Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More
When a reporter for the New York Times set out to test three genetic tests, she received extremely varied results.
The direct-to-consumer tests set back writer Kira Peikoff anywhere from $99 to $399. In two months, she received a full set of health results about her likelihood of contracting a variety of adult-onset conditions. But each of the providers she evaluated (23andMe, Genetic Testing Laboratories, and Pathway Genomics) interpreting the results differently. In one instance, she was warned about an elevated risk for rheumatoid arthritis; in another, a provider deemed her risk to be minimal.
This isn’t the first time that journalists have evaluated these tests, and it won’t be the last. Academics like Stanford’s Hank Greely have devoted years of research to exploring the ethical and legal implications of the new biomedicine technologies, particularly those related to genetics.
Peikoff received the results just as one of the companies, 23andMe, received a stern warning letter from the U.S. Food and Drug Administration over concerns about the clinical validity of its data. This set off a small firestorm in the process and prompted debate about the future of genetic testing.
Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.
How should we interpret damning reports like this? Should we avoid genetic testing altogether? This is still such a nascent sector, so is it better to wait until the technology evolves? I asked Greely, a genetics professor and the director at the center for law and biosciences at Stanford University, to weigh in.
VentureBeat: Were you surprised that the New York Times writer received three different results?
Hank Greely: Not at all. The results were inconsistent, as the interpretations were all over the map. The analytic validity is strong, I believe. But each of these companies calls the various medical effects based on a different set of studies. What makes the results even more confusing is that the companies do not have the same baseline for what they consider “average.”
VentureBeat: If the results vary so much, do you think there’s much point in ordering a genetic test at all?
Greely: With a few exceptions, the data derived from SNPs [pronounced snips] won’t show us anything particularly powerful or clear. You need full genome or whole exome sequencing [sequencing of the exome – the protein-encoding parts of all the genes – is beginning to dominate the headlines, thanks to its ability to diagnose diseases that were previously undiagnosable] rather than SNP chips, to get truly meaningful results. Exome and genome sequencing is not yet the norm. So for now, I think it’s better for people to save their money and not purchase any of these tests.
VentureBeat: 23andMe is now offering consumers their raw data. It’s not interpreting the results without FDA approval. Is it worthwhile to have access to that data?
Greely: Unless you are a world-class genetics researcher focused on correlations between SNPs and genetic disease, then no. You won’t be able to glean anything from the raw data. I would save my money and wait to sequence my full exome and genome and then pay for a full assessment with a clinical geneticist and genetic counselor.
VentureBeat: Do you predict that we’ll be able to cheaply sequence our exome, and potentially even genome, next year?
Greely: In 2014, it wouldn’t surprise me if you could get your whole exome sequenced for under $1,000. We just need one of these “next-generation sequencing technology” startups to work. I don’t have a strong sense yet of whether it will be a huge tradeoff between whole exome and whole genome. I might consider waiting a bit longer to get my whole gene sequenced for the same price.
VentureBeat: Any other predictions for 2014 for genetics?
Greely: It may be easier to match relatives in a database if the FBI moves ahead with its plan to increase the number of “CODIS markers.” You may soon definitively be matched in a DNA database to a sibling or cousin you never knew existed. We may see more criminal justice subpoenas or search warrants levied against SNP chip companies. A DNA sample could be matched to a crime scene.
We will also see fights in the courts about the language of genetic privacy laws and statutes. My strongest prediction is that genetics will get even more fascinating and unpredictable.
- FDA orders 23andMe to ‘discontinue marketing’ its genetic testing service
- 23andMe remains optimistic despite FDA issues: ‘We are not going anywhere’ (exclusive)
- Can you trust Facebook with your genetic code?
- 23andMe surrenders to FDA: Rebrands testing kits, stops offering genetic analysis
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.