Hearing Screening After Chemotherapy: A Study On Childhood Cancer Survivors
The Study Background
The survival rates of children with cancer have increased steadily since the mid-1970s, from 58% to 85% in developed countries. However, these improvements are tainted by immediate and late effects of treatment. Geenen et al. and Oeffinger et al. reported that up to 75% of all childhood cancer survivors (CCS) experience at least one late effect due to cancer therapy. One of the serious late effects is ototoxicity, which occurs in about 50% of pediatric cancer cases treated with platinum-based chemotherapy, cranial irradiation, or both.
Ototoxicity involves damage to the cochlea, or auditory nerve, and sometimes the vestibular system, leading to hearing loss, vertigo, and/or tinnitus. Currently, over 150 drugs, including aminoglycosides, glycopeptides, macrolides, platinum-based chemotherapeutic drugs such as cisplatin and carboplatin, loop diuretics, and salicylates, are considered ototoxic. Hearing loss was detected in 20–60% of cases where the drug was cisplatin. However, in lower doses (50-60Gy), there was no clinically significant hearing loss.
Some authors reported that though ototoxicity is usually irreversible, the incidence of hearing loss is variable, due to the method of administration, tumor location, age of the patient, the cumulative and total dose of the drug, and individual susceptibility. Furthermore, previous studies on hearing loss after platinum-based chemotherapy were either single-center or selected clinics, had relatively small samples, were restricted to specific diagnostic groups, or had short follow-up windows. These limitations prevented the findings from being extrapolated. Furthermore, since there is no consensus on the definition of hearing loss, it is considered a binary outcome with an artificial cut-off that differs between studies across the world.
The study aims to describe the severity of hearing loss using audiogram data among a representative national cohort of CCS and identify the clinical risk factors.
Methods
4628 CCS from the Swiss Childhood Cancer Registry who were diagnosed with cancer at ≤18 age and treated with platinum-based chemotherapy between 1990 and 2014 were analyzed. Patients with a normal hearing before starting cancer treatment and an available post-treatment audiogram were included. The study excluded deceased patients at the time of data extraction.
Audiograms, platinum-based chemotherapy and cranial radiation-related information, cancer-related information, and demographic data were extracted. The primary outcome of the study was hearing function. The Munster Ototoxicity Scale was applied and grouped the participants into i) no hearing loss (Münster Ototoxicity grade 0), ii) mild (Münster grade 1), iii) moderate (Münster grade 2a-2c), or iv) severe hearing loss (Münster grade ≥3).
The SIOP Boston Ototoxicity Scale was applied to grade the participants based on the severity of their hearing loss: i) no hearing loss (SIOP Boston grade 0), ii) mild (SIOP Boston grade 1), iii) moderate ((SIOP Boston grade 2), or severe hearing loss ((SIOP Boston grade ≥3). Munster grade ≥2b and SIOP Boston grade ≥2 were considered a clinically relevant hearing loss.
To assess ototoxic treatment and potential risk factors in these childhood cancer survivors, the authors extracted the protocol of platinum-based chemotherapy and cranial radiation therapy (CRT), treatment arm, type of platinum agent, and the total dose per cycle in mg/m2. The total cumulative dose of administered cisplatin and carboplatin was recorded separately. The total cumulative dose of cisplatin was divided into four: i) no cisplatin; (ii) ≤300 mg/m2; (iii) 301–450 mg/m2; and (iv) >450 mg/m2. Similarly, total carboplatin dose was divided into four categories: i) no carboplatin, ii) 1500 mg/m2, iii) 1500-3000 mg/m2, and iv) > 3000 mg/m2.
The correlation between ototoxicity and clinical variables was assessed using the chi-square test. Multivariable log-binomial models were adjusted for age at diagnosis, prior cancer diagnosis, sex, cumulative dose of cisplatin and carboplatin, concomitant CRT, and hematopoietic stem cell transplantation (HSCT). Stata version 15.1 (Stata Corp LP, Austin, TX, USA) was used for all statistical analyses.
RESULTS
Of the 4628 childhood cancer patients from the Swiss Childhood Cancer Registry, only 542 were treated with platinum-based chemotherapy and were alive during data collection in 2015. Of those eligible for the study, 153 (28%) had no available audiogram, 6 (1%) had lost hearing before cancer treatment, 99 (18%) had no post-treatment audiogram, and 14 (3%) had non-classifiable audiograms.
This study analyzed data from 270 (50%) childhood cancer survivors, among whom 154 were males (57%) and 116 (43%) were females. The median age at the most recent audiogram was 13.5 (interquartile range IQR: 9.3-17.0) years, the median age at cancer diagnosis was 6.8 (IQR: 2.1-11.7) years, and the median time since diagnosis was 5 (IQR: 2.5-8.1) years.
CNS tumors were the most common diagnosis (n = 104; 39%), followed by neuroblastomas (n = 62; 57%). About 154 patients were treated with cisplatin, 62 with carboplatin, and 54 with both platinum drugs.
Mild hearing loss was found in 53 survivors, 78 had moderate and 75 had severe hearing loss. Severity of hearing loss was higher in patients i) diagnosed at a younger age (odds ratio OR 5.4, 95% confidence interval [CI]; 2.5-12.0 FOR <5 years; ii) received treatment in early years (OR 4.8; 95%CI: 2.1-11.0 for 1990-1995); iii) had higher cumulative cisplatin doses (OR 13.5, 95% CI: 4.7-38.8 for >450 mg/m2); iv) received concomitant CRT (OR 4.4, 95% CI: 2.5-.8); and v) HSCT (OR 2.7, 95% CI: 1.0-7.2)
DISCUSSION
In this nationwide cohort of childhood cancer survivors, the authors found that three out of four CCS treated with platinum drugs experienced some degree of hearing loss at the latest follow-up/ and one out of four CCS suffered from severe hearing loss. Survivors with CNS tumors and melanoblastoma had a high prevalence of mild, moderate, and severe hearing loss. Consistent with previous studies, the present study found <5 years at cancer diagnosis, high cisplatin dose, concomitant CRT, and HSCT to be the risk factors for higher severity of hearing loss.
Several authors reported hearing loss as a binary outcome. Lack of consensus on the definition of hearing loss makes comparisons between ototoxicity studies in childhood cancer survivors difficult. Camet et al. and Langer et al. differentiated several severity grades of hearing loss. The current study found treatment in an earlier period was a risk factor for higher severity of hearing loss.
Skinner et al. reported hearing loss in 73% of the children and adolescents diagnosed with solid tumors and treated with cisplatin. Brock et al. found hearing loss in about 48% of the 29 children with different cancer diagnoses, treated with cisplatin. The authors found similar figures in the present study (57%). However, the authors noted that though a cumulative cisplatin dose of 301-450 mg/m2 increases the risk of hearing loss, a dose greater than 450 mg/m2 did not further increase the risk. In this study, carboplatin was found to be less toxic than cisplatin. However, the literature is conflicting. Qaddoumi et al. and several others observed an association between carboplatin and hearing loss, especially in patients with retinoblastoma and neuroblastoma.
Limitations
The study has some limitations: i) retrospective study design, ii) lack of data about aminoglycosides or otoprotective drugs, and iii) this study included only CCS who were alive at the time of data extraction and had been treated with platinum-based chemotherapeutic drugs.
Oncology Related Tools
Other
Latest Research
