Prevalence and risk factors for ototoxicity after cisplatin-based chemotherapy

Ototoxicity is a prominent side effect of cisplatin-based chemotherapy. There are few reports, however, estimating its prevalence in well-defined cohorts and associated risk factors. Testicular cancer (TC) survivors given first-line cisplatin-based chemotherapy completed validated questionnaires. Descriptive statistics evaluated the prevalence of ototoxicity, defined as self-reported hearing loss and/or tinnitus. We compared patients with and without tinnitus or hearing loss using Chi-square test, two-sided Fisher’s exact test, or two-sided Wilcoxon rank sum test. To evaluate ototoxicity risk factors, a backward selection logistic regression procedure was performed. Of 145 TC survivors, 74% reported ototoxicity: 68% tinnitus; 59% hearing loss; and 52% reported both. TC survivors with tinnitus were more likely to indicate hypercholesterolemia (P = 0.008), and difficulty hearing (P < .001). Tinnitus was also significantly related to age at survey completion (OR = 1.79; P = 0.003) and cumulative cisplatin dose (OR = 5.17; P < 0.001). TC survivors with hearing loss were more likely to report diabetes (P = 0.042), hypertension (P = 0.007), hypercholesterolemia (P < 0.001), and family history of hearing loss (P = 0.044). Risk factors for hearing loss included age at survey completion (OR = 1.57; P = 0.036), hypercholesterolemia (OR = 3.45; P = 0.007), cumulative cisplatin dose (OR = 1.94; P = 0.049), and family history of hearing loss (OR = 2.87; P = 0.071). Ototoxicity risk factors included age, cisplatin dose, cardiovascular risk factors, and family history of hearing loss. Three of four TC survivors report some type of ototoxicity; thus, follow-up of cisplatin-treated survivors should include routine assessment for ototoxicity with provision of indicated treatments. Survivors should be aware of risk factors associated with ototoxicity. Referrals to audiologists before, during, and after cisplatin treatment is recommended.


Introduction
Cancer survivors are at risk for both acute and chronic adverse effects related to cancer treatment. Research focused on common treatment toxicities are at the forefront for survivorship care planning due to the increased life span of cancer survivors [55]. More than 18.1 million patients were considered cancer survivors in 2022 increasing the 5-year survival rate from just 39% in 1960s to over 68% [2]. The increases in survival rates are associated with improved treatment and surveillance options [4]; however, an estimated 50-80% of patients will live with chronic health conditions from cancer treatments [2]. Some of the most common chronic health conditions in cancer survivorship include heart disease, hypertension, respiratory difficulties, endocrine complications, early onset menopause, fertility complications, arthritis/osteoporosis, and neurotoxicities including ototoxicity [7,66,67]. Cisplatin is one of the most ototoxic chemotherapy drugs used in oncology practice, with indications including cancers of testis, cervix, and head-and-neck [58,65]. Consequences of cisplatin-related ototoxicity experienced by cancer survivors includes hearing loss in up to 80% patients [3,29,41,44,83] which is often accompanied with tinnitus [14,22]. Cisplatin chemotherapy rapidly causes damage to the inner ear and central auditory pathways [65,74]. This sensorineural hearing loss is characterized as bilateral, symmetrical, and permanent [65]; thus, becoming a chronic health condition with no preventive or protective measures available for adultonset cancer survivors in regard to ototoxicity secondary to cisplatin treatment. In fact, cisplatin is retained in the inner ear (cochlea) indefinitely [9], and there are reports that cisplatin-related hearing loss may be progressive [40], but it is unknown if there are continuous or long-term damages related to this retention.
