NEW YORK – Ambry Genetics this week revamped its hereditary cancer genetic test menus, adding some genes to its panels while removing others.
These changes are fueled by the lab's focus on gene-disease validity and go against years of a "more-is-better" approach in the industry to creating panel tests. For customers who need to gauge additional genes that are not on the latest edited panels, the lab is still offering broader testing panels as add-on options.
When gene panels were introduced more than a decade ago, labs tended to add new genes to testing panels quickly after a tie to disease was uncovered. "When panels were designed originally, we were just testing everything because we didn't know [what to test]," said Jessie Grzybowski, associate director and medical science liaison at Ambry Genetics. "We had only traditionally been testing cancer patients for two genes," BRCA1 and BRCA2.
With improvements in sequencing capabilities, the cost of adding genes to panels was also not a steep barrier. "There has been a trend toward testing more and more genes because technology has made it easier and cheaper to do so," noted Eric Lee, director of consulting firm Preciselee Genetics, which helps develop and validate genetic tests, accredit testing labs, and interpret and report genetic testing data, in an email.
But, he added, "sometimes, this expansion has outpaced our understanding of how to use the results in the clinic."
The evidence linking certain genes to cancer risk can shift over time as more studies are conducted. For instance, the National Comprehensive Cancer Network (NCCN) guidelines on assessing high-risk familial or genetic risk previously said that the NBN gene was associated with an increased risk of breast cancer. However, the guidelines have since been updated and now note that while NBN is found on some hereditary cancer testing panels, there is "insufficient evidence of an association with breast and/or ovarian cancer."
Similarly, recent studies have shown more tenuous links between mutations in CHEK2 and colorectal cancer risk, and the NCCN last month removed its recommendation that patients with CHEK2 mutations undergo enhanced colorectal cancer screening.
"It's not that we got this wrong," said Jada Roberts, a genetic counselor at the University of Oklahoma's Stephenson Cancer Center, of field's prior understanding of this gene-disease link. "It's just that as more evidence came out, we started to see a weaker and weaker correlation. And, now, we can't really say that the relationship is causative."
Aliso Viejo, California-based Ambry is updating its entire hereditary cancer menu based on data that have emerged in recent years that support, or don't support, links between certain genes and hereditary cancer risk. Tempus AI announced last week that it has entered into an agreement to acquire Ambry for $600 million from Konica Minolta, which had itself acquired the firm in 2017. Meagan Farmer, senior marketing manager at Ambry, said the acquisition should not affect the rollout of the new testing menu.
According to Grzybowski, there is a group at Ambry, dubbed the gene team, that is dedicated to reviewing evidence on disease-gene validity. "We've spent a lot of time looking at the genes that have been on our panel [and] evaluating, 'Does it still make sense to keep them there? Has their gene-disease validity changed?'" she said. "Maybe we thought they were moderate-[evidence genes], but now, we know they're limited-evidence [genes], and maybe they're not adding any benefit to the patients who are getting this test."
Some of the new Ambry gene panels are guidelines based, meaning that each gene on the panel has associated NCCN management guidelines. The firm, for instance, recently edited its CancerNext panel, which now gauges 39 genes, by adding eight genes, mostly ones associated with adult renal cancer predisposition, and removing three.
"Focusing on genes with established clinical management guidelines is a positive shift and ensures that findings are clinically actionable," Lee said. "While some relevant findings might be missed by testing fewer genes, it's important to balance the benefits of broader testing with the potential risks of acting on information that may not be reliable or useful."
Panels that stick closely to guidelines may appeal to busy healthcare providers, particularly primary care doctors or doctors without a genetics background who want test results they can act on clinically and that inform patient care. "I can understand why the smaller panels would be ideal in those situations, especially for the providers that just aren't used to genetics," said Kaitlin Stokes, a cancer genetic counselor at Houston Methodist West in Texas. "Primary care providers know a lot, but genetics is not their specialty."
Roberts added that, similarly, "patients don't tend to love having just kind of gee-whiz information where there's not really anything to do with it."
The practice of focusing genetic tests on genes with stronger underlying data is not that widespread in the testing industry, in Lee's view. "Offering 'more' genes or features is an easy way to make a test seem better," he said. "It's far simpler to appeal to this mindset than to argue that focusing on fewer, well-understood genes is better."
Ambry's Grzybowski said that other labs have been quicker to add genes to their panels than Ambry has, and she is not aware of other players in the space editing test panels with the same level of focus on gene-disease validity as Ambry.
Farmer, who directed the University of Alabama at Birmingham's cancer genetic counseling program for a decade before working for digital health company My Gene Counsel and then joining Ambry Genetics in 2022, added that she hopes Ambry is leading the way on this change in approach to developing panels. "Just as a genetic counselor, I do hope that other labs will go in this direction [of a greater focus on gene-disease validity]," said Farmer. "Even if they don't have a peer-reviewed, published framework, I hope that they are being more careful."
Lee noted that even those labs that do choose that approach typically do not abandon all the extra genes or features. "Instead, they make it opt-in, which may be seen as less commercially risky and makes broader testing a conscious choice on the part of patients and doctors," Lee said.
In that vein, Ambry offers a CancerNext-Expanded panel that assesses up to 90 genes, all the cancer predisposition genes the firm tests for, not just those with clinical management guidelines.
It also has separated out genes with limited or preliminary evidence to other, opt-in panels. These genes have some data tying them to hereditary cancer risk, but the link has yet to be fully established. For instance, its ColoNext Panel features a base panel of 20 genes — MBD4 has been added to the base panel, while CHEK2 has been removed, in line with the new recommendations — but customers can add on a panel of six genes, including not only CHEK2 but also ATM, CTNNA1, MLH3, RNF43, and RPS20.
Both Farmer and Grzybowski cautioned that results from the limited evidence panels can only be negative or a variant of uncertain significance (VUS), as there is too much uncertainty to say for certain that a variant in one of these genes is pathogenic. "Limited evidence means we don't have enough evidence to even know if the gene is doing what we think it's doing," Grzybowski said.
These opt-in panels may then appeal to genetic counselors or other genetics providers who are versed in what ambiguous test results may mean as well as to patients that want a comprehensive analysis and can handle uncertainty.
Roberts said her approach to choosing a panel comes down to why the patient was referred to her. Typically, if someone came to her already with a cancer diagnosis, she would take a broader approach and select even the limited evidence genes for testing, as the results could help point to an explanation for that patient's diagnosis. However, if someone came to her for testing based on their family history, she'd likely suggest a narrower approach. "For somebody who is coming in to learn about their risks to then take action upon those risks, I probably wouldn't do genes that are preliminary evidence. They are not going to give us a lot of useful information," Roberts said.
She noted, though, that she has had a patient who wanted to know everything and has come back for testing time and again when new panels are released
Stokes, meanwhile, said she prefers ordering the largest panel possible, especially as insurance companies may only pay once for testing. Even if there is a VUS or other finding in a gene with no associated clinical management guidelines, the patient may keep that result in mind for the future, as the evidence may change.
Ambry's approach, Farmer said, aims to provide clinicians the flexibility to order its set panels with or without add-on tests based on their needs and provide payors more transparency into what's being ordered. "This may result in fewer medical necessity denials than for custom panels where [payors] do not know what gene content is being analyzed," she said.