Testing of children and adolescents with cancer revealed that 8.5% had germline mutations in cancer-predisposing genes, according to a report published online November 18 in the New England Journal of Medicine. Some were commonly mutated genes, such as TP53, seen in some pediatric cancers; others were new and were not yet associated with malignancies, such as Ewing's sarcoma.
"This clinically significant study increases the awareness of the fact that childhood cancers do not develop by chance and there are genetic factors that will have an impact on the clinical management of these patients, and ultimately of their families," John M. Maris, MD, Giulio D'Angio Endowed Professor of Pediatric Oncology at the Children's Hospital of Philadelphia, in Pennsylvania, who authored an accompanying editorial, told Medscape Medical News.
"This is a foundational study and for the first time provides an estimate of children who have a hereditary predisposition of cancer," corresponding author James R. Downing, MD, president and CEO of St. Jude Children's Research Hospital, Memphis, Tennessee, told Medscape Medical News.
"Most expected these rates to be low, and these observations have significant implications for clinical practice, some of which are already being instituted by us," he said.
Knowledge about the presence of mutations "may influence clinical management by directing cancer care, enabling presymptomatic genetic testing of relatives, guiding family-planning measures, and facilitating the institution of potentially lifesaving measures for cancer prevention and surveillance," the authors write.
The Pediatric Cancer Genome Project
The report was part of a larger project to define the landscape of somatic tumor mutations called the St. Jude-Washington University Pediatric Cancer Genome Project (PCGP).
For this study, investigators looked for germline mutations by sequencing DNA from blood samples of 1120 children using next-generation sequencing, which included whole genomes (595 patients) or whole exomes that look at only coding sequences (456 patients). In 69 patients, investigators used both whole genome and whole exome sequencing.
The PCGP cohort included children (median age, 6.9 years) and adolescents with leukemia (52.5%), central nervous system (CNS) tumors (21.9%), and non-CNS tumors (25.6%). The cohort included more patients with leukemia and adrenocortical tumors compared with what was expected on the basis of the population in the Surveillance, Epidemiology, and End Results (SEER) database.
On the basis of recommendations from the American College of Medical Genetics and Genomics and from medical literature, 565 genes were chosen for analysis, reflecting five broad categories — autosomal-dominant cancer-predisposing genes (60 genes), autosomal-recessive cancer-predisposing genes (29 genes), tyrosine kinase genes (23 genes), tumor suppressor genes (58 genes), and other cancer-associated genes (395 genes).
The 60 genes from the autosomal-dominant category were selected "because of the potential effect of germline mutations on clinical practice, including avoidance of radiation, choice of surgical approach for tumor resection, donor testing and selection of hematopoietic stem-cell transplantation, possible or proven benefits of tumor surveillance and early cancer detection, and risk-reductive surgery," the authors write.
For a control, the investigators analyzed data from 966 persons from the 1000 Genomes Project who did not have any known cancer and from an autism study, which included 515 persons with autism and 208 persons without autism.
Prevalence and Identification of Germline Mutations
From their analysis, the investigators found an 8.5% (95 patients) prevalence of pathogenic or probably pathogenic mutations in the PCGP cohort compared with a prevalence of 1.1% and 0.6% in datasets from the 1000 Genome Project and the autism study, respectively.
When the cases with hypodiploid acute lymphoblastic leukemia and adrenocortical tumors, which were present in the PCGP cohort at a greater-than-expected proportion, were excluded, the prevalence of germline mutations was 5.6% — still significantly higher compared with the control cohorts.
The prevalence of germline mutations was highest in patients with non-CNS tumors (16.7%). Prevalence was 8.6% for patients with CNS tumors and was lowest in patients with leukemia (4.4%).
The investigators found germline TP53 was the most commonly inactivated gene (50 patients), followed by mutations in APC, BRCA2, NF1, RB1, and RUNX1. Germline TP53 mutations were seen in adrenocortical tumors (69%), hypodiploid acute lymphoblastic leukemia (19%), and choroid plexus carcinoma (25%) — consistent with reports from other studies — and were associated with loss of heterozygosity.