Some cancers, such as testicular cancer (TC), are identified in young adulthood, thus the retention of cisplatin in the cochlea is especially troubling as these patients have upwards of 50 years of lifespan ahead of them. A 2018 study [33] found a significant effect of cisplatin-related hearing loss that exacerbated age-related hearing loss in TC survivors; however, in a conflicting later study from this group [70], age-related hearing loss appeared similar to cisplatin-related hearing loss, supporting the theory that cisplatin may not exacerbate agerelated hearing loss. Regardless of the mechanism(s) of action of cisplatin-related hearing loss, there is increased concern for the potential long-term consequences of hearing loss that have been reported in the general population, including clinically relevant decreases in communication and health-related quality of life [10,11,17,54], decreased social interactions [63,72,73,82], and increased risks of accelerated cognitive decline and dementia [47][48][49]. Thus, hearing loss and tinnitus related to cisplatin-based chemotherapy are important to understand in cancer survivorship. To add to the state of the science, we report here the prevalence of ototoxicity in a well-defined cohort of adult-onset cancer survivors and report-related risk factors. We evaluated hearing loss and tinnitus in a subset of TC survivors from a large multi-center investigation [The Platinum Study; [29,30,38]].

Patients
From 2012 to 2018, The Platinum Study enrolled cisplatintreated TC survivors at eight cancer centers including Indiana University [29,30,38]. These survivors completed comprehensive physical and audiological examinations and questionnaires [15,29,30,38,84]. We administrated a subsequent questionnaire to participants enrolled at Indiana University under the ethical approval of the governing Institutional Review Board. Here we report on the 145 patients that completed this survey as of February 11th, 2022. As of this date, 223 pts at Indiana University had consented to the study, representing a response rate of 145/223 = 65%. The participants at Indiana University were similar to participants in the overall Platinum Study.

Surveying patients
Patients were asked to complete a 20-page questionnaire through electronic entry via REDCap, or via classic paperand-pen collection methods. The survey was a composite of validated instruments including those from the Patient-Reported Outcomes Measurement Information System [PROMIS; [1]], the EORTC Chemotherapy-Induced Peripheral Neuropathy-20 Scale [EORTC-CIPN20; [62]], and the Scale for Chemotherapy-Induced Long-Term Neurotoxicity [SCIN; [56]]. Additional validated questions assessed demographic and social determinants of health, and patient-reported adverse health outcomes. Our interdisciplinary research team overseeing survey construction included oncologists, biostatisticians, psychometricians, geneticists, clinical pharmacologists, hearing scientists, and epidemiologists. The majority of participants indicated that it took less than 30 min to complete the survey.

Identifying patients with ototoxicity
Ototoxicity was defined as patients reporting tinnitus, hearing loss, or both tinnitus and hearing loss. Patients were identified with hearing loss if they reported hearing problems on the EORTC-CIPN20 and SCIN instruments, or indicated they had difficulty hearing in crowds, or reported hearing aid use. Similarly, patients with tinnitus were identified if they indicated "ringing or buzzing in their ears" or through the tinnitus-specific question in the SCIN instrument that asked, "Have you suffered in the last 4 weeks from ringing or buzzing in your ears (i.e., tinnitus)?".

Statistical analyses
Descriptive statistics are provided as frequencies (percentages) for categorical variables or medians [interquartile range (IQR)] for continuous variables. Bivariate comparisons between patients with and without tinnitus or hearing loss were conducted using a Chi-square test (or two-sided Fisher's exact test if more than 20% of table cells had expected counts less than 5) for categorical variables, or the twosided Wilcoxon rank sum test for continuous variables. To evaluate risk factors for ototoxicity, we performed logistic regression with a backward selection procedure, separately for each outcome, including the following variables in the initial modeling step: age at survey evaluation, body mass index (BMI), diabetes, hypertension, hypercholesterolemia, tobacco history, noise exposure, history of family hearing loss, and cisplatin dose. These variables were selected based on common variables that are typically reported in models assessing auditory