It is expected that genetic sequencing will advance both specialised and general healthcare leading to more personalised care based on an individual’s genome. It can be used to identify disease-specific mutations and those associated with more complex conditions. However, as discussed previously on this blog, understanding the effect of a single mutation can be difficult without comparison between healthy and disease patient samples. Sharing patient data accumulated by private and academic labs, voluntary databases, healthcare providers and large scale sequencing efforts such as the 100,000 genomes project, can provide researchers with the larger datasets required for accuracy.
Sharing personal data, in an anonymised form, with research partners can only be done with patient consent. When asked about willingness to share data for research the majority (75-90%) of patients with an existing disease would agree (Genetic Alliance UK, 2014, Darquy et al., 2015 (Europe), Oliver et al., 2015 (USA)) compared to only a third of healthy participants (Sanderson et al., 2015 (USA)). This discrepancy is most likely the result of different motivations between patients and families with a disease – who rely on research to advance understanding and treatment options – and those interested in their own general health.
One of the main reasons given by healthy patients for not wanting to share data is the risk that anonymised data could be re-identified with negative connotations for the patients and their families. This is also a common concern for disease patients where potentially, due to a limited patient population, simply having knowledge of the primary disease and local care provider could enable patient identification. Re-identification is an unavoidable possibility and patients must assess the acceptability of such a risk.
Some databases are also shared with commercial partners and in surveys most rare disease patients were happy with this, acknowledging that treatment development requires such collaborations. However, some expressed concerns over whether the results from commercial research would be made publically accessible for other groups to use or would be kept proprietary and used solely for profiteering (Genetic Alliance UK, 2014, Darquy et al., 2015).
Generally patients report that they would want to know what research is being conducted with their data and have access to any results including incidental findings (Genetic Alliance UK, 2014, Darquy et al., 2015). However, disclosure of incidental findings, which could impact on patient health, should be handled by trained professionals who can explain the potential implications to patients – this may be difficult for some research groups to ensure. Some research programmes therefore will not return this information to patients, whilst others, including the 100,000 genomes project, will only return consequential findings that are actionable (Genomics England). Details regarding disclosure of incidental findings should be included in consent terms.
Patients who agree to share data often give “broad consent” for their data to be shared with approved users with valid research questions, rather than individual consent for every project. For research outside this remit additional consent would be required with patients having the option to decline. An upcoming study by Genentech using Parkinson disease patients’ data submitted to genetic testing company 23andMe will require additional patient consent as it requires in-depth sequence analysis of individual records rather than being based on the existing anonymised and aggregated data (Adam & Friedman 2015).
Whilst sharing genetic data is important for progressing research donors should ensure they are personally comfortable with how, when, who-with and why their own personal data will be shared before consent is given. They should also be aware of if and how incidental findings will be reported, and the potential impact of such results.
- Adam S, Friedman JM (2015). Individual DNA samples and health information sold by 23andMe. Genet M Jun 18. PMID: 26087174.
- Darquy S, Moutel G, Lapointe AS, D’Audiffret D, Champagnat J, Guerroui S, Vendeville ML, Boespflug-Tanguy O, Duchange N (2015). Patient/family views on data sharing in rare diseases: study in the European LeukoTreat project. Eur J Hum Genet. Jun 17. PMID: 26081642.
- Genetic Allience UK (2014). What do patients with rare genetic conditions
think about whole genome sequencing in the NHS? http://www.geneticalliance.org.uk/docs/final-100kgp-summary-report-12-11-14.pdf
- Oliver JM, Slashinski MJ, Wang T, Kelly PA, Hilsenbeck SG, & McGuire AL (2012). Balancing the risks and benefits of genomic data sharing: genome research participants’ perspectives. Public health genomics, 15 (2), 106-14 PMID: 22213783
- Sanderson SC, Linderman MD, Suckiel SA, Diaz GA, Zinberg RE, Ferryman K, Wasserstein M, Kasarskis A, Schadt EE (2015). Motivations, concerns and preferences of personal genome sequencing research participants: Baseline findings from the HealthSeq project. Eur J Hum Genet. Jun 3. PMID: 26036856.
One thought on “Sharing genomic data to advance research”
In thinking about this issue, people tend to focus on government and nonprofit medical research use of genomic data. Of course 23andMe is for-profit, but it has fairly successfully convinced its customers that any medical research it does with their data is for the public good. Only recently, with disclosure of some patents and business deals with pharma companies, have their customers started wondering “Hey! Maybe my own personal DNA data ISN’T really being used for public good.” (Though of course one can argue that therapies that can cost patients $100,000’s per year -costs born by customers, insurers, and the government- ARE developed by pharma companies for the public good.) Of course with for-profit companies there is more motivation than with government agencies and nonprofit medical research to dissemble, to make the ultimate use of the customers’ data sound much more noble than it really is.
A glaring example of this is AncestryDNA/AncestryHealth, who just announced a collaboration with Calico, an offshoot of Google/Alphabet with goals in very laudable arenas. (A major project is a collaboration with Peter Walter at UCSF regarding stress response modulators. https://en.wikipedia.org/wiki/Calico_%28company%29#Modulators_of_Integrated_Stress_Response The stress response has been mentioned on this blog many times. Peter is an absolutely brilliant researcher who deserves support, who I was honored to collaborate with and who coauthored a paper with me.) Notably, Calico is clearly a division of a for-profit company.
The point of the announcement by Ancestry was to prod customers to think that their health, pedigree, and genomic DNA microarray results will be combined and used for public good. The branch of Ancestry involved is “Ancestry Human Diversity Project;” even the name hints that it’s for public good. However the Informed Consent form gives almost NO restrictions on the use of Ancestry’s data by third parties, including marketing firms etc.
Ancestry is famous for NOT being transparent, for even hiding the “Informed Consent” form. As Judy Russell said, “So when AncestryDNA says that many hundreds of thousands of its more than 1 million test takers have agreed to participate in the research study, I don’t believe for one minute that more than a tiny fraction knew that (a) they didn’t have to agree and (b) if they did agree, they were agreeing to disclose every last jot and tittle of their family history.”