Translational Cancer Mechanisms and Therapy

Clinical and Genome-wide Analysis of Cisplatin-induced Tinnitus Implicates Novel OtotoxicMechanismsOmar El Charif1, Brandon Mapes1, Matthew R. Trendowski1, Heather E.Wheeler2,Claudia Wing1, Paul C. Dinh Jr3, Robert D. Frisina4, Darren R. Feldman5,Robert J. Hamilton6, David J. Vaughn7, Chunkit Fung8, Christian Kollmannsberger9,Taisei Mushiroda10, Michiaki Kubo10, Eric R.Gamazon11,12, Nancy J.Cox12, Robert Huddart13,Shirin Ardeshir-Rouhani-Fard3, Patrick Monahan3, Sophie D. Fossa14,Lawrence H. Einhorn3, Lois B. Travis3, and M. Eileen Dolan1

Abstract

Purpose: Cisplatin, a commonly used chemotherapeutic,results in tinnitus, the phantom perception of sound. Ourpurpose was to identify the clinical and genetic determinantsof tinnitus among testicular cancer survivors (TCS) followingcisplatin-based chemotherapy.

Experimental Design: TCS (n ¼ 762) were dichotomizedto cases (moderate/severe tinnitus; n ¼ 154) and controls(none; n ¼ 608). Logistic regression was used to evaluateassociations with comorbidities and SNP dosages ingenome-wide association study (GWAS) following qualitycontrol and imputation (covariates: age, noise exposure,cisplatin dose, genetic principal components). Pathwayover-representation tests and functional studies in mouseauditory cells were performed.

Results: Cisplatin-induced tinnitus (CisIT) significantlyassociated with age at diagnosis (P ¼ 0.007) and cumulativecisplatin dose (P ¼ 0.007). CisIT prevalence was not signifi-cantly greater in 400 mg/m2-treated TCS compared with 300

(P ¼ 0.41), but doses >400 mg/m2 (median 580, range402–828) increased risk by 2.61-fold (P < 0.0001). CisITcases had worse hearing at each frequency (0.25–12 kHz, P <0.0001), and reported more vertigo (OR¼ 6.47; P < 0.0001)and problems hearing in a crowd (OR ¼ 8.22; P < 0.0001)than controls. Cases reported poorer health (P < 0.0001)and greater psychotropic medication use (OR ¼ 2.4; P ¼0.003). GWAS suggested a variant near OTOS (rs7606353,P ¼ 2 � 10�6) and OTOS eQTLs were significantly enrichedindependently of that SNP (P ¼ 0.018). OTOS overexpres-sion in HEI-OC1, a mouse auditory cell line, resulted inresistance to cisplatin-induced cytotoxicity. Pathway analy-sis implicated potassium ion transport (q ¼ 0.007).

Conclusions: CisIT associated with several neuro-otological symptoms, increased use of psychotropic medica-tion, and poorer health.OTOS, expressed in the cochlear lateralwall, was implicated as protective. Future studies should inves-tigate otoprotective targets in supporting cochlear cells.

IntroductionApproximately 9.6% of the U.S. population experiences tinni-

tus, the phantom perception of sound characterized by persistentringing, buzzing, beeping, or hissing in the ears (1). Severalstudies have linked tinnitus to increased anxiety, depression,insomnia, and reduced productivity and quality of life (QoL;refs. 2, 3). Nevertheless, the pathogenic mechanisms of thisdisorder remain poorly understood (4), and no pharmacologic

agents are approved to treat tinnitus (5); treatment is typicallylimited to cognitive behavioral therapy and management ofassociated anxiety (5, 6). Despite its etiologic ambiguity, severalrisk factors have been shown to contribute to the development oftinnitus, including age, hearing loss, and the administration ofototoxic drugs (7, 8).

Cisplatin is one of the most commonly used chemotherapeu-tics, and is known for its potent ototoxic effects. It can lead to

1Department of Medicine, University of Chicago, Chicago, Illinois. 2Departmentsof Biology and Computer Science, Loyola University Chicago, Chicago, Illinois.3Department of Medical Oncology, Indiana University, Indianapolis, Indiana.4Departments of Medical and Chemical & Biomolecular Engineering and Com-munication Sciences & Disorders, Global Center for Hearing and SpeechResearch, University of South Florida, Tampa, Florida. 5Department of MedicalOncology, Memorial Sloan-Kettering Cancer Center, New York, New York.6Department of Surgical Oncology, Princess Margaret Cancer Centre, Toronto,Ontario, Canada. 7Department of Medicine, University of Pennsylvania, Phila-delphia, Pennsylvania. 8J.P. Wilmot Cancer Institute, University of RochesterMedical Center, Rochester, New York. 9Division of Medical Oncology, Universityof British Columbia, Vancouver, British Columbia, Canada. 10RIKEN Center forIntegrative Medical Science, Yokohama, Japan. 11Clare Hall, University of Cam-bridge, Cambridge, United Kingdom. 12Vanderbilt University Medical Center,

Nashville, Tennessee. 13RoyalMarsdenHospital, London, UnitedKingdom. 14Depart-ment of Oncology, Oslo University Hospital, Radiumhospital, Oslo, Norway.

O. El Charif and B. Mapes contributed equally to this article.

Corresponding Authors: M. Eileen Dolan, University of Chicago, 900 E 57th St.,Rm 7100, Chicago, IL 60637-1470. Phone: 773-702-4441; Fax: 773-702-9268;E-mail: edolan@medicine.bsd.uchicago.edu; and Lois B. Travis, Indiana Univer-sity Melvin and Bren Simon Cancer Center, 535 Barnhill Drive, RT433, Indiana-polis, IN 46202. Phone: 317-274-4875; E-mail: lbtravis@iu.edu

Clin Cancer Res 2019;25:4104–16

doi: 10.1158/1078-0432.CCR-18-3179

�2019 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 25(13) July 1, 20194104

irreversible bilateral hearing loss in up to 80% of patients, withmany experiencing permanent tinnitus (9–11). Although fewstudies have rigorously assessed tinnitus following cisplatin-based chemotherapy, approximately 40% of patients appear toexperience at least some degree of tinnitus (6). This is substan-tially higher than the rate observed in the general U.S. population,and in patients with testicular cancer who did not receive cisplat-in-based chemotherapy (12%; ref. 9). Severe cases of tinnitus arealso markedly increased in cisplatin-treated patients when com-pared with the general population (22% vs. 1%–2%; refs. 4, 9).One study of 1,409 testicular cancer survivors (TCS) noted that19% of TCS treated with 100–400 mg/m2 cisplatin exhibitedsymptoms of moderate/severe tinnitus, with that prevalencereaching 42% in the dose-intensive cisplatin chemotherapygroup (12), suggesting a dose-dependent effect. It has beenpreviously demonstrated that cancer survivors with neuropathy,hearing loss, and tinnitusweremore likely to report poorer qualityof life (QoL) than those with neuropathy alone (13). Anotherinvestigation confirmed worse perceived stress among cancersurvivors with tinnitus (14).

Despite this documented effect on QoL, little is known aboutthe risk factors and mechanisms of cisplatin-induced tinnitus(CisIT). The subjective nature of the disorder and lack of modelsystems hamper the elucidation of its pathophysiology. Whilegenotype–phenotype associations promise to identify underlyingmolecular alterations, to our knowledge, no genome-wide asso-ciation study (GWAS) of CisIT has yet been conducted. Wepreviously characterized ototoxicity in our North Americancohort of TCS receiving homogeneous cisplatin-based treatment.Among 488 TCS, 40% experienced some degree of tinnitus and15.8% reported moderate/severe tinnitus. Furthermore, tinnituswas significantly correlated with reduced hearing at each frequen-cy examined (P < 0.001; ref. 9). In this study, we comprehensivelyevaluate the role of clinical characteristics and modifiable riskfactors on the development of tinnitus after cisplatin-basedchemotherapy in a cohort of 762 TCS, and identify geneticdeterminants of CisIT through GWAS. In addition, we conduct

pathway analyses and in vitro experiments to validate and con-textualize our results.

Materials and MethodsPatients and study design

All patients were enrolled in the Platinum Study, whichincludes eight cancer centers in the United States and Canada.Eligibility criteria included: men diagnosed with histologically orserologically confirmed germ cell tumor (GCT), age <55 years atdiagnosis and >18 years at enrollment, treatment with first-linecisplatin-based chemotherapy regimen for either initial GCT orrecurrence after active surveillance, with no antecedent chemo-therapy for another primary cancer, no salvage chemotherapy,and routine follow-up at the participating site. All patients weredisease-free at the time of clinical evaluation andprovidedwrittenconsent for study participation, access to medical records, andgenotyping. Study procedures were approved by the HumanSubjects Review Board at each institution. The studies were con-ducted in accordance with the U.S. Common Rule.

AssessmentsPatient data were either ascertained at clinical follow-up or

abstracted from medical records according to a standardizedprotocol (Supplementary Fig. S1). Data abstracted from medicalcharts included: age at GCT diagnosis, GCT tumor characteristics,treatment regimen with cumulative dose and number of cycles,and BMI at the initiation of treatment. Data collected at clinicalevaluation included age and BMI, audiometric studies, self-reported questionnaires, and genotyping. Pure-tone air conduc-tion thresholds were measured bilaterally on a subset of 602survivors at the frequency range 0.25–12 kHz to quantify hearing.The geometric mean of hearing thresholds across cisplatin-affected frequencies (4–12 kHz) was used to quantify cisplatin-induced hearing loss as described previously (9). Self-reportedototoxicity and neurotoxicity were assessed with two validatedquestionnaires: the EORTC-Chemotherapy Induced PeripheralNeuropathy 20 item questionnaire (CIPN20) and the Scale forChemotherapy-Induced Neurotoxicity (SCIN; ref. 15).