Four of the germline mutations were reported to be mosaic — mutations present only in a subset of normal cells and probably arising through the developmental process, Dr Maris explains in his editorial.
The analysis corroborated the correlation between patient genotype and tumor phenotype for several cancers: TP53 mutations and Li-Fraumeni syndrome–associated cancers; NF1 mutations and CNS tumors; RB1 mutations and retinoblastoma and osteosarcoma; and ALK mutations and neuroblastoma.
Importantly, new associations were also identified in the analysis: TP53, PMS2, and RET mutations with Ewing sarcoma; APC and SDHB mutations and neuroblastoma; and APC, VHL, CDH1, PTCH1, or SDHAmutations with leukemias.
Although loss of heterozygosity or inactivation of genes in both alleles was expected for most germline mutations, this was not always observed. Eight children had germline mutations in adult- onset cancer-predisposition genes — BRCA1, BRCA2, and PALB2. Cancers associated with these genes include leukemia, CNS tumors, neuroblastoma, osteosarcoma, and rhabdosarcoma.
"Although biallelic mutations of BRCA1/2 and PALB2 are known to cause Fanconi's anemia, there were no germline mutations or deletions involving the second alleles of these genes in any of the affected patients," the investigators write.
Family History Is Unreliable Guide
A family history is often used to identify a pediatric cancer population with a heritable disposition; however, in the study, only 40% of patients with pathogenic or probably pathogenic germline mutations had a family history of cancer.
The investigators believe that it is possible that the genes in the study were not precisely the ones identified from the family history. "Nonetheless...family history cannot be the sole indication used to guide the provision of genetic testing," the investigators write.
In his editorial, Dr Maris contends that this study is "just the tip of the iceberg in terms of the complexity of cancer susceptibility."
According to him, this study provides us with reason to pause "to reconsider how to approach the pediatric population with cancer.
"At a minimum, this work highlights the fact that family history alone is an unreliable guide to the likelihood of a cancer predisposition syndrome in any patient with a newly diagnosed cancer," he writes in his editorial.
"[Data from this study] provocatively suggest that germline sequencing should be routinely incorporated into clinical care, given that many pediatric patients with cancer have a genetic predisposition," Dr Maris writes in his editorial.
"We are moving very quickly to a place where genetic sequencing will be an important part of one's health records," Dr Maris told Medscape Medical News. "It is important that we begin with germline sequences in the pediatric population with cancer," he added.
Delineating germline sequences in children with cancers has important implications for clinical practice, he explained. "Knowing germline mutations will help us improve cure rates with more precise therapies," he said.
Already Incorporated Into Practice
Some of this is already being done at St. Jude Children's Research Hospital. Dr Downing explained to Medscape Medical News that this kind of analysis will alter the treatment approach taken for children with a germline mutation by influencing decisions on choice of surgical, radiation, or chemotherapy approaches.
Several approaches are also being pursued at St. Jude Children's Research Hospital on the basis of these findings. If a child has germline mutation(s), immediate family members such as parents and siblings are offered germline testing and are placed in appropriate programs for monitoring and surveillance. In addition, the genes identified in this study are being placed in a surveillance database that will be used monitor children through their adult life for secondary cancers.
Dr Downing shared information with Medscape Medical News on two efforts being pursued. In the St. Jude Children's Research Hospital Genomes for Kids clinical research study, all children enrolled will have whole genome, whole exome, and RNA sequencing performed on tumor and normal samples to determine somatic tumor mutations and germline mutations. An integrated analysis of these mutations will ultimately guide clinical decision making.
Patients with germline mutations can enroll in the new Hereditary Cancer Predisposition Clinic established at St. Jude Children's Research Hospital, which uses an integrated approach to cancer management. "In this clinic, hematologists, oncologists, medical geneticists, genetic counselors, and medical ethicists participate in cancer care and surveillance of children to establish management guidelines," Dr Downing told Medscape Medical News.
Because more childhood survivors will have children of their own, we need to invest now to enrich our ability to provide world-class and evidence-based cancer-predisposition counseling to families afflicted by childhood cancer, Dr Maris indicates in his editorial.
Dr Downing and Dr Maris have disclosed no relevant financial relationships.