To identify clinical and treatment-related predictors of brachial plexus-associated neuropathies after radiation therapy for head-and-neck cancer.

Three hundred thirty patients who had previously completed radiation therapy for head-and-neck cancer were prospectively screened using a standardized instrument for symptoms of neuropathy thought to be related to brachial plexus injury. All patients were disease-free at the time of screening. The median time from completion of radiation therapy was 56 months (range, 6-135 months). One-hundred fifty-five patients (47%) were treated by definitive radiation therapy, and 175 (53%) were treated postoperatively. Radiation doses ranged from 50 to 74 Gy (median, 66 Gy). Intensity-modulated radiation therapy was used in 62% of cases, and 133 patients (40%) received concurrent chemotherapy.

Forty patients (12%) reported neuropathic symptoms, with the most common being ipsilateral pain (50%), numbness/tingling (40%), motor weakness, and/or muscle atrophy (25%). When patients with <5 years of follow-up were excluded, the rate of positive symptoms increased to 22%. On univariate analysis, the following factors were significantly associated with brachial plexus symptoms: prior neck dissection (p = 0.01), concurrent chemotherapy (p = 0.01), and radiation maximum dose (p < 0.001). Cox regression analysis confirmed that both neck dissection (p < 0.001) and radiation maximum dose (p < 0.001) were independently predictive of symptoms.

The incidence of brachial plexus-associated neuropathies after radiation therapy for head-and-neck cancer may be underreported. In view of the dose-response relationship identified, limiting radiation dose to the brachial plexus should be considered when possible.