Hypertension, hypercholesterolemia, use of psychotropicmed-ication (antidepressants, anxiolytics, and antipsychotics), overallhealth, tobacco use, alcohol consumption, and noise exposurewere also ascertained at follow-up as patient-reported outcomes.Hypertensive subjects indicated receiving a diagnosis of highblood pressure and taking prescription medication for it atevaluation. Hypercholesterolemia subjects answered "yes, cur-rent" to the question: "Have you ever taken prescription medi-cation for high cholesterol?" The use of psychotropic medicationwas assessed as in Fung and colleagues (16), and was based onwhether a patient had been taking psychotropic medication for atleast one month at the time of clinical evaluation. Self-reportedhealth was rated on a poor–excellent scale in response to thequestion "Howwould you rate your health?" Values were numer-ically converted for association testing (poor/fair ¼ 1, good ¼ 2,very good ¼ 3, excellent ¼ 4). Exposure to noise was assessedusing two questions: "Have you ever had a job where you wereexposed to loud noise for 5 or more hours a week? (loud noisemeans noise so loud that you had to speak in a raised voice to beheard)" and "Outside of a job, have you ever been exposed tosteady loud noise ormusic for 5 ormore hours a week? (examplesare noise from power tools, lawn mowers, farm machinery, cars,

Translational Relevance

Cisplatin is one of the most commonly prescribed che-motherapeutics worldwide, but elicits several debilitatingoff-target toxicities, including cisplatin-induced tinnitus(CisIT; a perceived ringing/buzzing noise with no externalsound present). CisIT significantly affects perceived quality oflife, but its etiology remains poorly understood.We found thatcumulative cisplatin doses exceeding 400 mg/m2 to markedlyincrease CisIT susceptibility, and that survivors with tinnituswere more likely to experience hearing loss, vertigo, andproblems hearing in a crowd, as well as report poorer healthand greater reliance on psychotropic medication. Through agenome-wide association study, we identified genetic variantsin OTOS, a gene vital for the maintenance of normal hearing,to be associated with CisIT, and further demonstrated thatincreased OTOS expression was associated with reduced cis-platin sensitivity in silico and in vitro. These data provide novelmechanistic insights into CisIT susceptibility, andmay informclinicians of potential risk factors and comorbidities.

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4105

trucks, motorcycles, or loud music)". After numeric conversion(Yes ¼ 1, No ¼ 0), overall noise exposure was computed as thesum of noise exposure in and outside of work. Alcohol consump-tion was assessed as the self-reported response to the question"During the past year, how many drinks of alcoholic beveragehave you consumed on average? [1 drink ¼ 12 oz. beer (1 can orbottle), 4 oz. glass of wine, 1 mixed drink or shot of liquor]" withthe following options: rarely/never (0), 1–3/month (1), 1/week(2), 2–4/week (3), 5–6/week (4), 1/day (5), 2–3/day (6), 4–5/day(7), and 6þ/day (8). Tobacco use was assessed as the response tothe following twoquestions: "Have youEVER smoked cigarettes?"with "Yes" (1) and "No" (0) options and "Do you CURRENTLYsmoke cigarettes?"

GenotypingGenotyping was done as described previously (17, 18). DNA

was extracted from peripheral blood mononuclear cells. SNPswere genotyped in two phases at the RIKEN Center for integrativeMedical Science (Yokohama, Japan). Set 1 was genotyped on theHumanOmniExpressExome-8v1-2_A chip (Illumina) and set 2was genotyped on the InfiniumOmniExpressExome-8v1-3_Achip (Illumina). Samples were plated with inter- and intra-plateduplicates at random. Standard quality control (QC) was imple-mented in PLINK, and is outlined in Supplementary Fig. S2.Sample-level QC criteria included: sample call rate > 0.99, pair-wise identity by descent < 0.125, coefficient of inbreeding F < 6SDs from the mean, and genetically European as determined byprincipal component analysis (performed using SMARTPCA).SNP-level QC included: call rate > 0.99, and Hardy–Weinbergequilibrium (x2 P > 1 � 10�6). QC was first performed in eachset independently and then the two sets were merged followedby additional QC. QC for batch effects was performed usinga batch pseudo-GWAS where variants associated with batchpseudo-phenotype (10 genotypic PC-adjusted P < 10�4) wereexcluded. SNPs and samples passing QC criteria comprised theinput set for imputation with EAGLE phasing, performed on theMichigan Imputation Server using the Haplotype Reference Con-sortium (18–20). SNPswith imputationR2<0.8,MAF <0.01, andINFO scores > 1.05 or < 0.3 were excluded. Only subjects whopassed QC were included in this study (Supplementary Fig. S2).

Case–control definitionUsing SCIN (15), TCS were dichotomized to tinnitus cases/

controls based on responses to the question: "Have you had in thelast 4 weeks: Ringing or buzzing in the ears?" Cases responded"quite a bit/very much", and controls responded "not at all".Those answering "a little" were not included. We labeled "not atall", "a little", and "quite a bit/very much" as none, mild, andmoderate/severe tinnitus. TCS were also asked: "Do you have:ringing and buzzing in the ears?" Tinnitus cases responding "no"to this question were excluded from analysis.

Genome-wide analysesGWASwas done in PLINK v1.9 (22, 23) with logistic regression

assuming additive effects, with cumulative cisplatin dose, age atdiagnosis, self-reported noise exposure, and the first 5 geneticprincipal components (SMARTPCA (23)) as covariates. We per-formed imputation of missing covariates to avoid sample exclu-sion. Eight individuals with missing age at diagnosis wereassigned an imputed age calculated as age at clinical follow-upminusmedian follow-up duration (5.2 years). Twelve individuals

were missing values for cumulative cisplatin dose received,but each of these patients had information on the cycles ofcisplatin received. Because one cycle of cisplatin is equivalent to100 mg/m2, these patients were assigned an imputed cumulativedose by multiplying the number of cycles they received by 100(i.e., 3 cycles� 100¼ 300mg/m2). No subjects weremissing dataon noise exposure. Genome-wide significance was set to P < 5 �10�8. A list of expression quantitative trait loci (eQTL) wasextracted from GTEx v7 using all 48 tissues with eQTL analy-sis (25). Conditional analysis was performed by adjusting for thelead single nucleotide polymorphism (SNP) as a covariate. SNPfamily enrichment was assessed using the empirical Brown com-bined P value (26). Narrow-sense heritability was estimated witha genetic relationship matrix variance component model asimplemented by GCTA with the same covariates as GWAS (27).Pathway over-representation analysis was performed using GeneOntology (28, 29) pathways with the g:Profiler package (30) at asignificance threshold of FDR q < 0.05 after attributing genes theP value of the most significantly associated SNP in or within 25kilobases of gene start/end sites (GENCODE GRCh37.p12).

Gene expression and toxicity in cell linesOTOS expressiondata indifferent cancer cell lineswas obtained

from the Cancer Cell Line Encyclopedia (CCLE; ref. 31). Drugsensitivity to cisplatin and other antineoplastic agents, measuredas the area under the dose–response curve, was obtained from theGenomics of Drug Sensitivity in Cancer Project (32). Spearmancorrelation and linear regression was performed between expres-sion and sensitivity of cancer cell lines with nonmissing OTOSexpression in R 3.3.2.

Statistical analysisData are presented with medians (ranges). Age was modeled

continuously and in quartiles (ordinal). Cisplatin dose was trea-ted continuously and in dose categories: <300, 300, 400, and>400 mg/m2. Prevalence ratios (PR) were calculated to comparetinnitus prevalence by treatment or allele group, with normalapproximation for 95% confidence intervals (CI) and x2 P valuesreported. Logistic models were constructed to assess associations(OR) with tinnitus in univariable and multivariable contexts. Alltests were two-sided at a significance threshold of P < 0.05. Resultsare presented as OR/PR (95% CI). All statistical analyses wereperformed in R 3.3.2 and plotted with ggplot2 (33) unlessotherwise specified.

Cell cultureHEI-OC1, a mouse auditory cell line derived from the inner

ear organ of Corti, was cultured in complete media of DMEM(high glucose, L-glutamine and pyruvate) with 10% FBS (Corn-ing Inc.). L929, a mouse fibroblast cell line, was cultured incomplete media of EMEM with 10% FBS and 1% penicillin/streptomycin. Cells were passaged three times/week, plated at1 � 106 cells/T25 culture flask, and grown in a humidifiedchamber. HEI-OC1 was grown at 33�C with 10% CO2 and L929at 37�C with 5% CO2. The authenticity of the cell lines wasconfirmed by IDEXX BioResearch by measurement for inter-species contamination or misidentification, Case # 26495-2018(HEI-OC1) and #32923-2018 (L929). Cell lines were alsoassessed for Mycoplasma contamination using a MycoAlertMycoplasma Detection Kit (Lonza) at the time of acquisitionand prior to performing in vitro experiments.

El Charif et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4106

Plasmid preparationOTOS protein vector, pOTOS (pPM-N-D-C-His; catalog no.

PV401573) and blank control (pControl; PV001) vector werepurchased from Applied Biological Materials, Inc., with the blankcontrol vector sequence being verified by the manufacturer. Theplasmids were amplified into competent DH5a bacteria (LifeTechnologies Corporation), expanded using 50 mg/mL kanamy-cin selection, and isolated using PureYield Plasmid Midiprep(Promega Corporation). Confirmation of OTOS cDNA sequencewas performed at the University of Chicago DNA SequencingCore.