disorders [29,30,37,42,75]. During backward selection, a single variable was removed in each step, based on the highest P-value. This was done until all variables in a model had P < 0.05 in the last step, resulting in one model for each step. From among the set of models where all variables demonstrated P < 0.10, the final model was chosen with the lowest Akaike Information Criteria (AIC). The AIC is an approximation of cross-validation prediction error and therefore indicates model quality. The final models were used to evaluate associations between the variables and ototoxicity. Table 1 presents sociodemographic characteristics and health behaviors of the TC survivors (n = 145). In Table 1, we also show features of subgroups of patients reporting tinnitus (n = 98), hearing loss (n = 86), both tinnitus and hearing loss (n = 76), and those with no hearing loss or tinnitus (n = 37). Of 145 TC survivors, hearing loss was reported by 86 (59%) patients, while tinnitus was reported by 98 (68%) patients. There were more patients reporting tinnitus than hearing loss; however, 76 (52%) patients reported both hearing loss and tinnitus. Because subgroups with tinnitus, hearing loss, or both have overlapping patients (i.e., non-mutually exclusive subgroups) and are only mutually exclusive with the neither symptom subgroup (n = 37), we use Table 1 to descriptively evaluate clinically meaningful differences between the mutually exclusive neither-symptom (n = 37) subgroup and each of the other three subgroups, individually. Subsequent tables pursue statistical comparisons between appropriate groups. In Table 1, among 145 TC survivors, the majority were white and middle-aged [median age at survey completion, 46 years (IQR: [39][40][41][42][43][44][45][46][47][48][49][50][51][52]]. Patients without hearing loss or tinnitus (right-most column) were generally slightly younger [median age at survey completion, 40 years (IQR: [36][37][38][39][40][41][42][43][44][45][46]] than those reporting hearing loss, tinnitus, or both hearing loss and tinnitus (all three sub-groups had a median age of 48 years old at the time of survey completion). Similarly, those without hearing loss or tinnitus were only 9 years out from chemotherapy completion, while those with hearing loss, tinnitus, or hearing loss and tinnitus were 13 or 14 years out from chemotherapy completion. There were no clinically meaningful descriptive differences between those without tinnitus and hearing loss and the other subgroups relating to marital status, education, employment status and type of employment, income, or insurance. With respect to health behaviors, there were no meaningful differences in alcohol consumption; however, 89% of patients without hearing loss or tinnitus reported never smoking, while only 74-80% with hearing loss, tinnitus, or both never smoked. Table 2 shows detailed clinical features and adverse health outcomes for patients with tinnitus (n = 98) compared to those without tinnitus (n = 47), with the results of the statistical comparisons shown in the far-right column. TC survivors reporting tinnitus were significantly older at the time of survey completion than those without tinnitus (P = 0.004). The median age at time of survey for those reporting tinnitus was 48 years (IQR 40-53) compared to 43 years (IQR 36-47) for those not reporting tinnitus. Similarly, those reporting tinnitus had a longer duration since their last chemotherapy treatment, being 13 years (IQR 8-19) compared to 9 years (IQR 6-15) for those not reporting tinnitus (P = 0.012). Those reporting tinnitus were also significantly more likely to report hypercholesterolemia (44% vs. 21%; P = 0.008) and had received higher total doses of cisplatin (P < 0.001).

Results
A significant descriptive finding is that 97% of TC survivors with tinnitus reported suffering from tinnitus in the last 4 weeks. The majority of the total cohort of TC survivors (59%) reported hearing loss; and 78% of the patients with tinnitus also reported hearing loss, while only 21% of patients without tinnitus had hearing loss (P < 0.001). Patients with tinnitus indicated greater difficulty distinguishing between high-and low-pitched sounds (P < 0.001), reported difficulty hearing in crowds with background noise (P < 0.001), and also stated that their communication partners noted they had difficulty hearing (P < 0.001). Surprisingly, only 9% patients with tinnitus reported hearing aid use, which is an effective treatment for tinnitus for most patients with bothersome tinnitus [45]. Table 3 reports clinical features and adverse health outcomes for patients with hearing loss (n = 86) compared to those without hearing loss (n = 59). Those with hearing loss were older; the median age for those with hearing loss was 48 years (IQR 41-56) versus 43 years (IQR 35-49) for those without hearing loss (P < 0.001). Similarly, TC survivors with hearing loss reported longer duration since chemotherapy completion (14 years (IQR 9-20)), compared to those without hearing loss (9 years (IQR 6-14)) (P < 0.001). TC survivors with hearing Table 1 Sociodemographic characteristics and health behaviors for 145 cisplatin-treated TC survivors. a Subgroups are reported for the TC survivors with tinnitus (n = 98), hearing loss (n = 86), tinnitus and hearing loss (n = 76), and patients with no tinnitus and no hearing loss (n = 37) a Data are presented as median [interquartile range] or count (%) for a given column, unless otherwise noted in the table footnotes. All percentages are column percentages b Patients were classified with tinnitus based on the Scale for Chemotherapy-Induced Long-Term Neurotoxicity (SCIN; [56] or by indicating that they had symptoms of ringing or buzzing in their ears c Patients were classified with hearing loss based on affirmative responses to the European Organization for Research and Treatment of Cancer Chemotherapy-Induced Peripheral Neuropathy 20 Scale (EORTC-CIPN20; [62], Scale for Chemotherapy-Induced Long-Term Neurotoxicity (SCIN, [56], or by answering yes to either hearing aid use or having difficulty hearing in crowds d Includes "no college," "some college," and "other/not specified" e International Standard Classification of Occupations (ISCO) guidelines were used to group occupations into skill levels 1 & 2 and 3 & 4. Skill levels 1-2 included Automotive worker, building/construction worker, crafts and related trades, information technology, plant and machine operator and assembler, services and sales, skilled agricultural, forestry, and fishery. Skill levels 3-4 includes management, medical doctor, healthcare provider, professional, scientist or researcher, teacher/education, and technical professional. Data are missing from 20 patients; 17 from the sub-group reporting both tinnitus and hearing loss, 1 from the subgroup reporting tinnitus, and 2 from the subgroup reporting neither tinnitus nor hearing loss On disability 6 (4%) 6 (6%) 6 (7%) 6 (6%) 0 Retired 8 (6%) 6 (6%) 6 (7%) 6 (6%) 2 (5%) loss were also significantly more likely to have diabetes (8% vs. 0%; P = 0.042), hypercholesterolemia (50% vs. 17%; P < 0.001), hypertension (42% vs. 20%; P = 0.007), report tobacco use (26% vs. 12%; P = 0.043), and a family history of hearing loss before age 60 (21% vs. 9%; P = 0.044). Cumulative cisplatin dose was significantly related to hearing loss (P = 0.020). Patients with hearing loss indicated greater difficulty distinguishing between high and low pitches (P < 0.001), Table 2 Clinical features and adverse health outcomes stratified by presence or absence of tinnitus reported by 145 cisplatin-treated TC survivors. a Significant differences using an unadjusted bivariate analysis between those with or without tinnitus are shown in the far-right column BEP, bleomycin, etoposide, and cisplatin; EP, etoposide and cisplatin a Data are presented as Median [Interquartile Range] or Count (%) for a given column, unless otherwise noted in the table footnotes. All percentages are column percentages b Patients were classified with tinnitus based on the Scale for Chemotherapy-Induced Long-Term Neurotoxicity (SCIN; [56] or by indicating that they had symptoms of ringing or buzzing in their ears c Patients were classified with tinnitus based on the Scale for Chemotherapy-Induced Long-Term Neurotoxicity (SCIN; [56]   reported more difficulty hearing in crowds with background noise (P < 0.001), and also reported that their communication partners noted they had difficulty hearing (P < 0.001).
Not surprisingly, only those with hearing loss reported hearing aid use (P = 0.011); however, only 11% of all TCS with hearing loss reported hearing aid use.
The clinical characteristics of TC survivors shown in Table 2 and Table 3 are also the same variables we selected to evaluate risk factors for hearing loss and tinnitus in Table 4. Tinnitus was significantly related to age at survey completion (OR = 1.79 per 10 years; P = 0.004) and cumulative cisplatin dose (OR = 5.17 per 100 mg/m 2 ; P < 0.001). Hearing loss was significantly related to age at survey completion (OR = 1.57 per 10 years; P = 0.036), hypercholesterolemia (OR = 3.45; P = 0.007), and cumulative cisplatin dose (OR = 1.94 per 100 mg/m 2 ; P = 0.0498). Notably, hearing loss was marginally significantly (P < 0.10) related to family history of hearing loss before age 60 (OR = 2.87; P = 0.071).

