Originally published Feb. 19.
When a cohort of women and men who chose to purchase gene scans from consumer genetics firm 23andMe learned that they were positive for certain BRCA mutations that predispose them and potentially their family members to a heightened risk of breast and ovarian cancer, none described feeling overly anxious and a few even noted feeling "neutral," according to a recent study.
While not generalizable to the population at large, the findings suggest that early adopters of direct-to-consumer genetic testing, when they find out they are at risk for a serious disease, they seek medical advice and take actions with the guidance of healthcare providers.
Researchers from 23andMe and Stanford University published data in PeerJ this week from a small behavioral study that gauged the reactions of 32 customers after they learned through 23andMe's Personal Genome Service that they were positive for any of three reported BRCA1 and BRCA2 mutations associated with a heightened risk of hereditary breast and ovarian cancer. The study also gauged the responses of 31 customers who were non-carriers of the BRCA mutations.
The researchers, led by Uta Francke, 23andMe senior medical director, reported that 11 women and 14 men learned for the first time that they carried a BRCA1 185delAG or 5382insC mutation, or a BRCA2 6174delT mutation, which are particularly prevalent among those of Ashkenazi Jewish background. Meanwhile, five women and two men already knew their positive mutation status from previous testing.
After getting tested through 23andMe, 10 mutation-positive participants expressed "surprise" at learning their results, but none reported extreme emotional distress (i.e. crying, loss of sleep, suicidal ideation); four participants expressed being "moderately upset" in that they couldn't stop thinking about their results or felt some anxiety; and nine participants said they felt "somewhat upset," where they were initially disappointed but then the feeling dissipated. There were five women and six men who said they felt "neutral" in learning their positive-mutation status.
Among the non-carriers, learning their BRCA status didn't make them believe that they were free of cancer risk. All but one mutation-negative participant thought that her cancer risk was reduced, but most participants in this cohort seemed to understand that there are other genetic and environmental factors that could cause breast and ovarian cancer.
Most mutation carriers discussed their results with a medical professional, according to the paper. Among the 11 women who learned they were mutation positive for the first time through 23andMe, one had a prophylactic mastectomy, three planned to have mastectomies, three had oophorectomies, and four planned to have oophorectomies after they had children. Five women said they wanted to get breast exams or breast imaging tests, and seven said they would continue to monitor their health through regular breast exams. Mutation-positive individuals also discussed their results with at least 30 relatives, and this led to the identification of 13 additional mutation carriers.
Based on this data, the study authors concluded that providing genetic testing results directly to consumers, without the involvement of a physician, doesn't cause undue anxiety or spur irrational behavior, even if the markers are significantly linked with a serious illness. "Direct access to BRCA mutation tests … provided clear benefits to participants," wrote Francke and colleagues. "Given the absence of evidence for serious emotional distress or inappropriate actions in this subset of mutation-positive customers who agreed to be interviewed for this study," the study authors recommended broader testing for Ashkenazi Jewish women.
The study authors acknowledge some limitations of the study — namely the fact that 23andMe customers are not "entirely representative" of the overall population, and may represent a more "proactive more scientifically educated group" than the general public.
"The statistics about how anxious they were and how they followed up are very interesting, but it is unclear how they would transfer to a larger population of customers," Russ Altman, professor of bioengineering, genetics, and medicine at Stanford, told PGx Reporter over email. "Obviously, it is in the interest of 23andMe to show that DTC genetics provides welcome and useful information, even when it is potentially pretty bad news." Altman, while not involved with this latest study, has been a scientific advisor to 23andMe and still holds some stock in the company.
Since several mutation-positive 23andMe customers told family members to get tested, Altman noted that some customers and their relatives seemed to have "clearly benefited" from this increased knowledge. "However, I am not quite comfortable with the … statement … that 'broader screening ... should be considered,'" Altman said. "This was not really screening as customers elected to come to 23andMe for a complex (and sometimes unclear) set of reasons, which is very different from setting up a bus in front of the library and taking all comers."
It is also worth noting that Francke and colleagues only asked how participants felt after learning their results from 23andMe. They didn't track how their feelings might have changed after speaking to a healthcare professional and considering the test results in the context of their health and the potential implications for their families.
Still, the study authors chose to highlight the accounts of two women who said their anxiety levels increased after discussing their results with a medical professional. One woman felt pressured into surgical procedures by her doctor and another felt anxious upon hearing the various choices presented to her by a genetic counselor.