Cell viability assayAfter 24 hours of plating each cell line at 31,250 cells/cm2 into

96-well plates, the cells were transfected in Opti-MEM (LifeTechnologies) with the addition of a complexed DNA mixtureconsisting of 1 part Lipofectamine 3000 (Thermo FisherScientific) with 3 mg pOTOS or pControl, as per manufacturer'sinstructions. Transfection continued for 24 hours. Cisplatin(Sigma-Aldrich) was dissolved in a darkened hood using 0.9%saline. The solution was vortexed for 1 minute, sonicated for 10minutes with no heat, and then filtered through a 0.2-mm syringefilter to obtain a 10 mmol/L stock solution. Transfection mediawas exchanged for 5, 7.5, 10, 25, or 50 mmol/L cisplatin incomplete media or vehicle control. At 48 or 72 hours postdrugtreatment, 100 mL CellTiter-Glo (Promega Corporation) wasadded to each well, as per manufacturer's instructions. A total of135 mL of each well was transferred to a white Costar assayplate (Thermo Fisher Scientific) and luminescence was measuredon a Synergy-HT plate reader (BioTek Corporation). Four inde-pendent experiments were performed and within an experiment,each concentration was evaluated in triplicate. IC50 valueswere calculated through nonlinear regression of the log cisplatinconcentration versus normalized response curve using Prism6software.

Quantitative PCRCells were plated at 31,250 cells/cm2 into 12-well plates for

qPCR collections. After 24, 48, 72, and 96 hours of transfection(corresponding to 0, 24, 48, and 72 hours after cisplatin treat-ment), cells were pelleted following a brief trypsinization andstored at �80�C. RNA was isolated using the RNeasy Plus Kit(Qiagen) followed by reverse transcription of 200 ng RNA intocDNA with the High Capacity cDNA Reverse Transcription Kit(Thermo Fisher Scientific). Mouse Otos (Mm01292235_g1)expression was tested in HEI-OC1 and L929 cells, and wasfound to be minimally expressed. Expression of humanOTOS was determined with qPCR using TaqMan primerOTOS (Hs00964785_m1) and compared with mouse Actb(Mn00607939_s1), the endogenous control. Comparative DDCt

method was used to determine fold change of human OTOS ateach timepoint. Four independent experiments were performedand each sample was run in triplicate on a Life Technologies Viia7PCR machine.

ResultsCohort characteristics

Of 1,029 TCS who passed GWAS QC criteria, 608 (59.1%),265 (25.8%), and 154 (15.0%) reported none, mild, and mod-erate/severe tinnitus, respectively. Subjects were dichotomized to

cases (moderate/severe) and controls (none). Two cases wereexcluded for inconsistent responses. Cohort characteristics areprovided in Table 1. Median age at diagnosis and evaluation was31 (15–54) years and 39 (18–75) years, respectively. Medianfollow-up timewas 5.2 (1–37) years. A total of 91.6%were treatedwith either BEP (n ¼ 444, 58.3%) or EP (n ¼ 254, 33.3%). Theremaining 64 (8.4%) received vinblastine (n ¼ 6), carboplatin(n¼ 5), or ifosfamide (n ¼ 45) in combination with cisplatin, orhad missing cisplatin dose data (n ¼ 12).

Diagnosis and treatment risk factorsCumulative cisplatin dose was significantly associated with

tinnitus [OR per 100 mg/m2 ¼ 1.38; 95% confidence interval(CI): 1.1–1.7; P ¼ 0.007]. Tinnitus risk for 400 mg/m2-treatedTCS (n ¼ 358) was not significantly greater than those receiving300 mg/m2 (n ¼ 300; OR ¼ 1.14; 95% CI: 0.8–1.6; P ¼ 0.41).However, survivors who received >400 mg/m2 (n ¼ 30, mediandose ¼ 580 mg/m2, range ¼ 402–828 mg/m2) were at 2.61(95%CI: 1.8–3.9)-fold increased risk compared with those receiving�400 mg/m2 (P < 0.0001, Fig. 1A). Tinnitus was not associatedwith cumulative doses of bleomycin (OR ¼ 1.01, P ¼ 0.13),etoposide (OR ¼ 1.02, P ¼ 0.33), or ifosfamide (OR ¼ 1.06,P ¼ 0.87) when cumulative cisplatin dose was included as acovariate. Three of five TCS receiving both carboplatin (mediandose ¼ 1,800 mg/m2, range ¼ 5–3,646 mg/m2) and cisplatin(median dose ¼ 200 mg/m2, range ¼ 200–527 mg/m2) hadtinnitus (OR ¼ 7.37; 95% CI: 1.2–58.5, P ¼ 0.034). Tinnitus wasassociated with increasing age at diagnosis as a continuousvariable (OR by decade ¼ 1.31; 95% CI: 1.1–1.6, P ¼ 0.006) andby quartile (OR ¼ 1.22, P ¼ 0.013; Fig. 1B). Age at evaluationwas also associated with tinnitus (OR ¼ 1.30; 95% CI: 1.1–1.5;P¼ 0.002). No association with follow-up durationwas observed(age-adjusted OR ¼ 0.99; P ¼ 0.57).

Associations with self-reported health and neuro-ototoxicsymptoms

Self-reported overall health was significantly lower in casesthan in controls (OR ¼ 0.54; 95% CI: 0.4–0.7; P <0.0001, Fig. 2A). Tinnitus cases also reported using psychotropicmedications more frequently than controls (OR ¼ 2.40; 95% CI:1.3–4.4; P ¼ 0.003, Fig. 2B). Audiometry was performed on 602(78.4%) subjects. TCS with tinnitus had significantly worse hear-ing at every frequency examined (0.25 kHz–12 kHz, P <0.0001; Fig. 3A). Tinnitus was also associated with self-reportedhearing loss (OR ¼ 6.38; 95% CI: 4.9–8.5; P < 0.0001, Fig. 3B),problems hearing in a crowd (OR ¼ 8.28; 95% CI: 5.5–12.5; P <0.0001, Fig. 3C), and persistent dizziness or vertigo (OR ¼ 6.40;95% CI: 3.2–12.9, P < 0.0001; Fig. 3D), as well as symptoms ofsensory neuropathy (OR range 1.66–2.72; P <0.0001, Fig. 3E).

Associations with modifiable risk factorsTinnitus was significantly associated with hypertension

(age-adjusted OR ¼ 2.07; 95% CI: 1.3–3.4; P ¼ 0.003), andhypercholesterolemia (age-adjusted OR ¼ 1.96; 95% CI: 1.2–3.2; P¼ 0.009). In an age at clinical examination-adjusted modelwith both variables, tinnitus was significantly associated withhypertension (OR¼ 1.77; 95% CI: 1.02–3.0; P¼ 0.039), but nothypercholesterolemia (OR ¼ 1.6, 95% CI: 0.9–2.8, P ¼ 0.09).Tinnituswas associatedwithnoise exposure atwork (age-adjustedOR ¼ 1.93; 95% CI: 1.3–2.8; P ¼ 0.0005), and outside of work(age-adjusted OR ¼ 2.28; 95% CI: 1.6–3.3; P < 0.0001). Risk was

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4107

also associated with chronic (>15 years) smoking (age-adjustedOR¼ 2.07; 95%CI: 1.2–3.8; P¼ 0.005), but not increased amongever smokers (OR ¼ 1.26; 95% CI: 0.9–1.7; P ¼ 0.20) or withalcohol consumption (OR ¼ 0.92; 95% CI: 0.5–1.5; P ¼ 0.12).Association with chronic smoking remained significant afteradjusting for age, hypertension, and noise exposure (OR ¼1.79; 95% CI: 1.01–1.04; P ¼ 0.04).

Genome-wide association studyUsing GCTA's linear mixed model approach (27), we found

that additive SNP effects explained a large fraction of CisITvariance (h2 ¼ 0.81 � 0.42; P ¼ 0.006). In GWAS, no SNP metgenome-wide significance (Supplementary Table S1; Supplemen-tary Fig. S3). The first independent signal identified (rs7532231,OR ¼ 0.45, P ¼ 1.1 � 10�6) is 116 kilobases upstream of thelncRNA RP11–364B6.1. The second (rs6671895, OR ¼ 4.32, P ¼1.2 � 10�6) is intronic to the lncRNA RP5–884C9.2. The third(rs7606353, P ¼ 1.9 � 10�6, Fig. 4A) is a SNP 14 kilobasesdownstreamofOTOS (otospiralin), a gene implicated in auditoryfunction and survival of cochlear fibrocytes (34, 35). This variantis in near perfect linkage disequilibrium with rs74002277 inEuropeans (CEU R2¼ 0.96), a SNP in the 30-UTR region ofOTOSand which alters several transcription binding motifs (35). Theminor allele (G, 4% frequency) increased CisIT risk (OR ¼ 4.20;95% CI: 2.3–7.5). Because there has been prior evidence thatOTOS protects cochlear fibrocytes against cisplatin cytotoxici-ty (37), we hypothesized that variants associated with higherOTOS expression lowered CisIT risk. In GTEx, OTOS was signif-icantly expressed only in pituitary and thyroid tissues (Supple-mentary Fig. S4). We found that OTOS eQTLs were significantlyenriched in GWAS independently from rs7606353 (P ¼ 0.018).One haplotype block containing six thyroid/pituitary eQTLs wasdriving this enrichment (Supplementary Table S2; SupplementaryFig. S4). Theminor allele (A) of themost significant variantwithinthis block (rs10190781, 1.5% frequency) was associated withhigher CisIT risk (OR¼ 3.42; 95%CI: 1.5–9.5; P¼ 0.007; Fig. 4B)and lower OTOS expression (Fig. 4C), suggesting that higherOTOS expression is protective. Tinnitus risk increased with thenumber of risk alleles in either locus (OR¼ 3.77; 95%CI: 2.3–6.2;P ¼ 9.5 � 10�8; Fig. 4D).

