Op-jnci190078 1..12

Background: A total of 10%–20% of patients develop long-term toxicity following radiotherapy for prostate cancer. Identification of common genetic variants associated with susceptibility to radiotoxicity might improve risk prediction and inform functional mechanistic studies. Methods: We conducted an individual patient data meta-analysis of six genome-wide association studies (n1⁄43871) in men of European ancestry who underwent radiotherapy for prostate cancer. Radiotoxicities (increased urinary frequency, decreased urinary stream, hematuria, rectal bleeding) were graded prospectively. We used grouped relative risk models to test associations with approximately 6 million genotyped or imputed variants (time to first grade 2 or higher toxicity event). Variants with two-sided Pmeta less than 5 10 8 were considered statistically significant. Bayesian false discovery probability provided an additional measure of confidence. Statistically significant variants were evaluated in three Japanese cohorts (n1⁄4962). All statistical tests were two-sided. Results: Meta-analysis of the European ancestry cohorts identified three genomic signals: single nucleotide polymorphism rs17055178 with rectal bleeding (Pmeta 1⁄4 6.2 10 ), rs10969913 with decreased urinary stream (Pmeta 1⁄4 2.9 10 ), and rs11122573 with hematuria (Pmeta 1⁄4 1.8 10 ). Fine-scale mapping of these three regions was used to identify another independent signal (rs147121532) associated with hematuria (Pconditional 1⁄4 4.7 10 ). Credible causal variants at these four signals lie in gene-regulatory regions, some modulating expression of nearby genes. Previously identified variants showed consistent A R T IC LE Received: November 30, 2018; Revised: March 20, 2019; Accepted: April 29, 2019 © The Author(s) 2019. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 1 JNCI J Natl Cancer Inst (2020) 112(2): djz075 doi: 10.1093/jnci/djz075 First published online May 16, 2019 Article D ow naded rom http/academ ic.p.com /jnci/advance-articleoi/10.1093/jnci/djz075/5490201 by Intitute of C ild H elth/U niersity C olege Lndon user on 09 O cber 2019 associations (rs17599026 with increased urinary frequency, rs7720298 with decreased urinary stream, rs1801516 with overall toxicity) in new cohorts. rs10969913 and rs17599026 had similar effects in the photon-treated Japanese cohorts. Conclusions: This study increases the understanding of the architecture of common genetic variants affecting radiotoxicity, points to novel radio-pathogenic mechanisms, and develops risk models for testing in clinical studies. Further multinational radiogenomics studies in larger cohorts are worthwhile. Long-term side effects following radiotherapy affect the healthrelated quality of life for cancer survivors (1). Radiation dose and irradiated volume are the most important factors affecting toxicity risk but are broadly similar within patient populations. Known modifiers of the relationship between radiation exposure and risk of radiotoxicity include patient age, smoking history, concurrent treatments, and comorbidities. However, considerable interindividual variation in radiotoxicity remains unexplained after allowing for dosimetric and patient-level factors. An individual’s response to radiation is a heritable trait as evidenced by the similar cellular radiosensitivity of related individuals (2), intraindividual correlation in tissue response to therapeutic radiation (3), and the observation that rare mutations in some genes increase risk of radiotoxicity (4). Evidence suggests common variants may explain some of the remaining interindividual variation in radiotoxicity susceptibility (2,5). Simulation shows that the accuracy of models to predict radiotoxicity is improved when genetic risk variants are combined with clinical and dosimetric parameters (6). Genetic predisposition to radiotoxicity in nonsyndromic individuals is poorly understood. Preclinical studies suggest that the biologic mechanisms are complex, involving cell death, premature senescence, inflammation, tissue remodeling with development of fibrosis, and vascular damage. Large genetic studies are difficult because it is challenging to build cohorts associated with a phenotype that takes months to years to develop. In addition, the radiosensitivity phenotype varies by tumor site and means of assessment. In prostate cancer patients, common radiotoxicities include increased urinary frequency, radiation cystitis (characterized by urinary bleeding, pain, and inflammation), urinary retention, and rectal bleeding (7,8). Radiotoxicity prevalence in prostate cancer survivors varies by radiation delivery modality, time of follow-up, and method of symptom assessment (particularly for clinician vs patient scores). A large (n1⁄4 1571) prospective cohort study assessing clinician-assigned toxicity using the National Cancer Institute Common Terminology Criteria for Adverse Events reported the actuarial likelihood at 10 years of 15% and 9% for grade 2 or higher urinary and rectal toxicities, respectively (9). Despite approximately 50% of cancer patients undergoing radiotherapy, the collection of radiotoxicity data is sporadic and inconsistent. Most centers do not collect data routinely with the detail and standardization required to conduct radiogenomic studies. There is also a clear potential for clinical impact, for example, with an option for modified radiotherapy planning as increasingly conformal dose delivery techniques become available. The Radiogenomics Consortium was established to facilitate data sharing and develop methods for meta-analyses. Our first pooled genome-wide association study (GWAS) performed in prostate cancer survivors demonstrated that it was possible to meta-analyze data and identify genomic risk regions despite the heterogeneity inherent in radiotherapy cohorts (11). As part of a long-term goal to identify sufficient variants to develop a polygenic risk model for radiotoxicity, we undertook a larger meta-analysis on six cohorts comprising 3871 prostate cancer survivors. Secondary aims were to validate previously published single nucleotide polymorphisms (SNPs) and evaluate risk SNPs in Japanese cohorts. STROGAR guidelines (12) for reporting radiogenomic studies, which build on the STREGA and STROBE guidelines (13,14), were followed throughout.