Whole-organ histopathological study of recurrent nasopharyngeal carcinoma.

OBJECTIVES/HYPOTHESIS To study the histopathological features of recurrent nasopharyngeal carcinoma and the accuracy of preoperative magnetic resonance imaging (MRI). STUDY DESIGN Prospective. METHODS Whole-organ study of nasopharyngectomy specimens. RESULTS Between 2006 and 2009, 50 specimens were obtained after maxillary swing nasopharyngectomy and sent for whole-organ study. The tumors arose from the fossa of Rosenmüller (68%), posterior wall (18%), or roof of the nasopharynx (14%), and the majority (72%) took the form of an ulcerative tumor. The T-classifications of the recurrent tumors were: T1, 24%; T2, 48%; and T3, 28%. All the tumors appeared as islands of cancer cells separated by lymphoplasmacytic infiltrate and fibrosis. The tumor size measured by MRI correlated closely with that measured histologically, especially for the depth of invasion and parapharyngeal extension. For tumors with parapharyngeal extension, removal of the pharyngobasilar fascia was essential to ensure a clear margin on the surface of the petrosal internal carotid artery. None of the tumors showed invasion of the Eustachian tube. CONCLUSIONS Contrast MRI is accurate in assessing the local extent of recurrent nasopharyngeal carcinoma. During nasopharyngectomy, a radial resection margin of 15 mm should be taken with the underlying medial pterygoid muscle. For tumors with parapharyngeal extension, the pharyngobasilar fascia should be resected en bloc with the specimen. LEVEL OF EVIDENCE N/A.