Functional analysisWe tested the association between OTOS expression and cel-

lular sensitivity to cisplatin (as measured by area under the dose–response curve) in central nervous system (CNS) tumorlines (30, 31) and found a significant positive correlation (Spear-man r¼ 0.46, P ¼ 0.03; R2 ¼ 0.25, P ¼ 0.01; Fig. 4E), furthersupporting a protective OTOS function against cisplatin-induceddamage. To examine the specificity of this correlation, we com-pared the sensitivity of CNS tumor cell lines to eight antineo-plastic agents [5-fluorouracil (5-FU), bleomycin, bortezomib,cisplatin, cytarabine, docetaxel, etoposide, and vinblastine] basedon OTOS expression levels (Supplementary Table S3). We alsoexamined the relationship between cisplatin sensitivity andOTOSexpression in seven cancer cell line types (aerodigestive, breast,CNS, digestive system, lung, skin, and urogenital system; Sup-plementary Table S4). Of the cytotoxic drugs tested, only cisplatinsensitivity demonstrated a statistically significant correlation withOTOS expression in CNS lines. Furthermore, cisplatin sensitivitywas associated with OTOS expression only in CNS tumor linesand not cell lines of other tissue origins. These data suggest a

Table 1. Demographic features, clinical characteristics, and patient-reportedoutcomes of 762 male germ cell tumor survivors according to tinnitus status

CharacteristicsAll participants(n ¼ 762)

Tinnitus controls(n ¼ 608)

Tinnitus casesa

(n ¼ 154)

Age at diagnosis (years)b

Median (range) 31 (15–54) 30 (15–54) 35 (16–52)<20 43 (5.7%) 35 (5.8%) 8 (5.2%)20–29 300 (39.8%) 249 (41.5%) 51 (33.1%)30–39 243 (32.2%) 196 (32.7%) 47 (30.5%)40–55 168 (22.2%) 120 (20.0%) 48 (31.2%)

Age at assessment (years)Median (range) 39 (18–75) 38 (18–71) 43 (20–75)<20 6 (0.8%) 5 (0.8%) 1 (0.6%)20–29 134 (17.6%) 115 (18.9%) 19 (12.3%)30–49 274 (36.0%) 230 (37.8%) 44 (28.6%)40–49 202 (26.5%) 150 (24.7%) 52 (33.8%)50–59 123 (16.1%) 88 (14.5%) 35 (22.7%)�60 23 (3.0%) 20 (3.3%) 3 (2.0%)

Time since therapy completion (years)c

Median (range) 5.2 (1–37) 5.1 (1–37) 6.2 (1–35)�1 3 (0.4%) 1 (0.2%) 2 (1.3%)>1 and �5 362 (48.0%) 293 (48.8%) 69 (44.8%)>5 and �10 180 (23.9%) 145 (24.2%) 35 (22.7%)>10 and �20 159 (21.1%) 127 (21.2%) 32 (20.8%)>20 50 (6.6%) 34 (5.7%) 16 (10.4%)

Educational levelSecondary schoolor less

95 (12.6%) 62 (10.3%) 33 (21.9%)

Post-secondaryschool training

166 (22.0%) 129 (21.4%) 37 (24.5%)

College graduate 334 (44.2%) 276 (45.7%) 58 (38.4%)Post graduatetraining/degree

160 (21.2%) 137 (22.7%) 23 (15.2%)

Chemotherapyregimend

BEP 444 (58.3%) 355 (58.4%) 89 (57.8%)EP 254 (33.3%) 205 (33.7) 49 (31.8%)Other 64 (8.4%) 48 (7.9%) 16 (10.4%)

Number of cycles, platinum-basedchemotherapye

�2 15 (2.0%) 13 (2.2%) 2 (1.3%)3 333 (44.0%) 272 (45.1%) 61 (39.6%)4 389 (51.4%) 307 (50.9%) 82 (53.2%)>4 20 (2.6%) 11 (1.8%) 9 (5.8%)

Cumulative cisplatin dose (mg/m2)f

Median 400 (100–828) 400 (200–800) 400 (100–827.6)<300 41 (5.5%) 32 (5.4%) 9 (5.9%)300 300 (40.0%) 248 (41.5%) 52 (34.0%)>300 and <400 21 (2.8%) 15 (2.5%) 6 (3.9%)400 358 (47.7%) 287 (48.1%) 71 (46.4%)>400 30 (4.0%) 15 (2.5%) 15 (9.8%)

Reduced hearing in past four weeksg

Not at all 555 (73.1%) 521 (85.7%) 34 (22.5%)A little 127 (16.7%) 71 (11.7%) 56 (37.1%)Quite a bit/verymuch

77 (10.1%) 16 (2.6%) 61 (40.4%)

Problems hearing incrowdsh

Yes 222 (30.6%) 124 (21.2%) 98 (69.0%)No 505 (69.4%) 461 (78.8%) 44 (31.0%)

Requires hearingaidi

Yes 2 (0.3%) 2 (0.3%) 0 (0%)No 741 (99.7%) 602 (99.7%) 139 (100%)

Noise exposureNone 430 (56.4%) 365 (60.0%) 65 (42.2%)Work related ORother exposure

213 (28.0%) 163 (26.8%) 50 (32.5%)

Work relatedAND otherexposure

119 (15.6%) 80 (13.2%) 39 (25.3%)

(Continued on the following page)

El Charif et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4108

CNS tumor–specific association between OTOS expression andcisplatin sensitivity.

To functionally validate the role of OTOS gene expression incellular sensitivity to cisplatin, we overexpressedOTOS in two celllines, HEI-OC1 (a mouse auditory cell line) and L929 (a mousefibroblast cell line), and compared gene expression levelsand cellular proliferation following cisplatin treatment inOTOS-transfected cells and plasmid-transfected control cells(Fig. 5A). Overexpression of human OTOS in HEI-OC1 resultedin increased cell viability following cisplatin treatment at 5, 7.5,10, and 25 mmol/L concentrations for 48 hours (P ¼0.004; Fig. 5B) and 72 hours (P¼ 0.008; Fig. 5C). SupplementaryFigure S5 illustrates the effect of cisplatin on HEI-OC1 cellstransfectedwith humanOTOS andplasmid control. Furthermore,the cisplatin IC50 value of HEI-OC1 cells transfected with humanOTOS (18.68mmol/L, 95%CI: 13.95–25.01mmol/L)was 1.8-foldhigher than cells transfected with the control plasmid (10.66mmol/L, 95% CI: 6.04–18.82 mmol/L) at 48 hours. Overexpres-sion of OTOS was confirmed via qPCR, with human OTOSexpression increasing 69-fold at the time of drug treatment, 24hours following pOTOS transfection (Fig. 5C, inset). Similarly,overexpression of OTOS in L929 fibroblasts also markedlyreduced cellular sensitivity to cisplatin following treatment for48 hours (P ¼ 0.006; Fig. 5D) and 72 hours (P ¼ 0.03; Fig. 5E)concomitant with increased expression of OTOS (Fig. 5E, inset).The cisplatin IC50 value inOTOS-transfected cells (16.30 mmol/L,95% CI: 13.25–20.04 mmol/L) was 1.5-fold higher than cellstransfected with the control plasmid (10.96 mmol/L, 95% CI:8.82–13.61 mmol/L) at 48 hours.

Pathway analysisWe mapped SNPs to genes by proximity and performed path-

way enrichment with Gene Ontology terms. We identified 134coding genes with P < 5 � 10�4 (Supplementary Table S5). Tenpathways were significantly over-represented (SupplementaryTable S6), including nervous system process (q ¼ 0.005), potas-sium ion transmembrane transport (q ¼ 0.007), and sensoryperception of sound (q ¼ 0.005). Potassium transport moststrongly implicated KCNQ1 (rs2283200, P ¼ 8.9 � 10�6), which

Table 1. Demographic features, clinical characteristics, and patient-reportedoutcomes of 762 male germ cell tumor survivors according to tinnitusstatus (Cont'd )

CharacteristicsAll participants(n ¼ 762)

Tinnitus controls(n ¼ 608)

Tinnitus casesa

(n ¼ 154)

Audiometrically assessed hearing lossj,k

None/mild 238 (39.4%) 220 (45.4%) 18 (15.0%)Moderate 93 (15.4%) 75 (15.5%) 18 (15.0%)Moderatelysevere

129 (21.4%) 103 (21.3%) 26 (21.7%)

Severe/profound

144 (23.8%) 86 (17.8%) 58 (48.3%)

Peripheralneuropathyl

None 373 (49.0%) 329 (54.1%) 44 (28.6%)Mild 281 (36.9%) 215 (35.4%) 66 (42.9%)Severe 108 (14.1%) 64 (10.5%) 34 (28.6%)

Persistent vertigoor dizzinessm

Yes 36 (4.9%) 15 (2.5%) 21 (14.3%)No 702 (95.1%) 576 (97.5%) 126 (85.7%)

Self-reportedhealthn

Excellent 134 (17.6%) 121 (20.0%) 13 (11.7%)Very good 317 (41.7%) 264 (43.6%) 53 (45.5%)Good 266 (35.0%) 196 (32.3%) 70 (34.4%)Poor/fair 43 (5.7%) 25 (4.1%) 18 (8.4%)

Medication for anxiety, psychosis, and/ordepressiono

Yes 57 (10.1%) 38 (8.3%) 19 (17.75%)No 510 (89.9%) 422 (91.7%) 88 (82.2%)

Hypertension andon medicationYes 106 (14.0%) 69 (11.6%) 37 (24.2%)No 651 (86.0%) 535 (89.9%) 116 (75.8%)

Smoking statusp

Ever smoker 303 (39.9%) 235 (38.8%) 68 (44.2%)Never smoker 457 (60.1%) 371 (61.2%) 86 (55.8%)

Alcoholconsumptionq

�4 drinks perweek

487 (64.2%) 384 (63.4%) 103 (66.9%)

5–7 drinks perweek

172 (22.7%) 140 (24.3%) 32 (20.8%)

