Covid-19 between genomic mapping, genetic profiles and privacy

The 23andMe platform 

Recently on the weekend supplement, Life&Art, of the “Financial Times” there was an entire page dedicated to Anne Wojcicki, co-founder and CEO of 23andMe. A high-tech company with a seemingly bizarre name but today very well positioned to become one of the most important platforms for predictive medicine and personal care. 

The global pandemic could only turn the spotlight on 23andMe and Anne Wojcicki. Not so much for the service it offers, that is, the mapping of the personal genome or the spontaneous collection of genomic data, as for the databases it builds during this process. A rather delicate activity, as can be easily understood, but which doesn't seem to worry Wojcicki in the least, as she says she feels "on the right side of history". 

Technology can only help people to learn more about their biology and researchers to find the cures that are missing for genetic diseases, the FT journalist who interviewed Hannah Kuchlerl told. 

Now it happens that data, including genetic data, seems to be the key to the fight against the coronavirus. And 23andMe has information that no one else has. Information on the genetic variations of a very large sample. And, as we begin to understand, it is precisely these variations that cause Covid-19 to subtly hit a person hard, lightly or leave them symptom-free and therefore a loose cannon. 

However, 23andMe is not a non-profit company or happily dedicated to supporting searches, but a high tech company with investors such as Google, Genentech and venture capital funds. 23and Me's data collection model is based on consent, on voluntary adherence to a sort of Wikipedia of genomic information, so there is no Google or Facebook model behind it. The question then moves on to the use of data in relation to the reasons for engaging the donors of the same. This is where to investigate. 

It is precisely what they do that three professors from the University of Amsterdam (Josf Van Dijck, Thomas Poell and Martijn De Waal) investigate in their beautiful book Platform Society. Public values ​​and connected society, now also available in Italian thanks to Guerini e Associati. 

We're pleased to offer you, below, the full excerpt from the book that discusses 23andMe's data purpose and engagement model. 

Nothing new: technology is a two-faced Janus. 

A large hub of data 

The 23andMe platform began operating in 2006 as a personal genome analysis service, offering customers around the world a map of their DNA; ten years later, the platform was one of the world's largest genome mapping data hubs, having collected more than "320 million individual phenotypic data". 

The data is collected both offline and online. The offline method is to order a "genetic profiling service kit" from 23andMe and send a small amount of saliva; After paying between $99 and $199, customers receive a comprehensive overview of their genetic makeup, including a risk report that details how likely each individual is to have a genetic disorder or disease. 

In addition to the offline commercial transaction for genetic data testing, the company also reaches out to customers online by inviting them to make phenotypic data available through pop-up questionnaires. These supplementary data likely help compose an even more accurate profile of an individual's health status. 

From the beginning, 23andMe has wanted to promote its product as a diagnostic medical test, while the data it collects is apparently a by-product used in medical-scientific research. 

Problems with supervisory authorities 

In 2013, the US Food and Drug Administration (FDA) banned 23andMe's DNA test kits for providing consumers with inaccurate information based on predictive algorithms that turned out to be flawed. As a result of this moment of crisis, the platform has archived the medical component and shifted its attention from diagnosis to identifying the genealogy of individual customers. 

Following the FDA's intervention, 23andMe developed an "app that acts as a compass for your genome" for the Google Play Store, with the aim of showing customers "what the DNA says about you and your family" . 

Despite the new category in which the app was included, the platform's website still alluded to the purpose of offering personalized predictive medicine services. In 2014, after slightly modifying the rhetoric of its presentation, the company sought and obtained the approval of the British health authority, arguing that the kit was marketed not as a diagnostic test but as an "information product". 

From the UK, 23andMe could ship the test kit to customers in the UK and fifty other countries around the world. Since there are no global standards guidelines that allow for evaluation of a product's specifications, any company can look for those regional or national markets whose regulatory policies permit it to be distributed as a medical app. 

In 2015, the FDA approved the 23andMe test kit, limiting it to a few cases of specific ailments and pathologies, so the company was able to relaunch the modified version of its product in the United States, now enjoying the approval expressed by the FDA . 

The narration of 23andMe 

It is interesting to see how 23andMe solicits the transfer of data from customers on the basis of a double, closely interconnected argument: the promise to receive a personalized genetic profile and the promise to donate data on one's genotype and phenotype with the aim of helping research in the field of genetics and promote the common good. 

As the site reports, by sending a sample of your DNA, 

“You're not just learning about yourself; you are joining a community of motivated individuals who can collectively impact research and understanding of basic human nature» (emphasis ours). 

Thus, 23andMe appeals to users' need to express solidarity and a sense of community by evoking a "community of motivated individuals" - a term that refers to the fact that active patients or groups of users are involved in this effort. 

Despite the company's attempts to acquire patient platforms already active online, 23andMe has shown less interest in patient communities as a collective and more attention to different categories of patients as bearers of valuable data. 

As Harris, Wyatt and Kelly note, the rhetoric of 23andMe 

"It glides smoothly from notions of personalized health care to celebrating consumer participation in research as a form of 'gift exchange'." 

