By Paul Nicolaus
October 5, 2017 | One day, nearly a decade ago, Nita Farahany and her (now) husband spit into a tube on one of their first dates. After receiving the results, the couple signed into 23andMe and compared profiles. This, she admits, was their idea of fun. “We’re geeks,” she said.
As the pair perused the information gleaned from the tests, though, they came across a detail that wasn’t so fun to see—an increased risk of Alzheimer’s. Farahany’s husband has since taken this insight to heart, reading dozens of articles and transforming his diet by upping his intake of saturated fats.
“He also has an actual history of heart disease in his family,” she said, “so he probably will never get Alzheimer’s disease if you understand where I’m going with that.” Farahany disagrees with the diet decision; her father, a cardiologist, is horrified by it. But her husband believes that he is his own best advocate and the person most invested in the process of understanding his own genome.
A professor of Law & Philosophy at Duke University, Farahany shared this anecdote during a panel discussion at the 2017 Aspen Ideas Festival because it exemplifies one of the overarching concerns about direct-to-consumer (DTC) genetic testing—that the general public’s access to genetic information could result in misunderstandings or harmful health-related decisions.
Unlike traditional clinical testing that relies on gene sequencing or targeted testing of high penetrance Mendelian mutations, many DTC tests estimate risk by genotyping an array of common, low penetrance, single-nucleotide polymorphisms.
If the degree to which the general population has a disease is 1.2 and an individual’s risk is 2.4, then that person is twice as likely to manifest the disease, for example, but that genetic contribution is still just one tiny part of the overall story.
Even if the level of risk is greatly elevated there could still be an infinitesimally small risk of developing a condition. The problem, Farahany said, is that most do not understand that.
Access Versus Protection
These types of concerns came to a head in November 2013 when the U.S. Food & Drug Administration (FDA) directed a warning letter to 23andMe. Although alarm bells had previously been raised, there was essentially no regulation of these types of DTC genetic tests prior to this moment in time.
The FDA action resulted in the removal of 23andMe’s health-related testing products sold to U.S. customers and made an even larger splash as other companies called it quits and went off the market rather than engage with the agency.
But even after seeing the potential drawbacks of DTC genetic testing up close and personal, Farahany maintains a belief in the freedom to obtain this type of personalized health information. “I actually come down on the side of thinking everybody should have access to their information—even my husband, and even though I cringe when I see him drinking half & half,” she said.
Others don’t see it that way. “Nita and I are good friends and agree on many things,” said fellow panelist Hank Greely, a Stanford University law professor and director of the school’s Center for Law and the Biosciences. “23andMe is probably the deepest schism in our relationship.”
From his point of view, the intervention served as an appropriate break in the action. Of the 254 conditions the company reported on, there was a paper behind each of them, but one paper on genetic associations doesn’t mean much. “So I applaud what the FDA did,” he said.
There is a difference between ensuring the analytic validity of something and deciding whether the general public has a right to the interpretation of information, Farahany argued. If the FDA decides what information people have access to, that tramples on First Amendment rights, protects the old guard, paternalist system of medicine, and ignores the patient revolution underway.
“There are roughly 330 million people in the United States,” Greely noted. “There is one Nita Farahany.” The number of people who have the same depth of genetics knowledge might number in the thousands. If the FDA is regulating to protect the country, it needs to think about the 330 million or else people will get hurt.
Some experts have expressed fears that users could over-engage with results meant to inform—not diagnose. There’s also the other end of the spectrum to consider. Those who test negative for BRCA1 or BRCA2, for example, might decide to forego regular mammograms or breast self-examination. “And that,” he said, “could be a fatal mistake.”
PGen Study Takes Fresh Approach
Roughly a decade after their arrival, DTC genetic tests clearly remain controversial, and many of the same concerns tend to be raised over and again.
Part of the reason is that these tests are difficult to understand from a scientific perspective because, by definition, they are self-motivated and self-directed. As soon as there is an attempt to take a look under the hood in an experimental setting that essence is lost.
Prior studies have offered testing to a research institute’s employees or gone into a community and offered testing to individuals. The problem is that these individuals are different from those who are sitting at home on the Internet and decide one day that they’d like to spit in a tube and see what happens.
Or so says Alexis Carere, a postdoctoral fellow and genetic counselor at McMaster University in Ontario, Canada. Back in 2012, while still a doctoral student, she joined up with Robert Green, who directs the G2P Research Program at Brigham and Women’s Hospital and Harvard Medical School and led one of the first prospective studies of DTC genetic testing services.