Discussion
Among 145 TC survivors, the prevalence of ototoxicity was high with 74% indicating some type of ototoxicity (i.e., either tinnitus, hearing loss, or both tinnitus and hearing loss). Specifically, 68% TC survivors reported tinnitus and 59% reported hearing loss. Results of our ototoxicity prevalence is on the higher side, since the prevalence of ototoxicity in the literature associated with cisplatin-based treatments ranges from 4 to 90% [43], but our estimate is congruent with past reports from our team with similar patient populations and homogenous treatment regimens [3,22,29,76,83,84]. Within our cohort, there were more patients reporting tinnitus than hearing loss, with 52% of the patients reporting both hearing loss and tinnitus, indicating that not all TCS with tinnitus also noted hearing loss. Alarmingly, 97% of those with tinnitus reported they had suffered in the past 4 weeks from their tinnitus. Interestingly, 88% of participants with hearing loss also reported tinnitus and 87% of participants with hearing loss reported they suffered from tinnitus in the past 4 weeks.
In our cohort, TC survivors with tinnitus were older and had a greater interval in years since chemotherapy administration than those without tinnitus. Although some studies show a reduction in cisplatin-based tinnitus with time [57], we showed that our patients are still suffering for up to 19 years after chemotherapy from tinnitus, with 97% reporting acute distress within the past 4 weeks. This suggests the need for long-term survivorship care for TC survivors suffering from tinnitus. With only 9% of TC survivors reporting limited hearing aid usage in our study, there are concerns about lack of follow-up and knowledge on the usefulness of hearing devices for tinnitus relief [34,69]. There are no known FDA-approved medicinal treatments for tinnitus; however, there is evidence that hearing aid use can reduce the sensation of tinnitus, improve speech perception, and boost hearing [69]. The majority of our patients with tinnitus also reported hearing loss and their communication partners also indicated they had difficulty hearing, again reinforcing the need for audiological evaluation and appropriate rehabilitative treatments. In addition to tinnitus, those with hearing loss reported other comorbidities including diabetes, hypertension, and hypercholesterolemia. On multivariate analyses, we found that hypercholesterolemia was an associated risk factor for hearing loss, as it is in the general population [68]. Hypercholesterolemia is commonly defined as elevated levels of low-density lipoprotein cholesterol (LDL-C) or nonhigh-density lipoprotein cholesterol (HDL-C), diagnosed by a lipid profile, often related to a sedentary lifestyle, an unhealthy diet, with excessive consumption of saturated fats, trans-fatty acids, and cholesterol [50]. In addition to poor lifestyle choices, other associations with hypercholesterolemia include diabetes, excess body weight (mainly in the abdominal region), hypothyroidism, nephrotic syndrome, and cholestatic liver disease [50]. Hypercholesterolemia is also a risk factor for atherosclerotic cardiovascular disease [50,77]; thus, clinicians can discuss both hearing health and cardiovascular health during cancer treatment and beyond. Our TC patients with hearing loss also reported more prior or current use of tobacco than did those without hearing loss. In the general population, smokers are 1.69 times more likely to have hearing loss than non-smokers [16], and diabetes, hypertension, and hypercholesterolemia are common co-morbidities associated with hearing loss regardless of etiology [8]. Our patients with ototoxicity (compared to those without) had higher prevalence of both of these cardiovascular risk factors and as well as smoking, suggesting poorer health behaviors for these patients prior to cancer treatment. Clinicians should consider discussing the risk of ototoxicity with all patients, but special attention may need to be provided to those reporting poorer health choices to encourage them to improve their overall general health profile through cancer treatment and beyond.
