NEW YORK – Research showing that nearly 5 percent of prenatal specimens tested in a Laboratory Corporation of America lab harbored a pathogenic or likely pathogenic variant associated with hereditary cancer risk has raised questions about how and when clinicians should report a fetus' cancer risk to the parents.
In the study, published in the Journal of Molecular Diagnostics last month, the researchers reviewed 1,354 prenatal specimens that were sent for familial variant testing due to a family history of certain genetic disorders. They analyzed the specimens using Sanger sequencing for variants in 19 genes associated with hereditary cancer risk and for familial variants in 23 lysosomal storage disease (LSD) genes.
Lynne Rosenblum, lead author of the study who conducted this research while in a previous role as a clinical laboratory director at Labcorp, said she was surprised by the number of variants in this dataset associated with hereditary cancer risk or LSD variants and by the breadth of hereditary disorders found that have the potential to increase cancer risk.
"This type of testing, with the deliberate intent of prenatal testing for cancer risk, has been going on in Europe for decades, and it seems to be a new concern here in this country," she added.
Although testing for hereditary cancer risk in the prenatal setting can raise ethical issues, the guidelines are unclear as to how and when this assessment should occur and how this information should be presented to parents.
Some disorders that affect infants and children and can be detected by prenatal genetic tests, such as Noonan syndrome or Bloom syndrome, are also associated with risk of developing certain cancers. These can be risks for cancers that occur in childhood or later in life. In the case of Noonan syndrome, disease-associated variants in BRAF, KRAS, LZTR1, MAP2K, and PTPN11 genes may also confer risk for melanoma, glioblastoma, lung, and colorectal cancers, as well as for childhood leukemias. However, the cancer risk is considered by healthcare professionals to be a secondary or incidental finding because the primary reason for testing was to gauge the risk of other genetic disorders.
Rosenblum, who is now an assistant professor of pathology at Wake Forest University School of Medicine, and her colleagues found that across prenatal testing guidelines put forth by various professional societies, there was little consensus on how to handle knowledge of such secondary cancer risks. In a 2016 position statement, the National Society of Genetic Counselors (NSGC), for example, encouraged "deferring prenatal genetic testing for adult-onset conditions if pregnancy management will not be affected."
"Prospective parents have the right to make fully informed and autonomous decisions about reproductive options and pregnancy management," the group wrote in the position statement. "However, prenatal testing for adult-onset conditions denies the future child the opportunity to make this decision for him/herself as an adult."
The NSGC outlined potential ethical issues for conducting prenatal testing for these conditions, such as a child's right not to know about their potential hereditary risk, possibly due to concerns about stigma. Further, knowledge of future cancer risks at an early age could impact their life decisions, cause adverse effects on childhood including parents being overprotective or restrictive of an at-risk child, which could negatively affect a child's self-image. Then, there are privacy concerns.
The American College of Obstetricians and Gynecologists (ACOG), meanwhile, states in its guidelines that prenatal genetic testing should be used to "detect health problems that could affect the woman, fetus, or newborn" and be used to help inform pregnancy management.
Finally, the American College of Medical Genetics and Genomics (ACMG)'s guidelines on prenatal genetic testing recommend reporting incidental findings, or results not related to the primary test indication, for "highly penetrant pathogenic variants detected in genes unrelated to the fetal phenotype, but known to cause moderate to severe childhood onset disorders."
In the Journal of Molecular Diagnostics paper, the researchers found that 4.8 percent of the prenatal specimens sent for familial variant testing harbored pathogenic variants in established hereditary cancer risk genes. In the same dataset, three times as many specimens (15.6 percent) harbored variants in LSD genes. The connection to cancer risk is less well established for LSD gene variants, but Rosenblum and her colleagues were interested in exploring their prevalence because with more research they may become more clinically relevant.
"The proteins involved in lysosomal storage disorders may also have a role in the initiation of progression of cancer," she explained. "What we recognize as cancer risk genes now is likely to change and evolve, and we as geneticists should be thinking proactively."
