Working Toward Greater Precision in Childhood Cancers
Posted on by Dr. Francis Collins
Each year, more than 15,000 American children and teenagers will be diagnosed with cancer. While great progress has been made in treating many types of childhood cancer, it remains the leading cause of disease-related death among kids who make it past infancy in the United States . One reason for that sobering reality is our relatively limited knowledge about the precise biological mechanisms responsible for childhood cancers—information vital for designing targeted therapies to fight the disease in all its varied forms.
Now, two complementary studies have brought into clearer focus the genomic landscapes of many types of childhood cancer [2, 3]. The studies, which analyzed DNA data representing tumor and normal tissue from more than 2,600 young people with cancer, uncovered thousands of genomic alterations in about 200 different genes that appear to drive childhood cancers. These so-called “driver genes” included many that were different than those found in similar studies of adult cancers, as well as a considerable number of mutations that appear amenable to targeting with precision therapies already available or under development.
In one of the two studies reported in Nature, an NIH-funded team led by Jinghui Zhang at St Jude Children’s Research Hospital, Memphis, TN, compiled DNA data representing tumor and normal tissue from 1,699 patients with childhood cancers. Determining the DNA sequence from normal tissue allows for comparison between the mutations that were present at conception (called “germline”) and those that have arisen during life in the cancer cells (called “somatic”). These cancers included many forms of leukemia and a variety of solid tumors. Nearly all patients had received a cancer diagnosis at age 20 or younger.
The other study, led by Stefan Pfister at the Hopp-Children’s Cancer Center at the NCT Heidelberg, Germany, examined existing DNA data on tumor and normal tissue from 961 children, adolescents, and young adults. The analysis involved 24 distinct cancers, covering all the most common childhood cancers. The study also included many patients with various brain cancers, which weren’t represented in the other study.
Together, these studies offer a remarkably complete picture of childhood cancer. These so-called pan-cancer studies offer a way to analyze a massive dataset and sort out biological processes that are either common to certain cancers, or unique to an individual tumor or cancer type. While such investigations had been conducted in adult cancers, the new studies are the first to do so at this scale in childhood cancers.
By comparing DNA sequencing data in cancerous and healthy tissue, Zhang’s team uncovered 142 driver genes that are somatically altered in cancer cells. The genomic alterations included small, single-letter changes in the DNA “nucleotide” alphabet. But the majority (62 percent) represented changes in the number of copies of a particular gene in the genome or larger structural rearrangements. Interestingly, only 45 percent of these driver genes had been identified in previous studies of adult cancers.
Zhang’s team also uncovered 11 mutational signatures, which are specific mutation patterns that suggest their mechanism of origin. One of those signatures appeared quite unexpectedly to be consistent with mutational events caused by ultraviolet-light (UV) exposure. That signature occurred in eight patients with B-cell acute lymphoblastic leukemia (B-ALL), the most common type of childhood cancer. While more research is needed, the discovery makes the surprising suggestion that UV exposure, or perhaps other environmental factors that drive similar mutational signatures, may play a role in some cases of childhood leukemia.
The second study presents a remarkably similar overall picture despite analyzing distinct tumor types. Pfister’s team identified genetic alterations, including small mutations and larger copy number and structural rearrangements, in 149 cancer driver genes. Only about one-third of those genes had been associated with adult cancers previously.
Their analysis of normal and tumor tissue also indicates that about 8 percent of young people may have inherited a germline mutation from their parents that made them susceptible to cancer. That’s important because those kids may be at increased risk for developing secondary cancers and benefit from fundamentally different treatment approaches. It also suggests a significant proportion of families affected by childhood cancers may benefit from genetic counseling to determine whether other family members carry the same mutations, putting them at increased cancer risk.
Overall the studies confirm that childhood cancers often carry many fewer mutations than adult cancers. Many of the driver genes also appear to be involved in causing one specific type of childhood cancer. That’s different than adult cancers, in which mutations in particular genes often drive multiple forms of cancer.
The findings also have important implications for precision medicine treatment. In fact, an analysis by Pfister’s team suggests that about half of the childhood cancers in their study may be driven by genes for which targeted treatments are already available or in development. This discovery also highlights the need to test specifically for genes that are important in childhood cancers, many of which aren’t captured by tests designed for adults. This childhood-centered testing will help to ensure that kids receive the most-effective treatments to control their tumors.
In cases where a potentially promising, targeted treatment does not yet exist, this new list of driver genes now provides fodder for continued research and drug discovery. Toward that end, it will be important to continue to amass data on more childhood cancers and to conduct studies designed to explore the underlying biological mechanisms by which these genes drive cancer. To help speed that process along, the researchers have made all of the data from both studies freely available online to researchers and clinicians around the world.
 Cancer in Children and Adolescents.National Cancer Institute, August 24, 2017.
 Pan-cancer genome and transcriptome analyses of 1,699 paediatric leukaemias and solid tumors. Ma X, Liu Y, Liu Y, Alexandrov LB, Edmonson MN, Gawad C, Zhou X, Li Y, Rusch MC, Easton J, Huether R, Gonzalez-Pena V, Wilkinson MR, Hermida LC, Davis S, Sioson E, Pounds S, Cao X, Ries RE, Wang Z, Chen X, Dong L, Diskin SJ, Smith MA, Guidry Auvil JM, Meltzer PS, Lau CC, Perlman EJ, Maris JM, Meshinchi S, Hunger SP, Gerhard DS, Zhang J. Nature. 2018 Feb 28. [Epub ahead of publication]
 The landscape of genomic alterations across childhood cancers. Grobner S et al. Nature. 28 Feb 2018. [Epub ahead of publication]
Childhood Cancers (National Cancer Institute/NIH)
NCI-COG Pediatric MATCH (NIH)
ProteinPaint (St Jude, Memphis, TN)
Pediatric Pan-Cancer Study (German Cancer Research Center, Heidelberg)
The Cancer Genome Atlas (NIH)
Stefan Pfister (German Cancer Research Center)
Zhang Lab (St Jude, Memphis, TN)
NIH Support: National Cancer Institute