�2 drinks perday

100 (13.2%) 81 (12.3%) 19 (12.3%)

NOTE: Data presented as number (%) unless otherwise noted.Abbreviations: BEP, bleomycin, etoposide, and cisplatin; EP, etoposide andcisplatin; NA, not applicable.aRestricted to patients who reported "quite a bit" or "very much" tinnitus.Patients who reported "a little" tinnitus (n ¼ 265) are excluded from the tableand all analyses.bCategory excludes 8 participants for whom age at diagnosis was not recorded.cCategory excludes 8 participants for whom this variable was not stated.dBEP category includes patients who received only bleomycin, etoposide, andcisplatin; EP includes patients who received only etoposide and cisplatin. Theother category includes patients who received other antineoplastic agents,including ifosfamide (n¼45), vinblastine (n¼6), and carboplatin (n¼ 5), orwhowere missing dose data (n ¼ 12).eCategory excludes 5 participants for whom number of cycles was not stated.fCategory excludes 12 participants for whom complete dose data were notavailable.gCategory excludes 3 participants who did not answer this question.hCategory excludes 35 participants who did not answer this question.iCategory excludes 19 participants who did not answer this question.jCategory excludes 158 participants who did not complete audiometric testing.kASHA criteria defined hearing loss severity as the following: mild: 21–40 dB;moderate: 41–55 dB; moderately severe: 56–70 dB; severe: 71–90 dB; andprofound: > 90 dB; for at least one tested frequency for either ear (https://www.asha.org/public/hearing/Degree-of-Hearing-Loss).

lFollowing conversion of the Likert scale: "none, a little, quite a bit, verymuch" toa 0–3 numeric scale, each individual was assigned a summary statistic for thesensory subscale [Cronbach (a ¼ 0.88) and the motor subscale (a ¼ 0.78)by taking the mean of the response in the subscale: none (mean ¼ 0), mild(0 < mean � 1), severe (mean > 1; refs. 16, 17)].mCategory excludes 24 participants who did not answer this question.nCategory excludes 2 participants who did not answer this question.oCategory excludes 195 participants who did not answer this question. Parti-cipants could report more than one psychotropic medication including anxio-lytics, antipsychotics, and antidepressants. Psychotropic medications investi-gated include: alprazolam (n ¼ 8), aripiprazole (n ¼ 2), amphetamine þdextroamphetamine (n ¼ 14), amantadine (n ¼ 1), buproprion (n ¼ 9), carbi-dopa/levodopa (n¼ 1), citalopram (n¼ 8), clonazepam (n¼ 13), desvenlafaxine(n ¼ 1), diazepam (n ¼ 1), duloxetine (n ¼ 8), escitalopram (n ¼ 19), fluoxetine(n ¼ 5), fluvoxamine (n ¼ 1), hydroxazine (n ¼ 1), lisdexamfetamine (n ¼ 4),lorazepam (n ¼ 1), methylphenidate (n ¼ 13), nortriptyline (n ¼ 2), olanzapine(n ¼ 2), paroxetine (n ¼ 5), pramipexole (n ¼ 1), sertraline (n ¼ 7), trazodone(n ¼ 8), valproate (n ¼ 1), venlafaxine (n ¼ 10).pCategory excludes 2 participants for whom this outcome was not stated.qCategory excludes 3 participants for whom this outcome was not stated.

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4109

maintains high potassium concentrations in the endolymph ofthe inner ear (38, 39).

DiscussionTo our knowledge, this is the first investigation to consider the

effects of clinical, genetic, and lifestyle factors on CisIT risk. Here25.8% and 15.2% of cisplatin-treated TCS reported mild andmoderate/severe tinnitus at a median of 5 years after treatment.

Survivors treatedwith 100–400mg/m2 cisplatin appeared to havesimilar CisIT risk, but those given >400 mg/m2 were at 2.6-foldincreased risk, suggesting a nonlinear threshold–dose response.Follow-up time was not associated with CisIT, suggesting irre-versibility, but longitudinal data are needed to support thisobservation. Significant associations with tinnitus wereobserved for hearing loss, vertigo, sensory neuropathy, pooreroverall health, and use of psychotropic medications, as well as forhypertension, noise exposure, and chronic smoking. GWAS

Figure 1.

Tinnitus frequency by cumulative cisplatin dose and age at testicular cancer diagnosis (quartiles). A, Bar plot showing the frequency of tinnitus by cumulativecisplatin dose group (<300, 300, 400, and >400mg/m2) illustrates that the significant association between dose and tinnitus (logistic regression OR¼ 1.38, 95%CI: 1.1–1.7, P¼ 0.007) is due to the significantly increased risk of tinnitus in patients treated with doses > 400mg/m2 (post hoc x2 P < 0.0001). The number ofsubjects per category is presented on the x axis under the dose group label. B, Bar plot representing the frequency of tinnitus by age at diagnosis quartile showspositive association with cisplatin risk (OR by decade¼ 1.31; 95% CI, 1.1–1.6; P¼ 0.006). Error bars represent the binomial 95% CIs for each subgroup. N.S., notsignificant.

100%

75%

50%

25%

0%

80%

60%

40%

20%

0%Controls Cases Controls Cases(n = 606) (n = 460) (n = 107)(n = 154)

Tinnitus status Tinnitus status

P < 0.0001

P = 0.003

Freq

uenc

y

Freq

uenc

y of

psy

chot

ropi

c dr

ug u

se

Excellent

Very good

Good

Poor/fair

Self-reported health

A B

Figure 2.

Self-reported health and use of psychotropic medications by tinnitus status. A, Bar plot of the distribution of self-reported health according to tinnitus status.Patients were asked: "Howwould you rate your overall health?" and responded on a poor–excellent ordinal scale, which was converted to a numeric scale (0¼poor/fair, 1¼ good, 2¼ very good, 3¼ excellent). Logistic regression revealed significant negative correlation between tinnitus status and self-reported health(OR¼ 0.54; 95% CI: 0.4–0.7; P < 0.0001). B, Bar plot of psychotropic medication use by tinnitus status shows significantly higher prevalence of psychotropicmedication use in tinnitus cases (OR¼ 2.4; 95% CI, 1.3–4.4; P¼ 0.003). Patients were dichotomized to "Yes" and "No" psychotropic medication use based onreceiving medications from a list of frequently prescribed antidepressants, anxiolytics, and antipsychotics (Supplementary Methods; ref. 15). The number ofsubjects per category is presented on the x axis. Error bars represent the binomial 95% CIs for each subgroup.

El Charif et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4110

revealed independent signals implicatingOTOS that had relative-ly large effect sizes (ORs of 3.5–4.5) and were relatively rare(minor allele frequencies 1%–5%). We verified protective effectsofOTOS through eQTL analysis and functional validation in twocell lines, demonstrating that decreased OTOS expressionincreases susceptibility to CisIT. Whether thyroid/pituitary eQTLs

coregulate expression in cochlear fibrocytes is unknown, andhuman cochlear transcriptomic databases are necessary for moreprecise interpretations. However, we provide further evidence forOTOS's protective functions in silico using central nervous systemcell line data, and in vitro by overexpressing human OTOS in amouse auditory cell line routinely used for in vitrodrug ototoxicity

0

25

50

75

1000.25 0.5 1 1.5 2 3 4 6 8 10 12

Audiometric frequency (kHz)

Tinnitus

Tinnitus status

Tinnitus status Neuropathic symptoms

Tinnitus status

Controls

Cases

Hea

ring

thre

shol

d (d

B)

100%

75%

50%

25%

0%

Freq

uenc

y

Freq

uenc

y of

pro

blem

s he

arin

g in

a c

row

d

Freq

uenc

y of

per

sist

ent d

izzi

ness

or v

ertig

o

OR

Controls ControlsCases Cases(n = 607)

Controls Cases(n = 591) (n = 147)

(n = 585) (n = 142)(n = 153)

Difficulty hearingNone

A littleQuite a bit/very much

P < 0.0001

P < 0.0001

P < 0.000180%

60%

40%

20%

0%

80%

60%

40%

20%

0%

5

4

3

2

1

0

Pain (H

)

Pain (F

)

Tingli

ng (H

)

Numbn

ess (

H)

Numbn

ess (

F)

Tingli

ng (F

)

A

B

D E

C

Figure 3.

Associations of tinnitus withsubjective and objective hearingloss, problems hearing, vertigo,and neurologic symptoms. A, Boxplot of audiometric hearingthresholds (y axis, dB) across alltested audiometric frequencies(x axis, kHz) showing significantlyworse hearing in tinnitus cases ateach frequency (P < 0.0001 at eachfrequency). Box centers indicatemedians, hinges indicateinterquartile regions (IQR), andwhiskers indicate 1.5� IQRs. Pointsoutside the range of 1.5� IQR areshown. Bar plots of self-reporteddifficulty hearing (OR¼ 6.38; 95%CI: 4.9–8.5; P < 0.0001; B),problems hearing in a crowd(OR¼ 8.28; 95% CI: 5.5–12.5;P < 0.0001; C), and persistentdizziness or vertigo (OR¼ 6.40;95% CI: 3.2–12.9; P < 0.0001;D).Error bars represent the binomial95% CIs for subgroup. E, Forestplot showing the OR (centerpoints) and 95% confidenceintervals (error bars) of theassociation between tinnitus andneurotoxic symptoms from theEORTC-CIPN20. "H" denoteshands/fingers. "F" denotesfeet/toes. Responses toEORTC-CIPN20 items wereconverted from a none–very muchLikert scale to a numerical ordinalscale (0–3). Associations inB–Ewere evaluated usinglogistic regression adjusted forage at diagnosis.

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4111

screening (40) and a mouse fibroblast cell line. Pathway analysisimplicated potassium transport genes, including the cochlearpotassium transporter KCNQ1 (38, 39).

