Hearing loss in children with osteosarcoma

Abstract

9517 Background: Cisplatin is one of the most effective agents against osteosarcoma. Cisplatin is usually administered as a short IV infusion in most contemporary North American osteosarcoma treatment protocols. Ototoxicity from such protocols has not been well defined. The aim of this study was to determine the incidence and the risk factors for hearing loss in children with osteosarcoma. Methods: Eligible patients (pts) in this retrospective cohort study had osteosarcoma diagnosed and treated at our center from 1/1/1995 to 12/31/2004, were 3–18 years of age at diagnosis, and had a normal audiogram prior to the start of chemotherapy. All pts received cisplatin according to either standard practice or current open protocol. Most regimens prescribed a cumulative dose of 480 mg/m2 over 4 cycles. Pts with hearing loss at lower doses had cisplatin held on an individualized basis. Hearing was evaluated using pure-tone audiometry prior to the start of therapy, before each cycle of cisplatin administration, and off-therapy. The primary endpoint was any hearing loss as defined as > 20 dB decrease in hearing above 4,000 Hz. Fisher's exact test and logistic regression methods were used to identify univariate and independent predictors of hearing loss, respectively. Results: The incidence of hearing loss was 16/40 (40%). Six pts discontinued cisplatin early due to hearing loss. Seven of 9 pts (77.8%) who received cisplatin 120 mg/m2 as a short IV infusion on 1 day developed hearing loss compared to 9 of 31 (29.0%) pts who received this same dose as two daily infusions (p=0.018). On univariate analyses, age and cumulative cisplatin dose were not predictive of hearing loss. Logistic regression controlling for age, sex, race, and cumulative dose showed that administration of cisplatin 120 mg/m2 on 1 day compared to 2 days was an independent predictor of hearing loss (odds ratio of 11.6, 95% confidence interval = 1.62–82.87). Conclusions: Cisplatin given as 120 mg/m2/dose on 1 day was associated with greater risk of hearing loss than 60 mg/m2/dose on 2 days. Cumulative dose and age were not found to be predictors of hearing loss in this relatively small cohort. No significant financial relationships to disclose.