“It was a real shakeup for me for a little while," said one woman who was interviewed. "Not because of my reaction to the BRCA2 results from 23andMe, but to the total fear factor that was put in by all of the traditional medical people." She said that her doctor recommended that she get a mastectomy and oophorectomy within six weeks.
Another woman told the researchers that while she wasn't anxious after receiving her 23andMe results for BRCA mutations, she felt "slight anxiety" after hearing all her choices from a genetic counselor. "Whenever you have choices you have anxiety, because it's time to research and make right decisions and so on," the woman said.
23andMe told PGx Reporter that the women related these comments to researchers "spontaneously" and that that the investigators didn't specifically probe participants about how they felt after speaking to their doctors.
Tracking Emotions
Calling the PeerJ paper "remarkable," Robert Green, associate director for research at Partners HealthCare Center for Personalized Genetic Medicine, told PGx Reporter that the study shows that when people interested in learning more about their genes receive test results that show they have a "dangerous mutation," they do "pretty rational things" with that knowledge.
"They went to doctors, they went to genetic counselors, they explored their options. They may or may not have gotten surgery, but they thought about it," Green said. "It was remarkable for what it didn't show. It didn't show that people were outraged by receiving the information. It didn't show people reporting devastating or catastrophic emotional reactions to being surprised by the information. It showed people dealing with it."
Regulatory agencies, concerned about the potential impact of DTC genetic testing on the public's health, have reined in efforts by companies to sell such tests directly to consumers, without the need for a doctor's prescription. In 2010, after Pathway Genomics announced plans to market spit kits for its then-DTC testing service through Walgreens and CVS Caremark's pharmacies, the US Food and Drug Administration sent a letter to the company stating that the agency considers its saliva collection kit a medical device requiring approval before it hits the market. Subsequently, the FDA sent similar letters to 23andMe, Decode Genetics, Navigenics, Knome, and Illumina, and the US Congress held a hearing about the burgeoning DTC genomics industry (PGx Reporter 7/28/2010).
During this period, some industry observers expressed concern that marketing genetic tests to consumers may cause them excessive emotional distress or move them to make rash, harmful decisions. Phil Gingrey, a Republican congressman from Georgia and a former obstetrician/gynecologist, expressed concern at the Congressional hearing about how people would react if they learned their risks for serious diseases without the involvement of a doctor. "You've got a lot of hypochondriacs out there," he said at the hearing, "and we're going to make maniacs out of them."
Since then, several studies have shown that people who get tested through DTC genomics firms aren't necessarily more motivated to act on the results. For example, a few years ago researchers from the Scripps Translational Science Institute surveyed the reactions of more than 2,000 people who had learned their genetic predisposition for 23 diseases and conditions through Navigenics' Health Compass. The investigation revealed that the majority of study participants reported no significant changes in levels of anxiety, dietary fat intake, or exercise after receiving genetic test results (PGx Reporter 1/12/2011).
However, the Scripps study, as well as the Multiplex Initiative by the National Human Genome Research Institute, also showed that although a minority of people shared their reports from DTC testing with genetic counselors or their doctors, those who shared their test results with their healthcare providers appeared to be the most motivated to make changes to their lifestyle (PGx Reporter 7/29/2009, 11/12/2011).
Since BRCA mutations confer a 60 percent risk of breast cancer and a 40 percent risk of ovarian cancer on women who harbor them, mutation-positive women who want to reduce their risk with surgery can only do so after talking to their physicians. As such, a survey that only looks at how people feel after learning their test results, and doesn't factor in how they felt after speaking with a medical professional and being presented with a range of options, may not capture the full spectrum of emotions participants may experience.
"It is the total response of the customer to the DTC genetic news that needs to be evaluated," said Altman, who founded Personalis, a startup focused on delivering medical interpretation of human genomes for research. Personalis eventually plans to offer genome interpretation services for use in the clinic, but the company will work with healthcare providers in "a more conventional way" and not sell directly to consumers.
"There are those who focus on the ethics, policy, and psychology aspects of genetic testing in the community, and they are being understandably cautious" on the issue of selling tests DTC, Altman reflected. "I think there is no doubt that it may be more prudent to deliver potentially bad news in the setting of a [healthcare] provider who can provide support to a patient."
