Identification of Risk Factors and Comorbidities Associated with Treatment-Related Hearing Loss and Tinnitus and Comparison of Their Genetic Architecture with De Novo Etiologies

Although cisplatin has been in clinical use for over 40 years, it remains difficult to identify the subset of patients who may develop ototoxicity following therapy completion. Therefore, we sought to identify non-genetic and genetic risk factors for cisplatin-associated ototoxicity, and then evaluate whether these associations were shared with other etiologies. Through the use of pharmacokinetic modeling, we demonstrated that high levels of serum platinum were associated with multiple persistent cisplatin-induced toxicities, including tinnitus, and that patients with genetic variants in MYH14, a gene that affects kidney function, may be predisposed to retaining high levels of platinum for decades following completion of therapy. In addition, testicular cancer survivors who developed ototoxicity were more susceptible to persistent dizziness or vertigo, hypertension, and hypercholesterolemia, psychotropic drug use, and report poorer overall health. GWAS of cisplatin-induced hearing loss in testicular cancer survivors validated a previous association with rs62283056 (intronic to Mendelian deafness gene WFS1). Gene-based association analysis of cisplatin-induced hearing loss and tinnitus identified novel associations with TXNRD1 and WNT8A, respectively, genes relevant to renal and/or neuronal maintenance. We then evaluated an overall score of neurotoxicity burden in cisplatin-treated patients that was associated with numerous risk factors and comorbidities, including age at diagnosis or clinical examination, cumulative cisplatin dose, high serum platinum levels, tobacco use (ever smoker or current smoker), hypertension, persistent dizziness/vertigo, Raynaud Phenomenon, symptoms consistent with peripheral motor neuropathy, psychotropic drug use, and poorer overall health. Although GWAS identified no genome-wide significant SNPs, gene-based association analysis identified FAM20C in a gene-based association analysis, a gene that mediates cisplatin sensitivity in neurons, kidney, and bone, tissues vital for regulating cisplatin-induced toxicities. We also demonstrated that cisplatin-associated ototoxicity is associated with the same non-genetic risk factors and comorbidities as cranial radiation therapy (CRT)-associated ototoxicity and age-related hearing loss and tinnitus. GWAS of CRT-induced hearing loss and tinnitus implicated genetic variants involved in nervous system maintenance, particularly, ATXN1, a gene associated with the neurodegenerative disorder spinocerebellar ataxia type 1 that had its association with hearing loss validated in an independent replication cohort of pediatric cancer survivors. Interestingly, GWAS of age-related hearing loss and tinnitus also identified numerous genes vital to inner ear/nervous system function, as well as immune system regulation. Despite exhibiting similar phenotypic correlations, we were only able to detect shared genetic architecture between CRT-associated ototoxicity and age-related hearing loss in SNP-based enrichment analysis of GWAS results and polygenic risk score analysis. Taken together, these data indicate that elevated serum platinum levels may be of particular importance in identifying patients susceptible to developing multiple persistent toxicities following cisplatin-based chemotherapy, and that availability of circulating platinum is likely mediated by genetic variation in renal clearance and bone secretion. Further, we highlight the complexity of genetic predisposition to auditory disorders, indicating that similar phenotypes are not necessarily indicative of shared genetic architecture. Therefore, mechanistically based treatments for treatment-related ototoxicity will need to be validated across different etiologies.