Randomized controlled trial to test the efficacy of trans-tympanic injections of a sodium thiosulfate gel to prevent cisplatin-induced ototoxicity.

Abstract

e17560 Background: The hearing loss due to cisplatin cochlear damage is frequent and severe. Antioxidants, such as sodium thiosulfate (STS), can neutralize the effects of cisplatin, but, when administered systemically, they decrease its efficacy. In animal experiments, STS deposited in the middle ear reached the cochlea and reduced cisplatin ototoxicity. We conducted a randomized controlled trial to test the efficacy of trans-tympanic injections of a STS gel to prevent cisplatin-induced ototoxicity. Methods: Eligible participants were patients with symmetrical hearing treated for a locally advanced head and neck cancer with chemoradiation including 3 cisplatin cycles (100 mg/m2). For each participant, one randomly selected ear received the injections while the other ear did not. On the eve of each cisplatin treatment, a trans-tympanic injection deposited 0.1 ml of an immediately prepared STS-hyaluronate gel (0.5 M) on the round window. The main outcome was assessed blindly using the shift of hearing threshold in decibels (dB) from before chemoradiation to one month thereafter for pure-tone air conduction at 0.5-14 kHz frequencies. Adverse effects were noted according to CTCAE. Assuming a lower hearing loss of 7.0 dB for the treated ears, 0.90 power and 0.05 two-sided statistical significance, 25 patients were needed. The main outcome was assessed in a mixed linear model with hearing threshold shift as dependent variable and intervention, frequency and radiation dose to the cochlea as independent variables. Results: From January to December 2015, 13 patients were randomized. The trial was stopped in June 2016 for poor accrual. The average loss of hearing over all frequencies was 1.5 dB less for the treated ears than for the control ears. The difference was not statistically (p = 0.56) nor clinically significant, but was consistently in favor of the treated ears for all frequencies between 3 and 10 kHz. The intervention adverse effects were mild. Conclusions: Our trial suggests that STS deposited in the middle ear reached the cochlea but was not clinically effective. More work is needed to improve the efficacy of trans-tympanic administration of cisplatin antidotes. Clinical trial information: NTC02281006.