Participating users 

However, what may appear to be a gift exchange is in fact a data exchange, in which individual phenotypic data is transformed into economic value. When purchasing a DNA test kit, each customer is urged to grant their permission to make their genetic data available for research purposes, with a request formulated in terms of altruism and the common good: 

For scientists and researchers to accelerate improvements in health care standards, they need big data sets… from all of us. Your participation in research could help bring about scientific breakthroughs that enable disease prevention, better drug therapies, new disease treatments, and most importantly, the introduction of gene therapies. Once you have purchased your kit, you can choose to join this research revolution (emphasis ours). 

Users are addressed as "participants" in a "research revolution", which can open up a perspective for finding cures and preventive remedies for many diseases. As we learned from the fact sheet, more than 80% of customers choose to participate in the research. 

The terms 'research' and 'researchers' remain unqualified; they appear to refer to both public and private research, as if the data were made available without any restrictions to all researchers. But 23andMe users sign up to terms of service stating that it is up to the company to decide which third party customers' genetic data will be shared with. 

The privatization of data 

Already in May 2012 it had emerged that 23andMe would have privatized the revenues from the acquired data resources, since a patent for the "polymorphism associated with Parkinson's disease" had been assigned to the platform owner, sparking complaints from customers. Patients said they felt "cheated" for donating information to a company that subsequently profited from their data donation. 

In January 2015, pharmaceutical company Genetech paid 60andMe $23 million to access 23 DNA profiles of Parkinson's disease patients. It didn't take long for major pharmaceutical companies and drug product developers to start funding XNUMXandMe. 

Those who have invested in the company foresee a future where the combination of health databases will be the main resource for developing patentable drugs and treatments. When 23andMe announced it would enter drug research and development, its owners pointed out that the huge database of "research participants" would be its greatest assets in the face of stiff competition from companies " big pharma". 

23andMe's success in building a user-generated genomic database cannot be considered a separate event from what happens in the larger ecosystem of connective platforms, in which vertical and horizontal (platform) integration makes global expansion possible. 

The ecosystem of connecting platforms 

The strategy for finding the most effective return on investment model for a single app in the health sector depends in large part on a successful integration into the ecosystem. Worldwide distribution of 23andMe's genetic testing kit and genealogy app, despite regulatory bottlenecks, was made possible by its presence on the Google Play Store. 

Google's choice is not accidental: Alphabet-Google is the first and largest among 23andMe's financial investors, mainly through its affiliate Google Ventures. But the platform is not exclusively confined to the paths traced by Google within the ecosystem. 

In 2016, 23andMe also launched an Apple ResearchKit module to help researchers seamlessly integrate genetic information into app-based studies. Based on the results obtained in these studies, the researchers will build platforms for patients suffering from asthma and heart disease. 

The project is being presented as a "collaborative framework" with Stanford Medical School's Mount Sinai Asthma Health and MyHeartCounts apps; in this way 23andMe customers can upload their genetic information directly through these platforms, following an informed consent process. 

With more than a million customers worldwide, 80 percent of whom have consented to participate in research, 23andMe commercially markets its database as a gold mine for researchers. The data is not free, but researchers "will have a simple and low-cost way to incorporate genetic data into their studies" (23andMe, 2016a, italics ours). It is not specified what low cost refers to, but it is quite clear that researchers not included in the private organization will have to pay for data donated by users. 

Evolution in a global company 

As an industry platform, 23andMe has gradually evolved into a global genetic data company that is tightly interconnected with the infrastructure core of the platform ecosystem. Through the interconnection of hardware devices, cloud services, and software systems, relevant databases are gradually privatized, despite the original promise to use patient-donated information for collective goals. 

The ideal drive towards the community, which is expressed in the request to patients to donate their data to favor a greater good such as research, is transformed into an investment in connectivity, which helps companies like 23andMe to acquire value as it transforms patient data into tradable goods. As Ajana argues, shared data is “increasingly seen as a 'public good', a type of asset that could potentially benefit not only the individual but also society as a whole. 

In this context, solidarity becomes almost synonymous with data sharing and willingness to provide information. There is an ironic twist to this "data philanthropy" identified by Ajana: while patients want to contribute their data to the common good, tech companies could end up vastly privatizing common resources. 

We will return to this ambiguity later in the chapter after looking at several other examples of platforms. 

JOSF VAN DIJCK is Professor at the University of Utrecht. Previously, he taught Media Studies and was Dean of the Humanities faculty at the University of Amsterdam. She was President of the Royal Netherlands Academy of Arts and Sciences. She is the author of The Culture of Connectivity. A Critical History of Social Media (2013). 

THOMAS POELL is Associate Professor of New Media and Digital Cultures at the University of Amsterdam. His publications mainly focus on social media, popular protests and the role played by these same media in the development of new forms of journalism. He has co-authored several books, including The Sage Handbook of Social Media (2018) and Global Cultures of Contestation (2017). 

MARTIJN DE WAAL is a researcher at the Play & Civic Media research group at the Amsterdam University of Applied Sciences. His activities mainly focus on the study of the relationship between digital media, society and urban culture, with a specific interest in public space and civic media. The text by him The City as interface. How New Media are Changing the city it was published in 2012. In 2009 he was a visiting scholar at MIT, at the Center for Civic Media.