Our TC survivors with hearing loss were found to have complications discriminating different pitches, had problems hearing in large crowds with background noise, and also reported that their communication partners noted they had difficulty hearing. These are all classic manifestations of sensorineural hearing loss. Further, beyond communication concerns, untreated hearing loss is known to be associated with many negative health consequences, including cognitive decline, fatigue, social isolation, depression, and other negative health outcomes [13,19,46,53,63,72,73]. Untreated hearing loss leads to an annual average of $22,434 more in per-person healthcare costs due to increased hospitalizations, emergency department visits and outpatient urgent appointments [64]. We are surprised that only 10% of our TC survivors with ototoxicity reported hearing aid use, suggesting a gap in access, affordability, knowledge of treatment benefit, or lack of treatment adherence [78]. We more closely examined the nine patients that reported hearing aid use; 4/9 reported "quite a bit" and 5/9 reported "very much" difficulty hearing at time of survey. All nine hearing aid users reported difficulty hearing in crowds and background noise, and also reported tinnitus. Our study clinicians also indicate that our young survivors report stigma associated with wearing hearing aids, and that ototoxicity-related complaints shared with the clinicians lessen with time which may be due to the young survivors becoming more gradually accustomed to their hearing loss. Auditory rehabilitation including the use of hearing aids can enhance communication [71] and health-related quality of life [12,18,35,39]; and, cost-effective devices [27] and coverage are becoming more available to all patients with hearing loss [5].
Our patients with hearing loss were more likely to report a family history of hearing loss than those without hearing loss. We also considered family history of hearing loss in our models evaluating risk factors for ototoxicity. Family history of adult-onset hearing loss was a bivariately significant risk factor; and, although only marginally significant in our multivariable model, it should be further evaluated since considerable variability in the prevalence and severity of ototoxicity has been observed among patients receiving similar cisplatin doses, suggesting a role for genetic susceptibility. There are also genetic associations being discovered relating certain genes to ototoxicity [43,76,83,84]. Genome-wide association studies in TC survivors who received cisplatin identified genetic variants in WFS1 as significantly associated with hearing loss [83], and variants near OTOS associated with tinnitus although P-values were only suggestively significant [22]. In gene-based analysis, TXNRD1, which plays a key role in redox regulation, showed borderline significance with hearing loss and WNT8A (P = 2.5 × 10 -6 , a gene encoding a signaling protein important in germ cell tumors was significantly associated with tinnitus [84]. Genetic testing is becoming more accessible, but until it is widely available, it would seem prudent to ask patients about a family history of hearing loss as a possible way to identify those who might be at high-risk. Family history of hearing loss has been shown to be a risk factor in epidemiologic studies of sensorineural hearing loss with varying etiologies [52], and is a known risk factor for pediatric hearing loss [20]. However, we are unaware of any other studies that have evaluated self-report of familial hearing loss as a risk factor for cisplatin-related ototoxicity. Identification of patterns of inheritance of hearing loss could be informative not only for patient education, but also for ototoxicity management programs. Although clinical practice today cannot identify genetic predictors of susceptibility to ototoxicity before the administration of cisplatin-based chemotherapy, continued progress is being made in ototoxicity-related pharmacogenomics, with a goal of personalized medicine to help with the prevention of ototoxicity and other cisplatin-related health problems.
Although family history of hearing loss was a marginally significant risk factor for hearing loss in the final multivariate model, other identified factors included age, hypercholesterolemia, and total cisplatin dose. Only age and total cisplatin dose were found to be significantly associated with tinnitus in the multivariate model. Aligned with our reports, significant associations between cumulative cisplatin dose and ototoxicity are consistently reported [6, 21-23, 83, 84]. Associated risk factors are key for clinicians to be aware of when assessing cancer patients and providing survivorship/treatment planning. Our results indicate that patient age, total cisplatin dose, family history of adult-onset hearing loss, history of diabetes, hypercholesterolemia, and tobacco use should be queried and evaluated during oncological follow-up.
The long-term consequences of cisplatin-related ototoxicity have not been adequately studied in young adults [37]. Ototoxicity may interact with the effects of aging, leading to more severe hearing loss than that associated with only natural aging. Similar to other reports [37], we found that our TC survivors with ototoxicity were significantly older than our patients without hearing loss or tinnitus. Also, age at time of survey completion was significantly associated with hearing loss.
