A substantial number of children with cancer carry cancer-predisposing mutations inherited from a parent, according to a new study published Wednesday in The New England Journal of Medicine. Researchers examined the genes of more than a thousand children with cancer and found that 8.5 percent of them—most of whom had little family history of cancer—carry handed-down gene mutations that make them more susceptible to the disease. The figure might seem small, especially within a relatively rare diagnosis, but its implications are large—for the young patients and for their family members who might also be at risk—and could help doctors select more appropriate treatments.

Pediatric cancer is a mysterious case. Most cancers in the general population are caused by genetic mutations accumulated over a lifetime. But kids have not been around long enough to experience large amounts of UV rays, chemicals or other exposures that can lead to or exacerbate errors in DNA and spur cancer growth.

These findings "give us fundamental insights into the etiology of pediatric cancer—to help treat patients," says James Downing, chair of Childhood Cancer Treatment, president and CEO of St. Jude Children's Research Hospital and co-author of the new study, which is part of the Saint Jude–Washington University Pediatric Cancer Genome Project.

The researchers analyzed the genes of 1,120 pediatric cancer patients (median age 6.9), honing in on 60 genes that are associated with cancer predisposition to look for DNA errors passed down from a parent, known as germ-line mutations. These cancer-linked errors were present in 95 patients, most commonly in children with solid tumors outside of the central nervous system—and least frequently in those with leukemia. "This kind of work helps in understanding where the tumors are coming from—and also what are some of the underlying driving mutations," says Lisa Diller, chief medical officer of Dana–Farber/Boston Children's Cancer and Blood Disorders Center, who was not involved in the new work

For comparison, the team also analyzed these same genes in 966 (noncancer patient) adults in the general population as represented by the 1,000 Genomes Project and found these germ-line mutations in only about 1.1 percent of people.

The researchers uncovered data from the pediatric cancer population that might alter long-standing traditions. "A big surprise is the family history," says Jinghui Zhang, chair of the Computational Biology department at Saint Jude's and a co-author of the study. Or, more precisely, the lack thereof. Family histories have been a central tool for understanding the nature of a cancer and identifying persons who might be at risk. But Zhang and her colleagues found that of the patients with these germ-line mutations and with family cancer data, only 40 percent had a family history of cancer (which was a similar rate to the pediatric patients who did not have one of these flagged mutations). Furthermore, only 57 percent of the family cancers matched with the genetic predisposition.

Test and screen?
The heritable nature of these genetic mutations means that some in a patient's family—parents or siblings—might indeed harbor the same errors. But instituting testing and screening for all close relatives of a patient's is not an automatic next step. Although a family will likely want to know: "How likely is it that the germ-line mutation would progress to a harmful cancer?" Downing notes the answer will vary for different mutations. Heritable syndromes, including cancer predisposition, have a wide range of likelihoods of actually causing the disease, a concept known as penetrance. Many mutations—which exponentially raise risk for carriers against that of the general population—only go on to cause cancer in a relatively small fraction of carriers, Zhang explains. For most of these variants, however, "the penetrance is unknown," she says

Some of the risk might become known when a parent is diagnosed with cancer. Interestingly, a handful of children with cancer harbored germ-line mutations on the BRACA1 or BRACA2 genes, which are typically associated with adult-onset cancers. In her hospital, Diller says, they reassure mothers with BRACA-associated breast cancer that any risk they might have passed down to their offspring is only for cancer later in life. Now, Diller says, "we'll have to look at if we need to change that policy—whether we should consider children of BRACA carriers at a higher risk" for cancers much earlier in life.

In the meantime knowing that a patient has one of these mutations can help their doctors select treatment. "When a child has a mutation in these genes, it really tells us to stop and pause," Downing says. "We really need to approach that individual patient differently." Some gene signatures, for instance, raise the individual's susceptibility to the effects of radiation, a treatment that might otherwise be standard.

Additionally, Diller points out, having a germ-line mutation that is linked to a cancer in a paired organ, such as a kidney or retina, means that the other organ is also at higher risk. So instead of removing a kidney, as might be common in severe renal adenocarcinomas, doctors might prefer to do their best to save that organ, knowing that the other one might fall victim to cancer as well.

Bone marrow transplants, a common treatment for childhood cancers, might also need new considerations for these patients. Among the best marrow donors are matched siblings. But, as Diller notes, knowing these germ-line mutations' 50 percent probability of being present in a sibling, "you would want to make sure that the matched sibling is not also a carrier of the gene syndrome," she says.

Growing data, spreading syndromes
The findings provide the basis for a new program at Saint Jude's known as Genomes for Kids, or G4K, which will test all new admitted pediatric cancer patients for germ-line mutations. This will not only allow clinicians to use the results when deciding on treatment and family screening but help researchers pin down more links between heritable mutations and childhood cancer.

As more data come in, the frequency of these germ-line predisposition gene mutations will likely rise, Downing notes. "That frequency [of 8.5 percent] will probably be a lower limit," he says. "There will be more and more genes identified." And as the roster of potentially pathogenic gene mutations grows—and as more people have their genomes tested—Diller says we also need to be ready with better information about individual risk, along with helpful screening and lifestyle recommendations. For carriers of these mutations, Downing says, we still do not know if or when surveillance is appropriate. And that is no small issue, as Diller points out. Finding, for example, that a sibling of a cancer patient also carries the mutation, "you immediately make them into a patient rather than a healthy child—just because they carry a gene that we don't know how to interpret," she says.