Illumina, predicts that whole-genome sequencing of newborns will become routine within a decade. Flatley…
In a comment on my previous post, Tera Eerkes is skeptical about the utility of routinely performing whole-genome sequencing on newborns: I found this comment absolutely fascinating, given the recent reports on translational analysis, that indicate an actual lack of clinical utility of KRAS…
Lupski, J.R., et al. (2010). Whole-genome sequencing in a patient with Charcot-Marie-Tooth neuropathy. New England Journal of Medicine advance online 10.1056/nejmoa0908094 Roach, J.C., & et al. (2010). Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science : 10…
Over at the 23andMe blog The Spittoon, company co-founder Linda Avey expands on her vision for a novel model of genomic research, in which personal genomics customers contribute their genetic and health data to fuel research into the inherited and environmental triggers for disease. This is a model…

"As such, people will die from breast cancer and other diseases even though someone out there possesses the crucial genetic information they would have needed to avoid this."

I think this is overstating it a little bit: knowing some carries a BRCA mutation is not sufficient to prevent her death. I agree that there is meaningful medical intervention that could be done in such circumstances, but we're not yet at the point of magical gene therapy.

That being said, this is an incredibly important topic, and something Sanger is actively working on as we start to undertake these studies.

By Jeff Barrett (not verified) on 03 Nov 2009 #permalink

Hey Jeff,

Fair point - that was very badly worded. I've toned it down.

I also should have made it more clear that there are plenty of good researchers already thinking about these issues; I'm looking forward to seeing the types of solutions that people come up with to resolve the fairly daunting challenges involved.

Excellent article, Daniel, and a great counterpoint to the nonsense coming from the likes of DR. STEVE MURPHY THE GENE SHERPA.

At the recent TEDMED and companion BIL:PIL conferences, the question of whose data is it was the major topic of discussion. Among many great talks was Gilles Frydman's where he shared some interesting data from ACOR where patients sharing data had uncovered information that later changed clinical practice.

Ok lets stay calm and above all, let's keep some perspective here... Yes, there are reasons for and against returning research data to participants, but if on the pro side you are claiming that it will help overall health that is a relatively weak argument for the public at large. Looking at the big picture, closing all the McDOnalds and Burger Kings would do way more good for most people's health then giving them a whole bunch of data that neither they, nor most people (including scientists) can decipher. DOwnplaying the environment in all these genetic/genomic discussions is a big mistake and a real weakness of much of the "experts" discussing these issues now.

By HC Howard (not verified) on 03 Nov 2009 #permalink

You can't handle the truth.

Thanks Daniel.

Your commentary came at an opportune time because, as you may have seen, the issue of returning research results to participants was raised again at last week's TEDMED conference. This time it was Anne Wojcicki of 23andMe criticizing Kaiser Permanente's RPGEH study for not returning genetic data to their planned participant population of 100,000+ Kaiser patients. Today, Kaiser responded.

I've written up the salient points in a brief post over at the GLR for anybody that is interested:…

Definitely an interesting discussion.
However, most of the genotyping platforms used are 'for research-purpose only'. Many labs are not CLIA compliant. How many samples are thrown out from GWAS during QC for gender inconsistencies or others due to potential mix ups?
We should not confuse research with diagnostic services.

Dan Vorhaus quotes Cathy Schaefer's defence of the RPGEH no-return position:

We inform all participants about the purpose of collecting genetic and other information, and they volunteer to participate â at no cost to them â to facilitate this research, knowing their individual results will not be returned to them and that all data about them will be âde-identified.â We also inform participants that if we discover something in their data or samples that may be important to their health, we will contact them to learn if they want to have the information.

Why doesnât the RPGEH restructure so that it can return results to individuals? Because genetic information obtained through todayâs genome-wide studies has not been designed to be useful to individuals; it is designed for use in research. In most cases, it isnât known whether the variants tested for are actually implicated in a disease process, or are markers for variants that play a role in disease, and results from these tests are rarely actionable.

Were it not for the bit I bolded, it would be the usual inadequate defence of convenient position. It kinda made sense when we were measuring a few microsatellites, a bit less so with GWAS - though I did go through the Affy chip before we started the WTCCC to make sure we weren't going to deliberately test any known diagnostic SNPs and knowingly give ourselves this problem (*). But the position is fairly nonsensical with WGS, which can hardly fail to throw up the odd significant health finding.

But the bit in bold suggests RGPEH have considered whether it would be ethical to withhold information if it mattered, which suggests experience - it is easier to maintain a hardline stance if you've never had to deal with the consequences. Many years back I was working on a project where one of the "controls" was positive for one of the nastier BRCA mutations: we realised that we would have to find a way of telling the subject, even though we had said that we wouldn't - we routed this back through the ethics committee that approved this study, and so on to the subject's physician.

At which point, this reverts to being a discussion about paternalism - which is less absolute, and more a matter of personal choice: whether you would, or would not, sign up for a project where someone else gets to decide whether any findings are important enough to pass back to you.

(*) I also checked that there was no overlap between the GWAS chip and the police's DNA database markers, but that's another story ...

Ok, Gunn, you gotta stop.

Seriously. I have never said research subjects shouldn't get their data.......

Stop it now.

What my position is

1. If the study provides for releasing of data in a way which is likely to not harm the subject, OR if the study accounts for that likelihood of harm and has protocol in place to mitigate that harm. Then, results would meet the criteria of an IRB to be released.

2. Studies that have no IRB guidance shouldn't really be studies on humans.......It tends to erode the importance of the IRB. Which we all know was created to prevent atrocities such as those in Hitler's Germany and in Tuskegee.....

Seriously, I know IRBs have been frustrating scientists for years, but they serve a societal need: A governor on crazy or unethical scientists!!!