23andMe customers have to "unlock" their BRCA test results. Before viewing the results, the company recommends customers read about the association between the reported mutations and breast and ovarian cancer. According to Francke and colleagues, mutation-positive women in the study understood after learning their test results that their risk for breast and ovarian cancer was elevated and those who already knew this due to family history and ancestry said that knowing their BRCA status "made them more aware of the reality and prompted them to take action."
According to 23andMe, the added stress experienced by the two women highlighted in the paper occurred due to pressure and confusion from medical professionals. "The increased anxiety was not from finding out the serious implications (one noted that she was aware that surgery might be necessary), but from either (1) confusion engendered by all the options presented by the genetic counselor or (2) pressure from a doctor to get surgery," Catherine Afarian, a 23andMe spokesperson, said over email.
Regardless of whether study participants fully understood the implications of their test results by reading educational materials from 23andMe, the researchers didn't track their emotions after speaking with healthcare professionals. As such, the study doesn't offer any conclusive information about whether those interactions would have caused an increase in anxiety.
Green wouldn't comment on whether the emotional impact of participants' discussions with healthcare professionals should have been factored into the study design. As a general observation, he noted that "since the information was offered on those two people … it's hard to disentangle any anxiety they might have felt from the test to any anxiety they might have felt after they spoke" to healthcare professionals. "Maybe they are offering up a comment on the anxiety that accrued throughout the whole experience, starting with when they looked at the result and continuing through their discussions," he said. "They might look at it, not think about it, and put it away. But when they have to talk about it, that becomes a very anxious moment."
Although no one in the study had any significant emotional distress from finding out their BRCA mutation status through 23andMe, the fact that all female mutation carriers sought the advice of a doctor may be sufficient evidence for some critics to assert that these are medical tests that should be administered by a doctor. "You could argue that it is very alarming that a company would recommend an expensive and potentially life-altering diagnostic test to its customers, without ever meeting them (or even knowing for sure that the person on the other end of the email has been correctly matched to the DNA sample)," Altman said.
On the flip side, 23andMe and proponents of DTC genetic testing would argue that consumers with direct access to their genomic predisposition for diseases understand the risks and seek advice from their doctors before acting, so there is no harm. "It is pretty compelling that their actions lead to some people getting confirmatory testing, and screening of 30 relatives, 13 of whom were carriers," Altman noted. "If this information saves any lives, then you have to take notice of this as one important mechanism to help customers seek appropriate health information."
Study Weaknesses
In the PeerJ paper, Francke and colleagues compared their findings to the REVEAL trial, which was published in the New England Journal of Medicine in 2009 with Green as the lead author. That study found that children of Alzheimer's patients who learned whether or not they harbored APOE mutations – associated with heightened risk for the disease – didn't experience significantly more anxiety than those who didn’t know their APOE status.
"In contrast to our study of DTC customers who received unexpected results, the REVEAL study compared disclosure and nondisclosure groups of participants (that both were pre-screened and underwent extensive pre-test counseling and post-test psychological follow up) and failed to find significant differences in test-related distress," Francke and colleagues note.
The REVEAL study, unlike the survey by Francke et al., was a randomized trial. Furthermore, while it may not have found a difference in anxiety between groups that learned their APOE status versus those who didn't, when researchers looked at emotional distress according to study participants' APOE4 mutation status, "APOE4-negative people had a significantly lower level of test-related distress than did the APOE4-positive subgroup."
Additionally, REVEAL looked at APOE variants associated with Alzheimer's, which occur in up to 30 percent of the population and in approximately 40 percent of Alzheimer's patients. Meanwhile, BRCA mutations are highly penetrant in terms of their association to breast and ovarian cancer, but are relatively rare, occurring in five percent of breast cancer patients and up to 15 percent of ovarian cancer patients.
Green said he was "flattered" by Francke and colleagues' comparison to the REVEAL study. The two studies are comparable, according to Green, in that they are both the first of their kind in exploring the psychological impact of genetic testing. Francke and colleagues "are citing us as a type of study that's looked at the impact of disclosing information with a particular methodology and did not find serious adverse events, at a time in this field when people were quite concerned about finding very detrimental effects if you returned susceptibility results," Green said.
Altman agreed that although REVEAL and the study by Francke et al. are different in design and in terms of the tested markers, the researchers in PeerJ "are making a reasonable point that studies are suggesting that customers may not be suffering grave harms with genetic knowledge." However, he emphasized that because this study was done only on 23andMe customers, the results may not generalizable to the overall population, where there are people with less resources and education.
