Should All Breast Cancer Patients Undergo Genetic Testing?

Veronica Hackethal, MD

August 09, 2019

Imagine having to counsel a patient who has been diagnosed with breast cancer about the the possibility she may need to have her stomach removed.

That's what happened to Marc Tischkowitz, MD, PhD, an oncologist affiliated with Cambridge University Hospitals National Health Service (NHS) Trust in the United Kingdom. The patient with breast cancer was referred to him after the treating physician received the results from a multigene panel for breast cancer. The test showed that she carried a gene tied to increased risk for a certain type of stomach cancer, the treatment of which includes prophylactic removal of the stomach.

"We were in the situation in which this woman asked one question: is my breast cancer hereditary?" he said.

"We didn't answer that question. Instead, we gave her a completely different problem to worry about."

The gene in this patient was CDH1, which has been linked to up to 70% increased risk for diffuse stomach cancer, a rare but very aggressive type of cancer. Carrying a CDH1 mutation may also increase the risk for lobular breast cancer by up to 50%.

Multiple factors complicated this case. The patient's CDH1 mutation was a variant of uncertain significance (VUS). Evidence is inconclusive about whether this patient's genetic variant causes diffuse gastric cancer. Also, she had ductal breast cancer, which has not been linked to the CDH1 gene. And she had no family history of gastric cancer, so her risk for gastric cancer would probably be lower than that of someone with a CDH1 genetic mutation who also had a family history of gastric cancer. But it's unclear how much lower her risk would be.

All of this made it difficult to discuss the risks that she was facing.

For Tischkowitz, this patient's experience illustrates one reason to question the use of large, multigene panels in breast cancer.

Together with colleague Amy Taylor, PhD, they wrote a letter to the editor of the Journal of Clinical Oncology in response to a recent study that was published in the journal. That study has added fire to an already heated debate.

The study found that nearly half of breast cancer parients in whom there is a pathogenic or likely pathogenic variant are missed by current testing guidelines. The authors, led by Peter Beitsch, MD, from the Dallas Surgical Group–TME/Breast Care Network in Texas, recommended that all patients who have been diagnosed with breast cancer undergo expanded panel testing.

A flurry of letters to the editor in response to this conclusion show that there are many in this field who disagree and who think this is not a good idea.

At the heart of the controversy is a much larger, longer-running debate: who should make complex medical decisions, the patient or the medical establishment?

Landmark Decision

Let's backtrack to 2013, when the US Supreme Court made a landmark decision in a case concerning the tumor suppressor genes BRCA1 and BRCA2, the most important genes involved in breast cancer.

Up to 10% of all female breast cancers are thought to be due to mutations in BRCA1 or BRCA2.

The Supreme Court's decision effectively barred patents on naturally occurring genes. That, along with next-generation sequencing, which enables testing for multiple genes in one assay, contributed to a large drop in the price of genetic testing. The net effect has been a considerable expansion in the use of multigene panels for breast cancer.

Laboratories across the United States now provide genetic screening panels that may contain over 100 genes. Clinicians have a wide range of options to choose from: single-gene testing, multigene panels with high-risk genes only, panels with genes for other common cancers, such as colon or ovarian cancer, and comprehensive panels with genes for both common and rare cancers.

Critics of large multigene panels say that they create more questions than answers. No good evidence exists regarding some of these genes; some are only remotely related to breast cancer; guidance is murky on how to manage patients with positive results; and larger panels turn up more VUS, they argue.

The ensuing confusion can cause undue anxiety and harm patients.

"It's this dangerous situation in which you're giving information based on limited or wrong data to someone who has been diagnosed with breast cancer and is already vulnerable," Tischkowitz commented to Medscape Medical News.

It's human nature to want larger panels — bigger usually sounds better, he pointed out.

"But more genes actually creates more problems. It's not the way to go," he argues. "This sort of approach can really ruin lives.

"People who support large panels are saying basically it's fine to test for all these genes, and we're saying hang on a second, be careful," Tischkowitz said.

Wave of the Future?

Others in the field argue that multigene panels are the wave of the future.

Quoting evidence suggesting that current testing misses roughly 50% of cases involving clinically actionable mutations, proponents argue that current guidelines for genetic testing are too restrictive.

Larger panels may guide choice of targeted treatments and enable cascade testing of families to help prevent future cancer. And testing with the biggest panel first is more efficient — ordering smaller panels can waste valuable time for a patient who has been diagnosed with cancer.

