WASHINGTON (GenomeWeb) – A symposium at the American Association for Cancer Research's annual meeting here on Monday evening took up the debate on the merits of population-wide genetic testing.
Session chair and Dana Farber Cancer Institute researcher Judy Garber began by quoting Mary-Claire King, who said in 2014 that every woman in the US 30 years or older should be tested for BRCA1 and BRCA2 mutations, regardless of her race or ethnicity. In fact, Garber said, several groups have done studies to analyze the costs and feasibility of offering BRCA screening on a population-wide level. Some researchers have even suggested going beyond BRCA testing to offer genetic tests for conditions such as Lynch syndrome or other high-penetrance conditions.
However, though the idea of population-wide screening is starting to become more popular, some researchers and clinicians are still hesitant. Some of the analyses have shown that such an undertaking could be more expensive than it's worth. Questions have arisen about whether researchers themselves understand enough about what certain mutations mean to human health and whether they could communicate that understanding to patients, Garber said. What would be done to lower the rate of false positives? Are there enough genetic counselors to aid patients in learning about genetic testing and understanding the results? Would there be a system in place to track down patients and update them on their results when new research allows for new understanding?
On the other hand, studies have shown that cancer patients who've had their DNA tested for mutations have found that their germline DNA sometimes contained mutations that didn't match their tumors, Garber added. In a study of Lynch syndrome, patients who had broader genetic testing done found that they had mutations in genes like BRCA1 and BRCA2 even though they had no family history indicating such mutations might exist. What do we lose out on if we don't look more widely, Garber asked.
Ephrat Levy-Lahad, a researcher from the Shaare Zedek Medical Center in Israel, shared her experience in testing Ashkenazi Jewish women for BRCA mutations.
Among Ashkenazi Jews, there are three common BRCA mutations, and about 2.5 percent of people in that population are carriers. Other BRCA mutations are rare in this population, Levy-Lahad added, so limiting screening to these three mutations increases testing accuracy.
Over time, the goal of identifying mutations has gone from testing in high-risk families to testing patients with specific cancers to prevent second malignancies and direct treatment. But testing a woman after cancer develops is a missed opportunity for prevention, Levy-Lahad said.
In order to prevent cancer in the first place, clinicians need to test unaffected women. Currently, deciding which healthy women to test for mutations is based on their family history of cancer. The problem with that, Levy-Lahad said, is that 50 percent of women don't have suggestive family histories, not necessarily because they're not at risk but because their families may be small or because the variants may come down through their father or other male relatives.
In order to determine whether and how to undertake a BRCA screening program, Levy-Lahad started by considering the principles for disease screening developed in 1968 by Glover Wilson and Gunner Jungner. These criteria include questions on whether a disease is considered an important health problem, whether there's a suitable test for the disease, and whether there are facilities for diagnosis and treatment.
Levy-Lahad chose to focus on the criterion of whether the existing test is acceptable to the population being tested. She considered questions of whether women should be recruited for testing by their doctors or at hospitals, or whether they could refer themselves for testing by learning about it through word of mouth, or advertising. The second challenge was how to return either positive or negative results in a way that would result in satisfactory experiences for the patients.
She and her team studied 1,771 participants. They found that recruiter-enrolled participants were older than self-referred participants, but that more women self-referred. A large proportion of people interested in participating had a family history of cancer. The screening found that about 1.8 percent of participants in the screening were mutation carriers, and after testing for several measures of psychosocial satisfaction and stress or anxiety markers, 88 percent to 95 percent of participants were highly satisfied with the experience.
The program showed that screening is feasible, Levy-Lahad concluded. Both open-access and medical enrollment models work well, there was high compliance with post-test genetic counseling, the information participants received was actionable, and there was minimal psychological harm.
However, she also saw that participants were older than is optimal, in their 40s and 50s, whereas a rigorous prevention screening program would need to test women in their 30s.
Genomics should be used for prevention of disease and not just to guide therapy, she added, suggesting that now may be the time to start implementing population-wide screening for certain genes or mutations.
McGill University researcher William Foulkes seemed to suggest that the success of BRCA testing in Ashkenazi women might not translate to population-wide genetic testing for other diseases, however.
Less than half of all people who need genetic testing actually get it, he noted, suggesting that there are indeed many people who could benefit from population-wide screening. And the costs have "dropped in incomprehensible ways," and are probably going to continue falling, he said, so the cost of doing testing is actually a very small part of the whole picture at this point.
But there are other considerations, he said. For such population-wide genetic testing programs to be successful, it would have to be pioneered in countries with specific criteria — a country with a high level of educated people, with an equitable healthcare systems, adequate financial resources, integrated health records, and other considerations. That would leave out countries like the US and Canada, Foulkes said.
And there would be numerous questions about these programs, he said. For example, what would the program test for? Does it make sense to only test for two genes? What about cardiac syndromes, or MAP, or Lynch syndrome? Would such a program require performing whole-genome sequencing on everyone, or would there be a set list of genes to screen for?
Foulkes also asked how the medical community is supposed to treat people who aren't actually patients — if people aren't ill, how do they get tested, and what do they do with the results? "Our ability to interrogate the genome exceeds our ability to understand it," he said. "Do we have sufficient knowledge to understand mutations in healthy people? Lifetime risk numbers aren't clear to people who want to make health decisions."
Precision is certainly an issue — new studies are constantly changing our understanding of each gene and how each variant confers risk. In some ways, BRCA1 and BRCA2 are outliers in that we understand them fairly well, Foulkes said.
There are also ethical considerations — if there's no clear plan for what to do with results other than simply returning them to patients with no explanation of what they mean for their health, "then that's research, not established clinical practice," he added.
Further, he said, it's important to consider whether population-wide screening will increase or decrease the health disparities and inequality that already exists in society.
One way to start introducing these kinds of tests into a larger population may be to begin with people who are already sick and let them spread the word to their families to get tested, Foulkes suggested. This wouldn't prevent cancer patients from getting cancer in the first place, but it would be a good way to immediately start testing affected people and their families, and gain more understanding of the genomics underlying various diseases.
While genomic testing can make sense for studying the germline DNA of cancer patients or for mutations that are already established as being fairly common in certain populations like Ashkenazi Jews, it remains controversial in other populations, Foulkes concluded. We need more information about genes themselves, and we need to know which genes we'll test for, what the results will mean, and how to do it in an equitable way.