The fact that the PeerJ study involved early adopters of DTC genetic testing doesn't bother Green so much. "At this moment in history, [this study] is probably fairly reflective of people information seeking about their own genetics," he said.
However, only a minority of 23andMe customers who were BRCA mutation positive agreed to participate in the trial. 23andMe invited 136 23andMe customers who were BRCA1 or BRCA2 mutation-positive and 160 who were mutation-negative to participate in the study. However, only 32 mutation carriers and 31 controls participated. Francke and colleagues acknowledged in the paper that "responses reported by the 16 men and 16 women who are BRCA carriers and agreed to be interviewed for this study may differ from those experienced by the 61 male and 43 female mutation-positive individuals who did not reply to our invitation."
This is the "prominent" weakness of the study for Green. "We don't really know what happened to the 104," he said. "Unfortunately, that's the price of conducting a naturalistic experiment, where people have purchased a service, they've received and looked at this information, and they have a choice as to whether or not they participate in follow up."
It's also likely that BRCA mutation-positive 23andMe customers who had a particularly emotional response may not have wanted to participate in the study. "We cannot exclude that somebody who had a traumatic response may not have wished to revisit that experience in an interview," the study authors wrote, pointing out however, that the "initial study participation rate was higher within the mutation-positive group than within the mutation-negative group."
Finally, over time the study participants may not be able to recall exactly their initial reactions to their BRCA test results. "The time interval between viewing their BRCA report and the interview varied between two years and three months," the study authors said, acknowledging that "the accuracy of recollections may have differed among participants."
Confirmatory Testing
Myriad Genetics is the sole provider of commercial BRCA mutation testing for hereditary breast and ovarian cancer, and holds patents covering numerous mutations associated with these diseases. Insurers have criteria in place about when BRCA testing for hereditary cancer risk is "medically necessary." Testing conducted by consumer genetics firms is not covered by insurance.
When 23andMe began offering testing for three BRCA mutations, many industry observers wondered why Myriad wasn't taking action to shut down the practice. Historically, Myriad has been aggressive in going after commercial and academic labs conducting BRCA testing for breast and ovarian cancer patients.
However, it seems that many women who find out they harbor a BRCA mutation for the first time through 23andMe get retested by Myriad's diagnostic before taking medical action. In such cases, 23andMe is driving more customers to Myriad for testing on its BRACAnalysis test, priced at more than $3,000.
23andMe's Afarian said that the researchers don't have complete information on whether confirmatory testing was performed on all mutation-positive women who had or will have surgery. "Some, but not all, of the mutation-positive women had already had the results confirmed by an independent lab," Afarian said, adding that the five women who spoke with a genetic counselor did so upon referral from their doctor, and then had repeat BRCA testing in a clinical lab "in most but not all cases."
"However, not all of the women had actually had surgery at the time of the interview," Afarian noted. "Some of those individuals may have had a confirmatory test after the interview and before surgery."
Due to Myriad's patent position, the company essentially holds a monopoly on BRCA testing for hereditary breast and ovarian cancer. As such, patients and healthcare providers have argued that women who want to confirm the Myriad test results can't go to another lab to get a second opinion test.
"There are many other possible BRCA mutations on which 23andMe does not report," Afarian said. "As a result, 23andMe could only serve to confirm mutation-positive status for these three specific BRCA mutations.
"Due to the increased incidence of these specific mutations in individuals of Ashkenasi Jewish decent, we do believe 23andMe can serve as an affordable, non-invasive screening tool for individuals who are Ashkenazi Jewish," she added.
The US Preventative Services Task Force in 2005 recommended against routine referral for genetic counseling and BRCA testing unless women have a family history BRCA1 and BRCA2 gene mutations. The task force is in the process of updating its recommendations in this regard, and has been gathering data on how accurate physicians' risk assessment methods are for selecting which patients should receive BRCA mutation testing; and what the benefits and adverse effects are of genetic counseling patients ahead of testing.
"New models are emerging for the delivery of genomic services both in the context of healthcare providers and directly to the consumers," said Muin Khoury, director of CDC's Office of Public Health Genomics. "We need to collect evidence to actually look at the benefits and the harms" of providing genetic testing broadly in the population. According to Khoury, the USPSTF will likely release a new report on BRCA testing for gauging familial breast and ovarian risk later this year or in 2014.