This NIH-funded “Impact of Personal Genomics (PGen) Study” has measured the perceptions and tracked the behaviors of individuals who purchased personal genomic tests prior to the FDA restrictions imposed on DTC testing companies 23andMe and, later, Pathway Genomics in 2013.
The research took a different approach by working with these two companies and asking their customers to participate in the study after they ordered testing on their own accord. The method allowed the researchers to get at the actual consumer base and review participants’ test results—two key advantages over many prior studies.
Since graduating from Harvard, Carere has maintained ties to the study and continues to publish scientific reports that shed light on everything from how consumers interpret their results to how those results impact actions taken.
Findings Add Nuance
One of these PGen papers published in September 2016 (doi:10.1038/gim.2016.141) took a look at prescription medication changes. It came about in direct response to the 2013 FDA letter, which posed the possibility that customers could use their results to self-manage drug treatments through dose changes or by ditching therapies altogether.
One of the key takeaways from the study, according to Carere, was that virtually every participant received a positive pharmacogenetic result. Over 9 in 10 were informed that they metabolize a certain drug poorly, for example, or are more prone to side effects.
This means that in the general population almost everyone is receiving some sort of a result that at least in theory could be used to alter or guide medication usage. Even so, only a small fraction reported making unsupervised changes six months after testing.
Out of 961 participants, 54 reported changing a medication, and the vast majority of those (45 out of 54) reported that they consulted with their physician before making the adjustment. Ultimately, less than 1% made the type of change the FDA was concerned about.
“We felt that that was pretty important information to have because the FDA as well as other critics of DTC toss around these concerns, these possible negative outcomes that could occur, but often without any data to support that they actually do happen,” Carere told Diagnostics World News.
On the other hand, there were instances where individuals or their physicians misunderstood test results. When one patient learned of an increased risk of adverse effects when taking a statin, she decided to cut her cholesterol drug dose in half without consulting her physician.
It turned out, though, that the drug she took was not a statin at all. To top it all off, she had reported a personal and family history of heart disease, high cholesterol, diabetes, and obesity. This was probably a very appropriate treatment for her to have, Carere said, and she took an inappropriate action on the basis of her genetic test results.
Another of the PGen studies, from March 2015 (doi:10.1038/gim.2015.34), dug deeper in this direction of understanding (or lack thereof), using confidence as the key tool.
“Time and time again studies have shown that genetic literacy among people taking these tests both before and after is quite high,” Carere said. These tend to be highly educated people who know enough about genetics that they want to order a test. “When we looked at our population, same thing. They basically maxed out our quiz,” she said.
Beyond this objective understanding of genetics, however, the researchers used a separate scale to examine self-efficacy. What they found was that people felt less confident in their genetics knowledge after testing, she said, which is an interesting phenomenon considering companies often pitch their DTC tests as an empowering educational experience.
One explanation is that participants were over-confident at the start. In that case, being taken down a peg or two after receiving complex genetic information may have been an appropriate response. Another explanation, though, is that they were poorly served by the results received.
Consider, for a moment, that 23andMe customers previously received hundreds of pages that included a page for every condition examined in addition to carrier status, drug response, nonmedical traits, and ancestry information. It is very possible that some found the report to be confusing, overwhelming, and ultimately unhelpful.
The research suggested that there was room for improvement, Carere said, and that is exactly what happened. When the FDA shut 23andMe down, they worked with the company to create new reports, requiring test pilots and educational assessments of a general public population to ensure the information shared could be better understood by recipients.
Genetic Counseling as Potential Solution
Despite any improvement in the communication of test results over the years, some have continued to call for a greater use of genetic counselors who are trained to help patients understand their results, and testing companies have become more willing to enlist this form of expertise.
In fact, Harvard’s Green is among those who have begun to harness their unique skill set in the form of a new start-up company. Along with Lisa Alderson and Randy Scott, who are former executives from testing company Invitae, he co-founded Genome Medical in 2016.
The company’s vision is to help shepherd in an era of genetic and genomic medicine that can lead to improved outcomes, improved care, and improved life while reducing the cost of healthcare delivery, said Alderson, who serves as CEO.
One of the biggest barriers of bringing genomic medicine into everyday healthcare is the small number of experts in the field. In the U.S., there are only about 2,000 geneticists and fewer than 5,000 genetic counselors, she pointed out, which isn’t enough to meet the needs of 330 million people.
To fill that gap, Genome Medical has emerged as a nationwide medical practice that includes genetic counselors and geneticists as well as primary care doctors and care coordinators. Together, this care team helps patients and providers as they access health related genetic discoveries using a telehealth platform.