"Those fighting this and other recommendations for expanding testing are fighting a losing battle that is morally wrong to begin with. The public wants more genetic testing, not less, and they can handle the results," commented Beitsch, first author of the recent study that poured fuel on the debate.

That study was conducted in 20 community and academic medical centers in the United States. It included data from 959 patients with breast cancer who had not already undergone genetic testing. About half of the participants met National Comprehensive Cancer Network (NCCN) 2017 criteria for genetic testing, and half did not.

NCCN recommendations for genetic testing in breast cancer are under review. Current guidelines are based on clinical risk factors such as age, hormone receptor status, Ashkenazi Jewish ancestry, and personal and family history of cancer.

In the study, participants underwent genetic testing with an 80-gene panel; the results were analyzed at Invitae genetics laboratory in San Francisco, California.

Results suggested there was no difference in pathogenic/likely pathogenic (P/LP) genetic mutations between women who met NCCN guidelines and women who did not (9.39% vs 7.9%, respectively; P = .4241).

The authors pointed out that most of these P/LP variants were in genes related to breast cancer or other hereditary cancers and were genes for which management recommendations exist.

Results using an 11-gene panel suggested that restricting testing to only those women who met NCCN guidelines would miss about 45% of women with P/LP variants: P/LP variants were found in 6.26% of women who met guidelines, vs 3.54% of women who did not.

Overall, 54.22% of women tested positive for VUS. Rates of VUS were "virtually identical" between women who did and those who did not meet NCCN guidelines, the authors write. But they do not report numbers, percentages, or level of significance for VUS.

These results led to a conclusion that may have been too bold for some critics.

"We recommend that all patients with a diagnosis of breast cancer undergo expanded panel testing.... [G]uidelines should be expanded immediately to include genetic testing of all patients with breast cancer," Beitsch and colleagues wrote.

Current Recommendations

The American Society for Breast Surgeons (ASBrS) issued consensus guidelines in February 2019 in which it recommended that all women with breast cancer undergo genetic testing. But the ASBrS recommendation leaves open how careful providers should be in using genetic screening for breast cancer.

Speaking on behalf of the ASBrS via email, Eric R. Manahan, MD, MBA, FACS, who is affiliated with Hamilton Medical Center, Dalton, Georgia, filled in the gap: "Multigene panels are supported by the society as they can be a more efficient and cost-effective way to evaluate genes that confer risk and impact management guidelines."

But he added, "The number of the particular genes and panels are debatable as to their actionability."

Although the NCCN guidelines do not address whether or not clinicians should use multigene panels, they do contain a statement on multigene testing. Mary B. Daly, MD, PhD, chair of the NCCN Guidelines Panel for Genetic/Familial High-Risk Assessment in Breast and Ovarian Cancer said:

"Our policy is for people to consider multigene testing when they think that it's more appropriate than going gene by gene. We always leave open the option of looking for one particular gene. I would say the majority of people are now tested with some sort of a multigene panel."

Yet she voiced qualms about universal testing.

"I think statements like that from the ASBrS can do a lot of harm because it raises questions in people's minds before we have all the right answers," she said.

That's because the groundwork has yet to be laid for the use of large panels, say critics.

First and foremost, a better understanding of the science of breast cancer genetics is needed.

"In pathology, they say garbage in, garbage out. Multigene panels are not a bad idea altogether, but they're not ready for prime time unless we understand the genetics and the molecular biology better," said Mehmet Sitki Copur, MD, FACP, from Mary Lanning Healthcare in Hastings, Nebraska.

He and two colleagues wrote a letter to the editor in which they "respectfully differ with the recommendation" made by Beitsch and colleagues.

Resources are an important part of the groundwork, and not all of those are in place, they argued.

Quality of testing, interpretation of results, and turnaround times vary by laboratory, and quality assurance is needed to ensure accurate results and appropriate follow-up.

Insurance Coverage and Costs May Be an Issue

For insurance to cover these panels, pre- and post-test genetic counseling is required. And even then, insurance may not cover. Although Medicare covers BRCA1/2 testing, Medicaid coverage for genetic testing varies by state. As of January 2018, fifteen states did not cover testing, noted Copur.

"Cost barriers to the receipt of counseling and testing cannot be ignored. The costs to patients may be prohibitive in the most vulnerable populations," said Copur, speaking from experience. The majority of his patients are insured through Medicare or Medicaid.

Moreover, in his Nebraska practice, Copur has access to a genetic counselor only once every 2 weeks, a fact that highlights the shortage of genetic counselors in some areas of the United States.