The researchers put the cancer risk genes into four categories: DNA repair and instability genes, RASopathy-Noonan syndrome genes, genetic disorder genes that confer hereditary cancer risk, and purely hereditary cancer risk genes. The first three categories include genes that, when altered, are primarily associated with childhood genetic disorders and tend to be included in prenatal genetic screening testing. But variants in those genes may also indicate an increased cancer risk as a secondary finding.
Among the DNA repair and instability genes tested in this dataset were ATM and BLM. ATM variants are associated with the genetic neurological disorder ataxia-telangiectasia, and BLM variants are associated with the growth disorder Bloom syndrome. Both conditions are also associated with early-onset cancers. There were two specimens affected by these genetic conditions and four specimens that were carriers of variants in DNA repair and instability genes in this dataset.
The RASopathy genes include those in the RAS/MAPK pathway, such as KRAS, BRAF, and LZTR1. Variants in these genes are often a sign that a fetus has Noonan syndrome. However, variants in RAS/MAPK genes may also increase the predisposition for several childhood and adult tumors. In this dataset, 14 specimens harbored variants in these genes.
The researchers also assessed the dataset for variants in genes that identify genetic disorders that confer cancer risk, such as APC, NF1, BTK, and MPL. Four prenatal specimens in the dataset harbored pathogenic variants in these genes. Pathogenic variants in APC cause familial adenomatous polyposis, which can cause colon polyps in children. As the child grows into an adult and the disease progresses, there's a risk of developing colorectal cancer and cancers of the small bowel, thyroid, and other organs. Other genes in this category, like NF1 and BTK, are associated with a risk of childhood cancers like leukemia, lymphoma, and glioma. The ACMG and some other professional societies recommend reporting secondary findings such as variants associated with cancer risk in several genes in this category, including APC, NF1, and RET.
Finally, the researchers explored genes only associated with hereditary cancer risk, including RB1, PALB2, TP53, and others. In their dataset, they only identified one case with a variant in such a gene: a familial RB1 variant, which can cause retinoblastoma that tends to manifest before the age of 5 and increases the risk for other primary tumors like osteosarcomas, soft tissue sarcomas, and melanomas.
For LSD genes, 210 specimens were assessed for 23 disease-linked genes. Of these, 46 fetuses were considered positive for these variants because they had homozygous, hemizygous, or compound heterozygous variants; 105 had heterozygous variants or were carriers of the variants; and 59 were negative for a familial variant.
The researchers acknowledged in the paper that testing and reporting these secondary cancer risk variants from prenatal testing can raise ethical issues, particularly since they cannot immediately affect pregnancy management or the child's medical management.
"One of the major differences between inherited cancer susceptibility and congenital disorders is that for [hereditary cancer risk] disorders, onset is typically later and penetrance can be more variable," they wrote. "The likelihood of developing cancer can range from a low or moderate increase above the general population risk to near absolute and may be mitigated by monitoring and medical intervention, thus generating special concerns regarding prenatal testing for HCR."
Rosenblum noted that, for now, whether these secondary cancer risk findings are reported can depend on the lab doing the testing and on the clinician delivering the report findings to families. When Rosenblum was at Labcorp, she said these secondary findings about cancer risk were typically included in reports in a section separate from the main test results.
She said guidelines outlining a list of genes and incidental findings that should be relayed in prenatal genetic testing reports could help ensure parents get information appropriately. She added that many professional societies in the genetics space have guidelines on reporting certain secondary and incidental findings when adults receive genetic testing, but not in the prenatal setting.
"There may be real benefits to informing the parents, [since their children] could have earlier intervention or earlier surveillance if it's a cancer that might have earlier onset," she said. "On the other hand, it could cause undue worry and stress if there's a very low risk or very low increase in cancer risk, like if it's still a fraction of a percent increase [with the pathogenic variant]. Then the question is, is that useful information?"