Skip to main content
Premium Trial:

Request an Annual Quote

Secondary Tumor Risk Variants Sought in Pediatric Cancer Survivors

CHICAGO (GenomeWeb) – At the American Society of Clinical Oncology meeting here this weekend, members of two research teams described efforts to identify germline risk variants for secondary neoplasm development in individuals treated for pediatric cancer.

During a session on pediatric oncology here Sunday, the University of Alabama at Birmingham's Smita Bhatia presented initial findings from a genome-wide association study involving thousands of previous pediatric cancer patients who did or did not develop meningioma, mainly benign tumors that spring from the meninges tissues that surround the brain.

Therapeutic exposures to radiation and chemotherapy are risk factors for the development of subsequent neoplasms, Bhatia noted, and meningioma is more common amongst individuals who have received higher doses of cranial radiation.

Still, she explained, there are significant differences in secondary tumor development within groups of individuals who have been treated for pediatric cancer.

In an effort to uncover genetic contributors to this process, the team brought together cases and controls of European descent from two large pediatric patient cohorts: the Childhood Cancer Survivor Study (CCSS), which includes individuals enrolled from 1970 to 1986 who survived cancer for at least five years, and the St. Jude Lifetime Cohort — a collection spanning the early 1960s to 2005 that includes individuals with childhood survival rates of 10 years or more.

The team drew from these cohorts to find 167 past pediatric cancer patients who developed meningioma years or even decades after being treated with radiation (the median age at meningioma diagnosis was 30 years old) and 5,732 pediatric cancer survivor controls who remained secondary neoplasm-free.

In these cases and controls, the researchers compared patterns at almost 17 million SNPs or small insertions and deletions, representing variants that were directly genotyped with Illumina 5M Exome or Affymetrix 6.0 arrays or imputation based on data from the 1000 Genomes Project.

Their search led to sites on chromosomes 4 and 7, Bhatia said, with the chromosome 7 SNP showing the most robust ties to secondary meningioma risk. That variant, which was found in 4 percent of cases and 1 percent of controls, falls near a gene called GPR85 that is selectively expressed in neurons and appears to influence neuronal growth.

Bhatia cautioned that follow-up studies are needed to confirm this association in larger patient populations and to explore the functional consequences of the variant, particularly within the context of exposure to radiation.

Even so, she and her colleagues argued that there may eventually be a clinical benefit to doing enhanced surveillance in pediatric cancer survivors deemed at risk of meningioma based on their radiation exposure and genetic risk.

The team is interested in following up on potential associations that they did not have the power to explore fully based on the size of the current GWAS. It has also started to do exome sequencing on pediatric survivors from Childhood Cancer Survivor Study and St. Jude cohorts.

During another presentation during the session, Xuexia Wang from the University of Wisconsin at Milwaukee introduced clinical and genetic approaches that are being used to try to predict secondary neoplasm risk in prior pediatric cancer patients.

For that effort — a Children's Oncology Group study led by Bhatia — investigators set out to develop risk prediction models to find childhood cancer survivors at greatest risk of developing secondary neoplasms in the central nervous system.

Using data on 82 childhood cancer survivors who developed meningioma or glioma and 228 pediatric cancer survivor controls, the team started out by searching for nearly 100 variants in and around 43 genes that were previously implicated in central nervous system tumor risk.

From there, the investigators tested a handful of predictive models, including models that took into account basic features such as age at primary cancer diagnosis to clinical, genetic, or combined models. Their secondary neoplasm risk prediction modeling appeared to be most accurate when genetic variants in several DNA repair-related genes were considered, Wang said, suggesting defects in this process might make secondary tumor formation more likely following radiation or other sources of therapeutic DNA damage.

Will Parsons, director of the Texas Children's Hospital pediatric center for personal cancer genomics and therapeutics, shared his thoughts on findings from both late effect susceptibility projects within the broader context of precision oncology.

As more patients survive childhood cancer, there is an increasing focus on following outcomes in those with successful treatment for their primary cancers, Parsons noted. But, he said, it is difficult to know whether such findings would ever alter upfront treatment for pediatric patients, particularly if variants associated with secondary neoplasm risk are found in a subset of unaffected controls.