Treatment-related risk factorsDespite adequate statistical power, significant differences in

tinnitus risk between TCS given 3 versus 4 cycles of cisplatin-basedchemotherapy were not evident. Brydøy and colleagues previous-ly showed that TCS treated with >4 cycles of cisplatin-basedchemotherapy had a2.21-fold increased risk of tinnitus compared

with those given 1–4 cycles (12), consistent with our findings.Bokemeyer and colleagues showed a dose-dependent increasedrisk for ototoxicity after cisplatin-based chemotherapy, but didnot specifically assess tinnitus or compare by dosage group (41).

Modifiable risk factorsAn approximate 2-fold association between hypertension

and CisIT was apparent among our TCS. The relationshipbetween hypertension and tinnitus in the general populationis well-established, with a recent meta-analysis estimating a

10

8

6

4

2

0

241.04 241.06 241.08 241.1 241.12

80%

60%

40%

20%

0%

80%

60%

40%

20%

0%

1.0

0.5

0.0

–0.5

–1.0

–log

10 (P

)

Genomic distance on chr2 (Mb)

0.80.60.40.2

r 2

rs7606353

rs10190781 Genotype rs10190781 Genotype

OTOS

MYEOV2

P = 0.007

P = 9.5 × 10−8 0.1

0.0

–0.1

–0.2

–0.3

–0.4

Tinn

itus

frequ

ency

Freq

uenc

y

Cis

plat

in lo

g 2 (A

UC

)N

orm

aliz

ed e

xpre

ssio

n

(n = 740)

(n = 690) (n = 71) (n = 7)

(n = 22) (n = 362) (n = 33) (n = 4)GG GGAG/AA AG AA

0 –6 –5 –4 –3 –21 2

Number of risk allelesNormalized OTOS expression

100

80

60

40

20

0R

ecombination rate (cM

Mb)

A

B

D E

C

Figure 4.

GWAS of cisplatin-induced tinnitus revealsgenetic loci nearOTOS gene. A, LocusZoom plotof the third most significant GWAS signal neartwo adjacent genes:OTOS andMYEOV2. Eachpoint represents a SNP. The x axis indicateschromosomal position. The left y axis shows–log10(P) of association with CisIT and the right yaxis and blue line indicate the recombination ratein centimorgans/megabase (cM/MB). The color ofeach variant indicates the linkage disequilibriumR2 with the top signal in the region, rs7606353(purple). B, Bar plot of CisIT frequency bygenotype ofOTOS eQTL, rs10190781. Theminorallele (G) increases CisIT risk (OR¼ 3.42; P¼0.007). Error bars represent the binomial 95% CIs.C, Boxplot ofOTOS expression in thyroid byrs10190781 genotype indicates that the minorallele is associated with lower expression ofOTOS. Data were obtained from the GTEx Portalon 04/28/18.D, Bar plot of number of risk allelesfor both rs7606353 and rs10190781 showsadditive allele effects [OR¼ 3.77; 95% CI: 2.3–6.2;P¼ 9.5� 10�8]. E, Scatter plot of cisplatinresistance as a function of normalizedOTOSexpression. Cisplatin resistance, measured as thearea under the cisplatin dose–response curve, forall central nervous system tumor lines (19 gliomaand 4 neuroblastoma lines) was extracted fromCancerRX and normalized OTOS expressionswere downloaded from the Cancer Cell LineEncyclopedia. Correlation was assessednonparametrically using Spearman rank method(r¼ 0.46, P¼ 0.03).

El Charif et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4112

pooled OR of 1.37 (P < 0.05; ref. 42). It has been hypothesizedthat hypertension causes damage to either the cochlear micro-circulation and/or the stria vascularis, which supplies bloodto the organ of Corti (43). This could explain the observedassociation.

Multiple studies have identified a relationship between tobaccouse and tinnitus in the general population (44–46) and we founda significantly increased risk of tinnitus in survivors with >15 yearsof tobacco use. It is possible that long-term tobacco use mayreduce peripheral blood flow, including to the inner ear micro-circulation. The role of tobacco use in CisIT was examined byBrydøy and colleagues, but significant associations were notapparent for either current or former smokers compared withnever smokers (40), similar to our results.

Noise exposure is a known risk factor for tinnitus (47), but itsrole in the development of the hearing disorder in other cisplatin-treated cohorts has not been evaluated to our knowledge.Although Bokemeyer and colleagues reported an associationbetween a history of noise exposure and cisplatin-induced oto-toxicity, this term included both tinnitus and hearing loss (41).

Sensory neuropathyIn our cohort, tinnitus was positively associated with symp-

toms of sensory neuropathy. This could suggest that inherentsusceptibility to cisplatin-induced neurophysiological damagemay partially underlie various neuro-otological symptoms,possibly involving pharmacokinetic pathways, intrinsic cisplat-in sensitivity, reduced regenerative capacity, or altered neuralplasticity. Another possible explanation is that of patient per-ception. Self-reported tinnitus, hearing loss, and neuropathy in

cancer survivors have been linked to the Impact of EventsScale-Revised (IES-R) hyperarousal scores (14), suggesting thatdifferences in perception and appraisal could partially explainour findings. One study showed that audiometric thresholdscorrelated with tinnitus awareness time and loudness, but thattinnitus-induced annoyance was associated with proneness toexhibit anxious responses (48, 49). These investigations high-light how both cochlear and central nervous mechanismscontribute to tinnitus.

Tinnitus pathophysiologyWhile single-variant analysis did not identify significant genetic

signals, we found that cumulatively, additive SNP effects explaina large portion of variance in CisIT, further supporting thehypothesis that CisIT is a polygenic trait with complex geneticunderpinnings (4). Several mechanisms of tinnitus have beenproposed, but no consensus on its etiology has been reached.Onemodel postulates that disruptions to the endocochlear potentialresult in inhibitory–excitatory imbalances affecting spontaneouscochlear activity and lead to tinnitus (50). Inner-ear fibrocytessupply potassium ions to strial cells which pump them into theendolymph tomaintain the endocochlear potential, the "cochlearbattery" driving auditory transduction (51). The link betweenendocochlear potential and spontaneous auditory nerve activityhas been established (52). Our GWAS points to the potentialimportance of this hearing mechanism in CisIT.

OTOS mRNA and protein expression localize to spiral limbusand spiral ligament fibrocytes (34), and overexpression protectsprimary spiral ligament fibrocytes against cisplatin cytotoxici-ty (37). In support of the protective effect of OTOS against

PlateCells

Transfection Cisplatin

Hours –48 48 72–24 240CellTiter-Glo CellTiter-Glo

qPCR qPCR qPCR qPCR

Rel

ativ

e ce

ll vi

abili

ty (4

8H)

Rel

ativ

e ce

ll vi

abili

ty (7

2H)

Cisplatin (mmol/L) Cisplatin (mmol/L)

1.00

0.75

0.50

0.25

1.00

0.75

0.50

0.25

0 10 20 30 40 50 0 10 20 30 40 50

pOTOS

pOTOS or pControl

pControl

HEI-OC1 L929

A

B D

EC

Figure 5.

Effects of OTOS overexpression on cisplatinsensitivity in HEI-OC1 mouse cochlear and L929mouse fibroblast cell lines. A, Experimental scheme oftransfection and cisplatin treatment, followed by cellviability and qPCR analysis. Cells were transfectedwith humanOTOS (pOTOS) or a control plasmid(pControl) for 24 hours prior to cisplatin treatment.Cell viability of HEI-OC1 cells transfected with eitherpOTOS (closed circles) or the control plasmid(pControl; open circles) following treatment withincreasing concentrations of cisplatin for 48 hours(P¼ 0.004; B) or 72 hours (P¼ 0.008; C). Inset,HumanOTOSmRNA expression in HEI-OC1 cellscorresponding to 0, 24, 48, and 72 hours aftercisplatin treatment (24, 48, 72, and 96 hours aftertransfection) depicted as pOTOS relative to pControl.Cell viability of L929 fibroblasts transfected witheither pOTOS (closed circles) or the control plasmid(pControl; open circles) following treatment withincreasing concentrations of cisplatin for 48 hours(P¼ 0.006; D) or 72 hours (P¼ 0.03; E). Inset, HumanOTOSmRNA expression in L929 fibroblastscorresponding to 0, 24, 48, and 72 hours aftercisplatin treatment (24, 48, 72, and 96 hours aftertransfection) depicted as pOTOS relative to pControl.Data shown are from three independent experiments,with each concentration or time point within anexperiment evaluated in duplicate or triplicate. Errorbars reflect SEM. Statistical significance wasdetermined using two-way ANOVA analysis withSidak correction for multiple comparisons.

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4113

cisplatin sensitivity, our in silico analysis indicated that cisplatinsensitivity is positively associatedwith increasedOTOS expressionin CNS tumor cell lines, a rough approximation of cell types thatwould be affected by cisplatin-induced neurotoxicity. While CNStumor cell lines are not an ideal proxy for assessing toxicity toinner ear cells, we further validated the protective effects ofOTOSby demonstrating that its overexpression in vitro reduces cisplatinsensitivity in two cell lines (HEI-OC1 and L929), with HEI-OC1more accurately reflecting inner ear cells that are adversely affectedby cisplatin.

In support of ourfindings, twoprior studies have elucidated theimportance of OTOS expression in mammalian hearing. Induc-ible knockdown of OTOS in guinea pigs caused severe degener-ation of the organ of Corti and deafness, along with damage tofibrocytes (34). Interestingly, although Otos�/� mice had dam-aged fibrocytes, they did not exhibit histologic abnormalities ofsensory hair cells, and experienced only moderate deafness (35).Delprat and colleagues attributed this difference to long-termadaptation to congenital Otos absence (34). Therefore, it is plau-sible that inducible knockdown, modeling key aspects of directauditory insults such as cisplatin administration, will acutely alterthe ionic composition of the endolymph, disrupting hair cellmetabolism and survival.

