Patient-Reported and Dosimetric Determinants of Trismus after Chemoradiation with Proton Beam for Nasopharyngeal Carcinoma: A Prospective Study

Purpose/Objective(s)Trismus is a common side effect following radiation therapy for head and neck cancer and may be most severe for patients with nasopharyngeal carcinoma. To date, prospective efforts investigating patient reported outcomes combined with dosimetric predictors are lacking.Materials/MethodsWe conducted a prospective, phase II clinical trial of adult patients with biopsy proven, stage IIB-IVB nasopharyngeal cancer treated with concurrent proton-based chemoradiotherapy (cisplatin 100 mg/m2 q3week), followed by adjuvant cisplatin and 5-fluorouracil. Trismus was assessed in a multifactorial fashion. Objective trismus was defined as a decrease in maximal inter-incisal distance (MID) of >20%, relative to baseline. Patient-reported trismus was assessed using trismus domain of the EORTC QLQ H&N35 questionnaire. The relationship between radiation dose to the muscles of mastication / temporomandibular joint (TMJ) and the development of trismus was evaluated using Cox regression.ResultsBetween 2006-2011, 24 patients were enrolled. Median age was 48 years old. Median follow-up was 3.0 years. 63% of patients had stage III/IV disease. The mean doses to the lateral pterygoids, medial pterygoids, suprahyoids, infrahyoids, masseters, temporalis, and TMJs was 54, 51, 16, 46, 14, 12, and 32 Gy RBE, respectively. Eleven patients developed objective trismus. The median decrease in MID at 24 months was 6mm (range 1-19mm, IQR 5-11mm). Nine patients developed patient-reported trismus. Association of presence vs absence of patient-reported trismus and decline in 2y MID was seen (median 10mm vs 5mm, p=.03). Dose (per Gy increase) to the lateral pterygoids was associated with patient-reported trismus (HR 1.16, 95% CI 1.01-1.33, p=.04) and magnitude of decline in 2y MID (β=0.48mm, 95% CI 0.01-0.96mm, p=.045) but not objective trismus (HR 1.07, 95% CI 0.96-1.19, p=.25). Dose (per Gy increase) to TMJs was associated with patient-reported trismus (HR 1.14, 95% CI 1.02-1.26, p=.02), objective trismus (HR 1.12, 95% CI 1.01-1.25, p=.04) magnitude of decline in 2y MID (β =0.53mm, 95% CI 0.12-0.95mm, p=.01). Patients with patient-reported trismus had higher mean dose to TMJ (36 vs 30 Gy, p=0.01) and lateral pterygoids (57 vs 52 Gy, p=0.03) than those who did not.ConclusionIn this prospective study incorporating a homogeneous patient population and treatment regimen, the development of patient-reported trismus after chemoradiation with proton beam was associated with increase in mean dose to the lateral pterygoids and TMJ and decline in 2y MID. Delineation and avoidance of these structures seems prudent.