NEW YORK – A new tool can flag variants that may be of germline origin and possibly associated with heritable risk of multiple myeloma from within somatic variants detected in patients via next-generation sequencing, researchers reported Sunday at the American Society of Hematology's annual meeting.
In a study published in Blood in 2021, a team led by Santiago Thibaud, a hematologist at the Icahn School of Medicine at Mount Sinai, demonstrated that pathogenic or likely pathogenic germline variants (PGVs) in genes linked to heritable cancers are common in multiple myeloma patients, particularly in those who have a family history of hematological malignancies and those who were diagnosed with multiple myeloma at a young age. Within germline whole-exome sequencing data from 895 newly diagnosed multiple myeloma patients, they found PGVs in 8.8 percent of the cohort.
In the current study presented at ASH, Thibaud's team aimed to find out whether PGVs can be inferred from variants detected through targeted somatic sequencing of multiple myeloma patients. "Since hereditary cancer genes are included in most commercially available assays, when we submit whole bone marrow aspirates for tumor profiling in myeloma, we need to interpret the results with caution," Thibaud said. "Most reported pathogenic variants may indeed be somatic in origin, but without concurrent normal tissue sampling, we cannot exclude some of these variants being present in the germline."
Thibaud said that more data from the literature are needed to more precisely estimate the likelihood a genetic variant is somatic, germline, or clonal hematopoiesis of indeterminate potential based on variant allele frequency.
"These types of flagging tools have been developed and published primarily for pan-cancer use, but it is critical to develop tumor-specific tools to flag variants needing referral for confirmatory germline testing and genetic counseling," Thibaud said.
The researchers conducted a retrospective review of 1,715 next-generation sequencing reports from 1,161 multiple myeloma patients at Mount Sinai generated between 2015 and 2022. The tests were conducted on a single testing company's 448-gene somatic profiling panel on patients' bone marrow aspirates.
Thibaud's team worked with genetic counselors at Mount Sinai to create a variant classification tool that could flag PGVs that might have germline origins. The resulting tool is designed to identify PGVs in 35 distinct hereditary cancer genes when the variant allele frequency is more than 10 percent, as well as 24 founder variants at any variant allele frequency.
Although most germline variants will have a variant allele frequency of 50 percent or more, Thibaud said the 10 percent threshold was chosen to optimize sensitivity because the tool is intended for screening purposes.
As part of the tool's development, investigators validated suspected PGVs as bona fide PGVs using germline whole-exome sequencing in one-third of patients whose samples had been previously collected and sequenced for somatic variants.
After that, Thibaud's team applied the classification tool to the entire patient cohort. In the 382 patients with confirmatory germline whole-exome sequencing, 41 out of 385 variants, or 10.6 percent, were flagged as suspected PGVs. In the 779 patients who did not have confirmatory germline whole-exome sequencing, 86 out of 779 variants, or 11.1 percent, were flagged as suspected PGVs.
Overall, the percentage of suspected PGVs flagged was 10.9 percent. They calculated the diagnostic sensitivity of the tool at 84 percent and the specificity at 97 percent for identifying bona fide PGVs.
Looking at the number of variants found per gene, Thibaud and his collaborators found that there were high numbers of variants in familial cancer risk genes such as BRCA1/2, CHEK2, APC, and ATM that clustered around biological DNA repair pathways.
"Our main point was to prove that through analyzing somatic mutational data, we could infer who was likely to have a PGV," Thibaud said. If the tool can be validated for clinical use, Thibaud said, "it would be indicated in someone who has one of these mutations in somatic testing to be referred for germline testing, because we know that there's a high likelihood that they'll be positive and that will have an impact on the patient and their family."
Although most genetic studies have been carried out in patients of white European ancestry, the Mount Sinai study benefited from a diverse patient population that was 20 percent Black and 40 percent non-white. "That is very interesting, considering the incidence of myeloma is twice as high in Black patients as compared to others," Thibaud said.
Thibaud said if the tool is successfully validated in prospective studies, it may be useful for guiding treatment decisions for patients, particularly when it flags germline PGVs that affect homologous recombination repair. In Thibaud's earlier study published in Blood, those patients had a greater likelihood of being diagnosed at a younger age, a greater chance of having had a prior cancer or a family history of cancer, and a tendency toward favorable responses to high doses of the chemotherapy drug melphalan and stem cell transplantation.
On the basis of the current findings, Thibaud's group has initiated a pilot study to test the feasibility of screening multiple myeloma patients for PGVs. In the study, they are referring multiple myeloma patients considered to be at high risk for being PGV carriers to receive genetic counseling about germline genetic testing and possibly cascade testing if they turn out to be a PGV carrier.
Out of 33 patients who have undergone germline testing in this pilot so far, researchers have found clinically relevant PGVs in 11 patients. "While these findings are preliminary, a prevalence of 33 percent suggests that we are successfully enriching for PGV carriers with our eligibility criteria for this screening methodology," Thibaud said.
Thibaud concluded at the meeting that the variant classification tool has promising diagnostic performance for identifying PGVs in multiple myeloma patients from somatic sequencing data, and that additional refinements and integration of new data can enhance its diagnostic performance. These findings also add to a growing body of evidence supporting implementation of germline genetic screening in high-risk subsets of multiple myeloma patients, he added.