Breglio and colleagues recently showed a reduction of endoco-chlear potential following cisplatin administration in mice, andpreferential accumulation of cisplatin in the stria vascularis (53).We linked potassium transport to CisIT in our pathway analysis,which most strongly implicated KCNQ1, an important cochlearion channel. This voltage-gated channel localizes tomarginal cellsof the stria vascularis where it maintains high Kþ concentrationsin the endolymph (38). Mutations in KCNQ1 result in deafnessand long QT syndrome (54).

Taken together, these findings point to the function of support-ing cells in the cochlear lateralwall in cisplatin ototoxicity andhaircell survival. Impairedmaintenance of the endolymph causes haircell death and permanent hearing loss (55), which may bepartiallymediatedby a reduced endocochlear potential.However,prior research has largely focused directly on cisplatin's cytotoxicmechanisms in hair cells (56). Our new findings, together withthose in the literature, suggest that novel strategies aimed atprotecting supporting cells in the cochlear lateral wall, perhapsby stimulatingOTOS expression, may prove beneficial. Very littleis known about the protein product of the gene, its functions, andits interactions. Better understanding of the functional proteinand its cellular localization and interactions is warranted. Apotential approach would be to use a gene expression–basedhigh-throughput screen to identify small molecules or siRNAcapable of markedly increasing OTOS expression without per-turbing the viability of auditory cells, such as HEI-OC1. Theidentification of compounds or siRNAs resulting in increasedOTOS expression could be evaluated for their effect on protectingcells from cisplatin damage. Importantly, the utility of thisapproach depends on the modulation not impacting the anti-neoplastic activity of cisplatin.

It remains unclear whether the effects of OTOS on tinnitusoccur directly or are mediated by effects on hearing loss.However, if OTOS protects against tinnitus indirectly by pre-venting hearing loss, it remains a viable candidate for cisplatinotoprotection. It is also unclear whether the actions of OTOSare uniquely pertinent to cisplatin or generalizable to otherototoxic insults like noise, although evidence of direct protec-

tion against cisplatin cytotoxicity is presented here and in theliterature (37). Future studies should elucidate the cellularfunctions and interactions of otospiralin to assess its viabilityas a target for cisplatin otoprotection.

ConclusionsStrengths of our study include the collection of detailed treat-

ment data, homogeneous cisplatin-based chemotherapy, long-term follow-up, quantitative hearing evaluations, the use ofvalidated instruments to assess neurotoxic symptoms, and thenovel implementation of genome-wide scans to nominatecellular mechanisms and targets. Our results point to cochlearmechanisms that have not yet been widely explored in cisplatinototoxicity, and may provide novel opportunities for targetedotoprotection.

An intrinsic limitation of all cross-sectional studies is theinability to make causal inferences. Furthermore, questionnaireswith more detailed information on the experience of tinnitus,including baseline measurements pre-chemotherapy, would pro-vide more conclusive evidence with regard to associations notedhere. However, given the young median age of patients at treat-ment, it is unlikely that tinnitus was pre-existing. It is important tonote that the results of our translational research could potentiallyimpact millions of patients worldwide annually diagnosed withcisplatin-treated cancers who are at risk for CisIT. Our findingsmay also inform tinnitus research in nonplatinum–treatedpatients. In view of our results, health care providers can improvemanagement of cancer survivors experiencing CisIT by recom-mendingmonitoring of blood pressure, encouragement of smok-ing cessation, and avoidance of additional ototoxic insults such asnoise exposure.

Disclosure of Potential Conflicts of InterestB. Mapes is an employee of and has ownership interests (including patents)

at Tempus Inc. D.R. Feldman reports receiving commercial research supportfrom Novartis, Seattle Genetics and Decibel, and receives royalties from reviewchapter of UpToDate. R.J. Hamilton is a consultant/advisory boardmember forBayer, Janssen, Amgen, Astellas, and Abbvie, and reports receiving commercialresearch support from Janssen and Bayer. C. Kollmannsberger is a consultant/advisory boardmember for Pfizer, Astellas, Bristol-Myers Squibb, EMD Serono,Ipsen, Eisai, Janssen, Merck and Roche. E.R. Gamazon is a consultant/advisoryboard member for City of Hope/Beckman Research Institute. L.H. Einhorn hasownership interests (including patents) at Amgen and Biogenidec, and is aconsultant/advisory board member for Celgene Advisory Board. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: O.E. Charif, B. Mapes, H.E. Wheeler, R.D. Frisina,D.J. Vaughn, C. Kollmannsberger, M. Kubo, N.J. Cox, L.B. Travis,M.E. DolanDevelopment of methodology: O.E. Charif, B. Mapes, H.E. Wheeler, C. Wing,R.D. Frisina, E.R. Gamazon, M.E. DolanAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M.R. Trendowski, C. Wing, D.R. Feldman,R.J. Hamilton, D.J. Vaughn, C. Fung, C. Kollmannsberger, T. Mushiroda,M. Kubo, R. Huddart, L.B. Travis, M.E. DolanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): O.E. Charif, B. Mapes, M.R. Trendowski,H.E. Wheeler, C. Wing, P.C. Dinh Jr., R.D. Frisina, D.R. Feldman,D.J. Vaughn, C. Kollmannsberger, E.R. Gamazon, P. Monahan, S.D. Fossa,L.H. Einhorn, L.B. Travis, M.E. DolanWriting, review, and/or revision of the manuscript: O.E. Charif, B. Mapes,M.R. Trendowski, H.E. Wheeler, P.C. Dinh Jr., R.D. Frisina, D.R. Feldman,R.J. Hamilton, D.J. Vaughn, C. Fung, C. Kollmannsberger, M. Kubo,

El Charif et al.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4114

E.R. Gamazon, N.J. Cox, R. Huddart, S. Ardeshir-Rouhani-Fard, P. Monahan,S.D. Fossa, L.H. Einhorn, L.B. Travis, M.E. DolanAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): B. Mapes, P.C. Dinh Jr., D.R. Feldman,R.J. Hamilton, M. Kubo, L.B. TravisStudy supervision: M. Kubo, S.D. Fossa, L.B. Travis, M.E. Dolan

AcknowledgmentsThe HEI-OC1 cell line was a generous gift from Dr. Federico Kalinec,

House Ear Institute, Los Angeles, CA. The L929 cell line was a generous giftfrom Dr. Wendy Stock at the University of Chicago. This work was supportedby the NIH Genetic Susceptibility and Biomarkers of Platinum-relatedToxicities grant (R01 CA157823), and Pharmacogenomics Research Network(PGRN)-RIKEN Global Alliance which is supported by the RIKEN Centerfor Integrative Medical Science and the NIH Pharmacogenomics Research

Network (GM115370). The Genotype-Tissue Expression (GTEx) Project wassupported by the Common Fund of the Office of the Director of the NationalInstitutes of Health, and by the NCI, NHGRI, NHLBI, NIDA, NIMH, andNINDS. The data used for the analyses described in this manuscript wereobtained from the GTEx Portal on 04/28/18. R.J. Hamilton acknowledgessupport of The Josh Carrick Charity and the Institute of Cancer Research andRoyal Marsden NIHR Biomedical Research Centre.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received September 27, 2018; revised February 13, 2019; acceptedMarch 25,2019; published first April 5, 2019.

References1. Bhatt JM, Lin HW, Bhattacharyya N. Prevalence, severity, exposures, and

treatment patterns of tinnitus in theUnited States. JAMAOtolaryngolHeadNeck Surg 2016;142:959–65.

2. Weidt S, Delsignore A, Meyer M, Rufer M, Peter N, Drabe N, et al. Whichtinnitus-related characteristics affect current health-related quality of lifeand depression? A cross-sectional cohort study. Psychiatry Res 2016;237:114–21.

3. Friberg E, Jansson C, Mittendorfer-Rutz E, Rosenhall U, Alexanderson K.Sickness absence due to otoaudiological diagnoses and risk of disabilitypension: a nationwide Swedish prospective cohort study. PLoS One 2012;7:e29966.

4. Vona B, Nanda I, Shehata-Dieler W, Haaf T. Genetics of tinnitus: still in itsinfancy. Front Neurosci 2017;11:236.

5. Langguth B, Kreuzer PM, Kleinjung T, De Ridder D. Tinnitus: causes andclinical management. Lancet Neurol 2013;12:920–30.

6. Han BI, Lee HW, Kim TY, Lim JS, Shin KS. Tinnitus: characteristics, causes,mechanisms, and treatments. J Clin Neurol 2009;5:11–9.

7. Cianfrone G, Pentangelo D, Cianfrone F, Mazzei F, Turchetta R, OrlandoMP, et al. Pharmacological drugs inducing ototoxicity, vestibular symp-toms and tinnitus: a reasoned and updated guide. Eur Rev Med PharmacolSci 2011;15:601–36.

8. Rybak LP, Mukherjea D, Jajoo S, Ramkumar V. Cisplatin ototoxicity andprotection: clinical and experimental studies. Tohoku J ExpMed 2009;219:177–86.

9. Frisina RD, Wheeler HE, Fossa SD, Kerns SL, Fung C, Sesso HD, et al.Comprehensive audiometric analysis of hearing impairment and tinnitusafter cisplatin-based chemotherapy in survivors of adult-onset cancer.J Clin Oncol 2016;34:2712–20.

10. Hartmann JT, Lipp HP. Toxicity of platinum compounds. Expert OpinPharmacother 2003;4:889–901.

11. Ruggiero A, Trombatore G, Triarico S, Arena R, Ferrara P, Scalzone M, et al.Platinum compounds in children with cancer: toxicity and clinical man-agement. Anticancer Drugs 2013;24:1007–19.

