Skip to main content
Premium Trial:

Request an Annual Quote

AACR Panel Discusses Challenges of Reporting Incidental Findings

SAN DIEGO (GenomeWeb News) — Reporting incidental or secondary findings may pose a challenge for the cancer field, a panel at this year's American Association for Cancer Research meeting said.

The panelists particularly focused their discussion on recommendations the American College of Medical Genetics and Genomics issued last year and updated about a week ago. Those guidelines list a set of some 56 genes for which variants should be reported back to patients undergoing clinical sequencing.

Many of the genes on that list are linked to increased cancer risk, as well as susceptibility to cardiac conditions.

Additionally, to search for cancer-causing variants, researchers and clinicians have been turning to sequencing the whole exomes or whole genomes of tumor-normal pairs. This, Kenneth Offit, a medical oncologist from Memorial Sloan-Kettering Cancer Center, noted, means that clinicians could uncover germline variants, like those on the list of incidental findings to be reported.

Further, some of the panelists argued that searching for and reporting secondary findings is a burden on oncologists, And as many research groups wonder whether they too have a duty to warn, there are concerns that such reporting may take away from the research enterprise.

"We have some serious issues we have to grapple with," Offit said during the session.

As he noted, tumor-normal pairs are typically compared to find cancer drivers in the tumor, and many labs perform a subtraction of the normal genome from the tumor genome, leaving only tumor-specific mutations behind. But other labs don't perform that subtraction, leaving the germline and any potentially disease-causing mutations visible.

Still, Sharon Plon, a professor of molecular and human genetics at Baylor College of Medicine, noted that the ACMG guidelines specifically mention such comparisons of tumor-normal pairs. The guidelines say that the normal sample should undergo the evaluation and reporting of secondary findings.

While Plon acknowledged that she was one of the outside experts ACMG consulted on the guidelines, she said the inclusion of this sentence was "a mistake."

She also questioned how the list of 56 genes was drawn up, noting that it is rather idiosyncratic and originally contained 57 genes until one was determined to not have enough evidence backing its inclusion.

ACMG last week updated its recommendations to allow patients to opt out of receiving such incidental findings. The lack of such an opt-out option was a major source of criticism of the guidelines when they were initially released.

Despite that update, Plon said that much of the burden on labs and oncologists is unchanged. Labs must still offer those findings and oncologists must still explain to patients what incidental findings are and what may be found.

A small percentage of cases do have incidental findings to report. As part of the BASIC3 study, which is assessing how sequencing information is relayed to pediatric cancer patients and their parents, Plon and her colleagues found that about 4 percent of patients had incidental findings.

Once a finding is reported back, that triggers a number of follow-up steps for patients, their families, and clinicians.

Commonly, she noted, after a family finds out that one child has a certain disease-causing variant, one of the first questions is about whether other children or family members could have the same variant.

For example, in one child who underwent sequencing to determine the cause of his cancer, Plon said, not only was a cancer-causing variant found, so was a variant linked to cardiomyopathy. Because his father and brother also had that variant, all three needed cardiac workups.

Similarly, Sloan-Kettering's Offit said some variants newly linked to cancer through genome-wide association or similar sorts of studies have varying levels of penetrance, making patient counseling trickier.

Further, curation and functional analysis of such variants is a time-consuming process, he added. "This is going to keep you pretty busy," he said.

While the ACMG guidelines specifically refer to clinical testing, the return of incidental findings could also affect the research enterprise, as the line between research and the clinic have sometimes blurred.

However, the report from The Presidential Commission for the Study of Bioethical Issues on incidental findings, which came out in December, makes a "distinct distinction" between research and the clinic, Offit said.

Indeed, Stephen Chanock from the National Cancer Institute added that how research and clinical testing are conducted have some stark differences. And he argued that the lines between research and clinical care should remain separate.

Clinical tests, for instance, take place in CLIA-certified labs that follow certain formalized steps, and those tests are expected to inform clinical decisions and actions.

Research tests, by contrast, may not be conducted in CLIA labs. Further, there are a number of hurdles, he added, to conducting research in a CLIA-certified lab, not the least of which are the different infrastructure needs and higher cost.

If researchers are supposed to report incidental findings, that also raises questions of to whom they are supposed to relay those results — such as to the PI, local doctor, or local IRB — how those results are conveyed and by whom, and who pays for it and follow-up testing. If the institute or study has to bear the costs, that'll take a bite out of what can go toward the research enterprise, Chanock said.

The research process is also different, Chanock said, noting that it is iterative as new findings build on older ones. This raises the further question of, as new discoveries are made, how long does the obligation to report findings back last? It becomes, he said, impractical.

According to Chanock, there are currently 114 genes known to be mutated in cancer susceptibility complexes, and that number will only increase.

The ACMG guidelines name some 56 genes, 23 of which are cancer-related, to be reported back as secondary findings, and that list, too will likely grow. He said that how such a list is updated and what criteria will be used needs to be established.

"That's where the community has to converge," Chanock said.

ACMG has noted that its list will evolve as more data is generated.

Baylor's Plon argued, though, that new guidelines for returning incidental findings need to developed by a professional cancer organization, such as AACR or the American Society of Clinical Oncology, that draws on the expertise of oncologists.