Genetic counselors are needed for informed consent, which many agree is critical, especially for large panels. Daly, from the NCCN, said she doubts that busy practices will have the time to appropriately obtain consent from patients.

Beitsch and colleagues argue that patients would want to know.

In a letter published in the Journal of Clinical Oncology in response to critics, they argue that opponents of large panels who challenge the value of testing genes beyond those that have proven associations with breast cancer have a narrow focus that "values the view of the specialist more than that of patients."

They argue that patients "seek overall health care, including risk and prevention of illness in all organ systems.

"We do not propose indiscriminate expanded panel testing, rather germline genetic testing of all patients with a personal diagnosis of breast cancer," they add.

Several of the critics noted that many of the study's authors have ties to Invitae and similar companies that are involved in genetic testing. Addressing this in an email to Medscape Medical News, Beitsch said that pharmaceutical, medical device, and testing companies routinely support studies that involve their products. All authors followed appropriate conflict of interest protocols, including institutional review board approval, data interpretation by non-Invitae physicians, and author disclosures, he added.

Beitsch agrees that physicians and patients need more education about genetic testing, especially about VUS.

For example, in the case of the patient with the VUS in her CDH1 gene cited above, Beitsch said there was about a 98% chance that it was a benign variant and that the patient should have been told not to act on it.

It is physicians who induce anxiety in the patient, not the test itself, he added.

"The need for more education should not stand in the way of expanded testing," he went on. "VUS counselling does not have to be complex from a provider or patient perspective. We simply tell our patients that these findings are not meaningful at this time, but as evidence changes we will follow up and make additional recommendations."

In their study, positive results involving pathogenic/likely pathogenic variants were found in just 25 genes, he continued.

"For those clinicians who are not comfortable using an 80-gene panel, data support multigene testing that includes these 25 genes," he said.

Which brings up the question: Is it possible to reach consensus about which genes to include in these panels?

Yes, say some physicians in the United Kingdom who have already reached consensus.

In 2018, the UK Cancer Genetics Groups wrote a consensus article in which experts agreed that a more targeted panel of eight genes that have proven clinical utility be included in breast cancer genetic screening panels. That panel is available in clinical laboratories in the United Kingdom. It is intended to serve as a national standard for adoption by the NHS. The group has also developed management proposals for patients whose test results are positive.

The ultimate goal is to test every woman with breast cancer, but that hasn't happened yet. To increase accessibility, genomic services in the Unied Kingdom are being reorganized, and a national test directory has been developed. Genetic tests will be covered by the NHS for patients who meet defined clinical criteria.

The United Kingdom's more controlled approach, compared to the relative free-market situation in the United States, reflects broader cultural differences between the two countries and their healthcare systems.

Yet Daly of the NCCN said that a consensus panel would not be unwelcome in the United States.

"I think it would be helpful [to have a consensus], but keep in mind that it won't be just one panel. There will be panels for different cancers and different syndromes," she said.

These panels will need to be based on penetrance estimates, or the likelihood that those genes lead to breast cancer. The main hurdle again lies in the science and the need for better data on breast cancer genetics, she added.

"If we were absolutely certain that we have the data to support testing for every breast cancer patient, we shouldn't let resource issues stop us," she said.

Some of the groundwork is already being laid to solve the scientific and resource problems.

Although researchers around the globe are working on pieces of the scientific puzzle, registries such as Ask2Me, Breast Cancer Information Core, ClinGen, ClinVar, the ENIGMA consortium, and iGAP (which supports universal testing for patients at risk for cancer) are cataloguing hereditary cancer susceptibility genes.

To improve the quality of laboratory testing, the Association for Molecular Pathology, the American Society of Clinical Oncology, and the College of American Pathologists have developed a consensus guideline for the interpretation and reporting of genetic variants in cancer.

The ASBrS is collaborating with other societies to develop educational materials to improve provider knowledge about testing and interpretation of results.

In the future, alternative service delivery models, such as telemedicine, may improve the shortage of genetic counselors and provide more equal access to testing.

All this is cause for hopeful, but what about that patient with the worrisome variant in her CDH1 gene?

The patient was lost to follow-up, so Tischkowitz is not sure what happened to her.

"I suspect she didn't have her stomach removed, and I suspect she's still living with anxiety because of it," he said.

Whether those comments would apply to every patient in a similar situation is a question in search of an answer.

Authors' and editorialists' relevant financial relationships are listed in the original articles.

J Clin Oncol. Beitsch et al, Full text; Taylor and Tischkowitz, Full text; Copur et al, Full text

Follow Medscape on Facebook, Twitter, Instagram, and YouTube

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....