Machine learning on genome-wide association studies to predict the risk of radiation-associated contralateral breast cancer in the WECARE Study.

Sangkyu Lee,Jung Hun Oh,John D Boice,Patrick Concannon,Joseph O Deasy,Xiaolin Liang,Jonine L Bernstein,Leslie Bernstein,Anne S Reiner,Charles F Lynch,Meghan Woods,Yonglan Zheng

doi:10.1371/journal.pone.0226157

Abstract

The purpose of this study was to identify germline single nucleotide polymorphisms (SNPs) that optimally predict radiation-associated contralateral breast cancer (RCBC) and to provide new biological insights into the carcinogenic process. Fifty-two women with contralateral breast cancer and 153 women with unilateral breast cancer were identified within the Women’s Environmental Cancer and Radiation Epidemiology (WECARE) Study who were at increased risk of RCBC because they were ≤ 40 years of age at first diagnosis of breast cancer and received a scatter radiation dose > 1 Gy to the contralateral breast. A previously reported algorithm, preconditioned random forest regression, was applied to predict the risk of developing RCBC. The resulting model produced an area under the curve (AUC) of 0.62 (p = 0.04) on hold-out validation data. The biological analysis identified the cyclic AMP-mediated signaling and Ephrin-A as significant biological correlates, which were previously shown to influence cell survival after radiation in an ATM-dependent manner. The key connected genes and proteins that are identified in this analysis were previously identified as relevant to breast cancer, radiation response, or both. In summary, machine learning/bioinformatics methods applied to genome-wide genotyping data have great potential to reveal plausible biological correlates associated with the risk of RCBC.

Highlights

Radiation-associated contralateral breast cancer (RCBC) is a rare adverse health outcome following radiation therapy for primary breast cancer
Estrogen receptor (ER) and progesterone receptor (PR) status was available with a missing rate of approximately 30%, but ERBB2 status was not available
Using Genome-wide association studies (GWAS) analysis combined with a non-linear machine learning approach, we created a prediction model of the risk of RCBC for young women who received a moderate level of radiation dose (> 1 Gy) to the contralateral breast

Summary

Introduction

Radiation-associated contralateral breast cancer (RCBC) is a rare adverse health outcome following radiation therapy for primary breast cancer. Young women at the time of exposure are at greater risk than older women [1,2,3,4]; the risk of RCBC has been found to be elevated in women 40 years of age, in those who received a scatter radiation dose > 1.0 Gy to a quadrant of the contralateral breast [5]. Genetic factors have been identified [16,17,18,19,20], suggesting that certain rare genetic variations in deoxyribonucleic acid (DNA) damage response genes, such as ATM, might make it difficult to repair the damage induced by radiation and eventually may lead to CBC. The candidate gene approach used in these studies has provided limited biological insights beyond the already known DNA damage response such as DNA repair or cell cycle arrest

Objectives

Methods

Results

Discussion

Conclusion