Study Evaluates Risk of Breast Cancer for Childhood Survivors of Cancer

Zhaoming Wang, PhD

Zhaoming Wang, PhD

Investigators at St. Jude Children’s Research Hospital are currently evaluating the monogenic and polygenic associations to subsequent breast cancer risk in survivors of childhood cancer.

Using the St. Jude Lifetime Cohort Study whole-genome sequencing data, investigators are aiming to identify the role of genetic variants and treatment exposures to subsequent risk of breast cancer in adult survivors of childhood cancer. The combined risk of 11 high- and low-penetrant variants form a polygenic risk score, according to investigators. Variants of high penetrance are BRCA1, BRCA2, CHD1, PTEN, STK11, and TP53; variants of low penetrance are ATM, CHEK2, NBN, NF1, and PALB2.

In findings presented at the 2018 AACR Annual Meeting, 34 of 1131 female survivors selected from the 3000-patient analysis were found to be carriers of pathogenic or likely pathogenic variants in the 11 predisposition genes. While investigators predict that individual genetic profiling in combination with rare pathogenic or likely pathogenic variants may provide a promising breast cancer risk-stratification method for childhood cancer survivors, further data are needed to confirm.

OncLive: Can you discuss your genomic analysis of childhood cancer survivors?

In an interview with OncLive, Zhaoming Wang, PhD, who specializes in computational biology at St. Jude Children’s Research Hospital and is co-lead author of the study, discussed the findings within the breast cancer cohort. Wang explained that this was the first assessment of the joint effects of rare and common genetic variations in survivors of childhood cancers.Wang: We presented a systemic characterization of the germline mutations among 3000 survivors of childhood cancer for the first time at the 2017 AACR Annual Meeting. We did whole-genome sequencing on these 3000 survivors and analyzed the likely pathogenic mutations among those germline mutations, and their genetic contribution to the risk of subsequent breast cancer.

At the 2018 AACR Annual Meeting, I presented on one of the most common subsequent neoplasms, which is breast cancer among female survivors. Before our study, we knew that chest irradiation to the breast tissue is a major risk factor for survivors as they move into adulthood. Literature has also reported the risk of chemotherapy agents, but genetic risk is unknown.

To build on top of last year's whole-genome sequencing data, we looked at not only the rare, highly penetrant mutations in the predisposition genes, we also looked at the genetic variants. We consider rare, highly penetrant mutations together with these genetic variants in the same model. Therefore, we tried to model the risk, including commonly used demographic clinical variables and adding in genetic factors, including the rare mutations and common variants. We built up this polygenic risk score with 172 common variants. Each person will receive a number to measure their risk of developing breast cancer in the general population. We are the first to apply such a score on survivors of childhood cancers.

How do you hope this information will be used moving forward?

What would be your take-home message for oncologists about this analysis?

We found that the survivors with the risk score in the top 20%—the highest quintile—had a 3-fold rate of subsequent breast cancer compared with the survivors with the polygenic risk score in the lowest quintile. Those are our main findings. Interestingly, we also observed that in the survivors younger than 45 years old, the polygenic risk score conferred risk to this subgroup. We have a very limited number of samples. Last year, we characterized 3000 samples, and now we are looking at female breast cancer, which is [1131] survivors. The estimates still need to be refined and further validated. We anticipate that clinically, considering these common low-risk variants together with the highly penetrant variation on top of these other risk factors, there can be a better strategy to inform future risk stratification for subsequent breast cancer among survivors of childhood cancer.Gene-wide association studies have been so successful, but people always have doubts over the clinical utility. When we get a bigger sample size, we will have a better picture of what is going on. We are at a time where we can make use of the data clinically combined with the rare mutations that explain a lot of heritable breast cancers, rather than sporadic.

Are there any next steps with this research?

The message to the clinician is that although we need more data to support a clinical recommendation, we are the first to provide some evidence to see the common variant effects. Science is progressing. We at least see promise in terms of changing practice clinically so that clinicians can better manage and find high-risk patients for screening and cancer prevention at early stages.The central life cohort is expanding, so we have at least 1000 or more survivors participating. We will sequence those patients, too. These 1000 will be added and will help us refine and validate. There is a multicenter cohort called Childhood Cancer Survivor Study, in which at least 30 different centers have participated nationwide. They already completed the whole-exome sequencing of 5700 patients, so we can put all the data together. That is one of the future directions that we are going to pursue.

In terms of the cohort, the median age is 36. In the general population, breast cancer won't occur that early. Continuously following up with this cohort and managing the changing risk of these patients is something we plan to do. We will continuously follow up with these people.

Is there is anything else that you would like to add?

Additionally, these analyses provide a working paradigm for studying the monogenic and polygenic mutations and their association with late effects, not just breast cancer. Survivors have a lot of disease burden. For example, there are monogenic and polygenic associations with obesity, and we can apply a similar model there. It is very exciting to present this information. Our institution has really committed and invested itself in these data. The data are available on the St. Jude Cloud, so we welcome outside collaborators to come in and work together to explore some interesting hypotheses.

Wang Z, Wilson CL, Easton J, et al. Monogenic and polygenic associations with subsequent breast cancer risk in survivors of childhood cancer: the St. Jude Lifetime Cohort Study (SJLIFE). Cancer Res. 2018;78(suppl 13; abstr 3007.). doi: 10.1158/1538-7445.AM2018-3007.