12. BrydoyM, Oldenburg J, Klepp O, Bremnes RM,Wist EA, Wentzel-Larsen T,et al. Observational study of prevalence of long-term Raynaud-like phe-nomena and neurological side effects in testicular cancer survivors. J NatlCancer Inst 2009;101:1682–95.

13. Miaskowski C, Paul SM, Mastick J, Schumacher M, Conley YP, Smoot B,et al. Hearing loss and tinnitus in survivors with chemotherapy-inducedneuropathy. Eur J Oncol Nurs 2018;32:1–11.

14. Miaskowski C, Paul SM, Mastick J, Abrams G, Topp K, Smoot B, et al.Associations between perceived stress and chemotherapy-induced periph-eral neuropathy and otoxicity in adult cancer survivors. J Pain SymptomManage 2018;56:88–97.

15. Postma TJ, Aaronson NK, Heimans JJ, Muller MJ, Hildebrand JG, DelattreJY, et al. The development of an EORTC quality of life questionnaire toassess chemotherapy-induced peripheral neuropathy: the QLQ-CIPN20.Eur J Cancer 2005;41:1135–9.

16. Fung C, Sesso HD, Williams AM, Kerns SL, Monahan P, Abu Zaid M, et al.Multi-institutional assessment of adverse health outcomes among NorthAmerican testicular cancer survivors after modern cisplatin-based chemo-therapy. J Clin Oncol 2017;35:1211–22.

17. Wheeler HE, Gamazon ER, Frisina RD, Perez-Cervantes C, El Charif O,Mapes B, et al. Variants in WFS1 and other mendelian deafness genes areassociated with cisplatin-associated ototoxicity. Clin Cancer Res 2017;23:3325–33.

18. DolanME, El Charif O,Wheeler HE, Gamazon ER, Ardeshir-Rouhani-FardS, Monahan P, et al. Clinical and genome-wide analysis of cisplatin-induced peripheral neuropathy in survivors of adult-onset cancer.Clin Cancer Res 2017;23:5757–68.

19. Das S, Forer L, Schonherr S, Sidore C, Locke AE, Kwong A, et al. Next-generation genotype imputation service andmethods. Nat Genet 2016;48:1284–7.

20. Loh PR, Danecek P, Palamara PF, Fuchsberger C, A Reshef Y, K Finucane H,et al. Reference-based phasing using the Haplotype Reference Consortiumpanel. Nat Genet 2016;48:1443–8.

21. McCarthy S, Das S, KretzschmarW,DelaneauO,Wood AR, Teumer A, et al.A reference panel of 64,976 haplotypes for genotype imputation.Nat Genet 2016;48:1279–83.

22. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al.PLINK: a tool set for whole-genome association and population-basedlinkage analyses. Am J Hum Genet 2007;81:559–75.

23. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets.Gigascience 2015;4:7.

24. AndersonCA, Pettersson FH,ClarkeGM,CardonLR,Morris AP, ZondervanKT. Data quality control in genetic case-control association studies.Nat Protoc 2010;5:1564–73.

25. GTEx Consortium. The genotype-tissue expression (GTEx) project.Nat Genet 2013;45:580–5.

26. PooleW, Gibbs DL, Shmulevich I, Bernard B, Knijnenburg TA. Combiningdependent P-values with an empirical adaptation of Brown's method.Bioinformatics 2016;32:i430–i6.

27. Yang J, Lee SH, Goddard ME, Visscher PM. GCTA: a tool for genome-widecomplex trait analysis. Am J Hum Genet 2011;88:76–82.

28. AshburnerM, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Geneontology: tool for the unification of biology. The Gene Ontology Consor-tium. Nat Genet 2000;25:25–9.

29. The Gene Ontology Consortium. Expansion of the Gene Ontology knowl-edgebase and resources. Nucleic Acids Res 2017;45:D331–D8.

30. Reimand J, Arak T, Adler P, Kolberg L, Reisberg S, Peterson H, et al. g:Profiler-a web server for functional interpretation of gene lists (2016update). Nucleic Acids Res 2016;44:W83–9.

31. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S,et al. The Cancer Cell Line Encyclopedia enables predictive modelling ofanticancer drug sensitivity. Nature 2012;483:603–7.

32. Yang W, Soares J, Greninger P, Edelman EJ, Lightfoot H, Forbes S, et al.Genomics of Drug Sensitivity in Cancer (GDSC): a resource for therapeuticbiomarker discovery in cancer cells. Nucleic Acids Res 2013;41(Databaseissue):D955–61.

33. HW. Ggplot2 Elegant Graphics for Data Analysis. New York, NY: SpringerScience; 2009.

34. Delprat B, Boulanger A, Wang J, Beaudoin V, Guitton MJ, Venteo S,et al. Downregulation of otospiralin, a novel inner ear protein,

Genetics of Cisplatin-induced Ototoxicity

www.aacrjournals.org Clin Cancer Res; 25(13) July 1, 2019 4115

causes hair cell degeneration and deafness. J Neurosci 2002;22:1718–25.

35. Delprat B, Ruel J, GuittonMJ, Hamard G, Lenoir M, Pujol R, et al. Deafnessand cochlear fibrocyte alterations inmice deficient for the inner ear proteinotospiralin. Mol Cell Biol 2005;25:847–53.

36. Ward LD, Kellis M. HaploReg: a resource for exploring chromatinstates, conservation, and regulatory motif alterations within sets ofgenetically linked variants. Nucleic Acids Res 2012;40(Database issue):D930–4.

37. Zhuo XL, Wang Y, Zhuo WL, Zhang YS, Wei YJ, Zhang XY. Adenoviral-mediated up-regulation of Otos, a novel specific cochlear gene, decreasescisplatin-induced apoptosis of cultured spiral ligament fibrocytes viaMAPK/mitochondrial pathway. Toxicology 2008;248:33–8.

38. Chang Q, Wang J, Li Q, Kim Y, Zhou B, Wang Y, et al. Virally mediatedKcnq1 gene replacement therapy in the immature scala media restoreshearing in a mouse model of human Jervell and Lange-Nielsen deafnesssyndrome. EMBO Mol Med 2015;7:1077–86.

39. Wangemann P. Kþ cycling and the endocochlear potential. Hear Res 2002;165:1–9.

40. Kalinec G, Thein P, Park C, Kalinec F. HEI-OC1 cells as a model forinvestigating drug cytotoxicity. Hear Res 2016;335:105–17.

41. Bokemeyer C, Berger CC, Hartmann JT, Kollmannsberger C, Schmoll HJ,KuczykMA, et al. Analysis of risk factors for cisplatin-induced ototoxicity inpatients with testicular cancer. Br J Cancer 1998;77:1355–62.

42. Yang P, Ma W, Zheng Y, Yang H, Lin H. A systematic review and meta-analysis on the association between hypertension and tinnitus. Int JHypertens 2015;2015:583493.

43. Yang S, Cai Q, Bard J, Jamison J,Wang J, YangW, et al. Variation analysis oftranscriptome changes reveals cochlear genes and their associated func-tions in cochlear susceptibility to acoustic overstimulation. Hear Res 2015;330(Pt A):78–89.

44. Shargorodsky J, Curhan GC, Farwell WR. Prevalence and characteristics oftinnitus among US adults. Am J Med 2010;123:711–8.

45. Nondahl DM, Cruickshanks KJ, Huang GH, Klein BE, Klein R, Nieto FJ,et al. Tinnitus and its risk factors in the Beaver Dam offspring study. Int JAudiol 2011;50:313–20.

46. Nondahl DM, Cruickshanks KJ, Wiley TL, Klein BE, Klein R, Chappell R,et al. The ten-year incidence of tinnitus among older adults. Int J Audiol2010;49:580–5.

47. Baguley D, McFerran D, Hall D. Tinnitus. Lancet 2013;382:1600–7.48. MoonKR,ParkS, JungY,LeeA,Lee JH.Effectsofanxiety sensitivityandhearing

loss on tinnitus symptom severity. Psychiatry Investig 2018;15:34–40.49. Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity,

anxiety frequency and the prediction of fearfulness. Behav Res Ther1986;24:1–8.

50. Norena AJ. Revisiting the cochlear and central mechanisms of tinnitus andtherapeutic approaches. Audiol Neurootol 2015;20Suppl 1:53–9.

51. Garcia Berrocal JR, Mendez-Benegassi I, Marti C, Ramirez Camacho R.[Intervention of spiral ligament fibrocytes in the metabolic regulation ofthe inner ear]. Acta Otorrinolaringol Esp 2008;59:494–9.

52. Sewell WF. The relation between the endocochlear potential andspontaneous activity in auditory nerve fibres of the cat. J Physiol 1984;347:685–96.

53. Breglio AM, Rusheen AE, Shide ED, Fernandez KA, Spielbauer KK, McLa-chlin KM, et al. Cisplatin is retained in the cochlea indefinitely followingchemotherapy. Nat Commun 2017;8:1654.

54. Vyas B, Puri RD,Namboodiri N, Nair M, SharmaD,Movva S, et al. KCNQ1mutations associated with Jervell and Lange-Nielsen syndrome and auto-somal recessive Romano-Ward syndrome in India-expanding the spectrumof long QT syndrome type 1. Am J Med Genet A 2016;170:1510–9.

55. Liu H, Li Y, Chen L, Zhang Q, Pan N, Nichols DH, et al. Organ of corti andstria vascularis: is there an interdependence for survival? PLoS One 2016;11:e0168953.

56. Sheth S, Mukherjea D, Rybak LP, Ramkumar V. Mechanisms ofcisplatin-induced ototoxicity and otoprotection. Front Cell Neurosci2017;11:338.

Clin Cancer Res; 25(13) July 1, 2019 Clinical Cancer Research4116

El Charif et al.