Findings in our study need to be interpreted within the context of its strengths and weaknesses. Strengths of our investigation include the well-defined cohort, homogenous cisplatin based chemotherapy treatment administered at a center of excellence for germ cell tumors, and use of validated surveys. However, adverse health outcomes were selfreported, without baseline data before treatment. Any crosssectional design has potential inherent limitations and does not permit us to rigorously infer causation of evaluated risk factors to adverse health outcomes. Nonetheless, many of the risk factors that were identified here for hearing loss and tinnitus have also been documented in the general population, and can thus inform the development of survivorship follow-up care.

Implications for cancer survivors and conclusions
Follow-up of cisplatin-treated survivors should include routine assessment for ototoxicity, asking about both tinnitus and hearing difficulty. Survivors should be aware of influences associated with an increased likelihood of tinnitus and hearing loss. These risk factors include cisplatin dose, cardiovascular risk factors, and family history of adult-onset hearing loss. Lifestyle modifications can be discussed with patients to improve physical health through exercise and adoption of a healthy diet. The best practice guidelines for cardiovascular risk factors include improved nutrition, physical activity, weight loss, and pharmacologic intervention for those at risk for atherosclerotic cardiovascular disease (Handelsman et al., 2020).
Referral to audiologists are needed since ototoxicity adversely affects health-related quality of life [60], and untreated hearing loss is associated with negative health outcomes [72]. An interdisciplinary approach with the audiologist, oncologist, radiation oncologist, oncology advanced practitioner, and oncology nurse should be used to achieve optimal patient outcomes [58]. Patients should receive otologic evaluations prior to cisplatin administration and have their hearing monitored during treatment and into survivorship [59,81], which will not only improve patient health but also provide critical data for improving future treatment plans and options.
New oncology patient appointments for patients with germ cell tumors should include a thorough history to obtain known risk factor information for ototoxicity. Providers should be alerted for patients with a family history of adult-onset hearing loss and cardiovascular risk factors since these may be associated with an increased risk for ototoxicity from cisplatin treatment; these potential risk factors are also known to be related to hearing loss in the general population too [8]. Patients should be educated with regard to the signs and symptoms of tinnitus and hearing loss and encouraged to report these promptly to their cancer care team so early interventions can take place. Interventions for some cancers, but not testicular cancer [32], include chemotherapy dose adjustment, the selection of an alternative regimen if one is available and equally efficacious, referral to the audiology team for further assessment, and starting the conversation early about possible chronic otologic effects. Survivorship care planning is an optimal treatment area in which to include baseline audiology assessments for cancer patients, with additional audiologic assessments as indicated. Given the known late effects of hearing loss and tinnitus, we should include annual assessments for ototoxicity especially as patients age throughout their cancer survivor journey. Formulating cancer care plans with the patient, family and interdisciplinary team will improve patient outcomes. There are no current preventive or protective measures available against ototoxicity secondary to cisplatin treatment. Please see Guthrie and Spankovich, this special section, for further discussion about promising potential pharmaceutical agents being developed to provide preventive, protective and treatment options [25,31,36,80]. For example, sodium thiosulfate (STS was granted fast-track designation from the FDA to prevent cisplatin-related hearing loss in pediatric hepatoblastoma [26,28,61], and intratympanic dexamethasone and sodium thiosulfate pentahydrate has shown some promise to potentially reduce cisplatin-induced hearing loss [51,79]. There are several severe side effects noted for systemically administered STS via intravenous infusion, therefore it must be used with considerable caution [24], and, while intratympanic administration reduces systemicadministration side effects, it is not always practical clinically [51]. Currently, there are no FDA-approved agents to prevent cisplatin-related hearing loss in virtually all cancer patients and until those are available, thorough survivorship planning including ototoxicity management is required. Funding Drs. Frisina, Dolan, Sesso, Dinh, Feldman, Monahan, and Travis were supported by 2 R01 CA157823 funded by the National Cancer Institute. Dr. Sanchez was supported by R01DC019408-01 funded by the National Institute of Deafness and Other Communication Disorders.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on request.

Declarations
Ethics approval This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee at Indiana University. Informed consent was obtained from all participants in this study.

Competing interests
The authors declare no competing interests.