Genomic Analysis of Hearing Loss in Childhood Cancer Survivors

• Author / Creator

  Baedke, Jessica

• The development of efficacious anticancer therapies has significantly improved the prognosis of childhood cancer patients at the price of predisposing survivors to extensive chronic health problems. Although not fatal, reducing treatment-related hearing loss (ototoxicity) among childhood cancer survivors (CCSs) is a priority since it is the most common toxicity of mainstay cisplatin chemotherapy regimens for pediatric solid tumors and a frequent sequela of the cranial radiation therapy often used to treat pediatric brain tumors. Disconcertingly, it is now becoming apparent that ototoxic effects are long-lasting, with impairment adversely impacting language development, neurocognitive functioning, psychosocial skills, and school performance. It is therefore critical to identify high-risk patients for ototoxicity to provide, if possible, alternative cancer therapies, targeted risk-based interventions, and follow-up care. Interindividual ototoxic variability, however, is not adequately described by demographic and clinical variables in CCSs and has led to the hypothesis that genetic susceptibility underlies these diverse responses. To address this knowledge gap, we have conducted the first genome-wide association study (GWAS) among 5-year CCSs of European genetic ancestry to identify single nucleotide polymorphisms (SNPs) associated with ototoxicity.

  Genome-wide SNP genotypes and clinically-ascertained ototoxicity data were obtained from the St. Jude Lifetime Cohort Study (SJLIFE), which follows 5-year CCSs for their lifetime. To allow for possible detection of both general and treatment-specific genetic variants associated with ototoxicity, four quasi-stratified analyses were performed. Specifically, the four analyses targeted: 1) those doubly unexposed to cisplatin and cochlear radiation ≥ 20 Gy in the worst ear, 2) those exposed to cisplatin but not cochlear radiation, 3) those exposed to cochlear radiation but not cisplatin, and 4) the union of these three treatment subgroups. Adjusting for genetic ancestry principal components and the non-genetic risk factors pertinent to each analysis group, four separate logistic regression GWASes were performed. For each GWAS, 709,023 autosomal SNPs passing quality control were iteratively tested for an additive association with ototoxicity, which was defined as having/not having a score ≥ 2 on the International Society of Pediatric Oncology Ototoxicity Scale.

  While no SNP attained genome-wide statistical significance (Wald p-value < 5 x 10-8) for an association with ototoxicity, our methodology allowed us to screen for and identify biologically-supported SNPs with suggestive significance. Our most significant hits localized to a 114 kb region (p-value range = 1.9 x 10-7 – 9.6 x 10-6, odds ratio range = 2.8 – 4.8). Most of these SNPs were broadly relevant across all treatment profiles, which further modulated the strength of the genetic effect, whereas others were only relevant in the absence of cisplatin or cochlear radiation exposure. The plausibility of these nine SNPs was compelling, with molecular support from regulatory, epigenomic, and transcriptional perspectives and phenotypic support from the causal linkage of mutations in this region to progressive hearing loss in mice. Additionally, several distinct SNP signals known to reside in or influence the expression of genes implicated in sound transduction were identified.

  These findings, which warrant further study, suggest these loci may have clinical utility in identifying high-risk ototoxicity patients for the provision of personalized cancer treatments and follow-up care. A whole-genome sequencing analysis is underway and a replication analysis is planned.

• Subjects / Keywords

  • SNP
  • Childhood cancer survivors
  • Carboplatin
  • Cranial radiation
  • Audiology
  • GWAS
  • Ototoxicity
  • Childhood cancer
  • Cisplatin
  • Hearing loss
  • Radiation
  • Genetics
  • Cochlear radiation

• Graduation date

  Fall 2019

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-hw8b-tt40

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.