The process typically begins with a counseling session that reviews personal and family health history and gathers an understanding of an individual’s interests and areas of concern. There is also consideration of whether testing is medically indicated and whether it would be covered by insurance.
When testing is pursued, Genome Medical helps obtain the appropriate test from the right lab while screening for the highest quality products available in the market. The company also facilitates the logistics involved, such as sample collection.
“With the test results in hand, we do clinical interpretation of those results in the context of the personal and family health history of that patient,” Alderson told Diagnostics World. “And then we return those results to the patient and we help counsel them on what that means—both for them and their family members.” Ultimately, this is turned into recommended clinical action.
While conventional, physician-mediated interactions with patients are expected to be the company’s real bread and butter, Genome Medical is also making its services available to consumers looking for advice before or after obtaining DTC genetic testing.
In line with that offering, the company recently announced a partnership with Helix, which spun out of genomics giant Illumina and counts both Green and Farahany among its scientific advisory board members.
Genome Medical’s “Genetic Navigator” service is designed to guide consumers through Helix’s marketplace for DNA-powered products that extend across a range of categories, some of which fall in the realm of entertainment, lifestyle, and recreation (think genetically matched wines and workouts) and others that fall within the realm of health.
“At the highest level, I think our vision is aligned,” Alderson said, “which is that individuals have a right to their genetic information.” That information is informative for the individual and, when in a health or medical setting, should be delivered with the benefit of access to medical experts who can interpret what it means.
Role of Counselors Debated
But should genetic counselors be devoting their expertise to the DTC realm at all? Traditionally, they have been found in hospital and clinic-based settings and have focused on medical tests ordered by physicians for specific indications.
While some say that DTC testing companies ought to have them on staff to better answer questions and explain test results to customers, others note the shortage of these professionals, the lower priority of this form of testing, and the potential conflict of interest that can arise when counselors are employed by testing companies.
Most would agree that having the ability to sit down with a genetic counselor to review results would almost certainly lead to a better understanding and a more appropriate response while helping dispel a lot of the concerns that people have, Carere said, but that’s not really the point of DTC testing or the role of genetic counselors right now.
“It’s not medical testing, it’s often not medically indicated, it’s often not medically actionable, and counselors have typically been a little bit resistant to these kinds of referrals because they are not at a loss for work,” she said.
Even so, companies like Genome Medical appear to view the ever-evolving realm of DTC testing as an opportunity that shouldn’t be shunned altogether. After all, in some cases, one quick question or basic screening initiated through a channel like Helix or 23andMe could kick off a much longer genetic journey.
To make sure its counselors aren’t beholden to testing companies in a way that may not be in the best interest of consumers, Alderson pointed out that the company performs a conflict of interest assessment with its hires and does not allow its part-time staff to be employed by a lab.
Evolving Our Understanding
As of April 2017, 23andMe received the FDA’s okay to market its Personal Genome Service Genetic Health Risk (GHR) tests that cover Parkinson’s, late onset Alzheimer’s, and eight other conditions. It’s the latest development in what has become an intriguing case study of sorts.
While this company was central to the PGen study and may be at the center of many DTC genetic testing conversations, however, Carere cautioned that broad generalizations made about the topic of DTC testing tend to fall apart.
“It’s not this one kind of test or this one kind of experience or this one kind of company,” she said. A massive assortment of tests fall under the DTC umbrella, ranging from the poor quality and potentially dangerous to the benign to the useful, reliable, and medically actionable.
One reason DTC has come under plenty of criticism over the years is because the types of testing most commonly thought of in this space have, in general, been quite different from those found in a clinical setting. Some have been purely recreational and others have offered up risk estimates for complex diseases.
“That’s very different from anything that is being used clinically,” she said. “There’s basically no evidence to show that those estimates are useful, that we can do anything with them, or that we should be changing medical management on the basis of them.”
Beyond that, they’re not regulated in the same way. Prior to the 2013 FDA action, 23andMe wasn’t regulated by the FDA and did not have to meet the same standards as a clinical lab. It becomes easy to envision, then, a bunch of shoddy Internet-based testing companies offering up unreliable results, Carere explained. That’s why DTC has gotten a bit of a bad rap.
On the other hand, it is plausible to come across a DTC test that is every bit as good as a clinical test, which is especially true when it comes to carrier testing for genetically inherited conditions.
“DTC has evolved a lot over the last 10 years,” she added. During that span of time, new data has emerged to address the concerns that were raised a decade ago and continue to be raised. “We need to kind of evolve our own understanding of DTC genetics.”
Paul Nicolaus is a freelance writer specializing in health and medicine. Learn more at www.nicolauswriting.com.