The influence of positive margins and nerve invasion in adenoid cystic carcinoma of the head and neck treated with surgery and radiation

Abstract

Purpose: Surgery is the primary treatment for adenoid cystic carcinomas arising from major and minor salivary glands of the head and neck. However, local recurrence is frequent because of the infiltrative growth pattern and perineural spread associated with these tumors. At UTMDACC, we have had a longstanding policy of using postoperative radiotherapy to reduce the risk of local recurrence and to avoid the need for radical surgery; this 30-year retrospective study analyzes the results of this combined modality approach.

Methods and materials: Between 1962 and 1991, 198 patients ages 13-82 years, with adenoid cystic carcinomas of the head and neck, received postoperative radiotherapy for known or suspected microscopic residual disease following surgery. Distribution of primary sites was: parotid: 30 patients; submandibular/sublingual: 41 patients; lacrimal: 5 patients; and minor salivary glands: 122 patients. Eighty-three patients (42%) had microscopic positive margins and an additional 55 (28%) had close (< or = 5 mm) or uncertain margins. One hundred thirty-six patients (69%) had perineural spread with invasion of a major (named) nerve in 55 patients (28%). Using radiation techniques appropriate to the primary site, a median dose of 60 Gy (range 50-69 Gy) was delivered to the tumor bed. Follow-up ranged from 5-341 months (median, 93 months). All surviving patients had a minimum of 2 years follow-up.

Results: Twenty-three patients (12%) had local recurrences with 5-, 10-, and 15-year actuarial local control rates of 95%, 86%, and 79%, respectively. Fifteen of the 83 patients (18%) with positive margins developed local recurrences, compared to 5 of 55 patients (9%) with close or uncertain margins, and 3 of 60 patients (5%) with negative margins (p = 0.02). Patients with and without a major (named) nerve involved had crude failure rates of 18% (10 out of 55) and 9% (13 out of 143), respectively (p = 0.02). There was a trend toward better local control with increasing dose. This was significant in patients with positive margins, in whom crude control rates were 40 and 88% for doses of < 56 Gy and > or = 56 Gy, respectively (p = 0.006). Actuarial 5-, 10-, and 15-year freedom from relapse rates were 68%, 52%, and 45%, respectively. Base of skull and neck failures were uncommon with or without elective treatment, developing in 2 and 3% of patients, respectively. Distant metastases were the most common type of disease recurrence, developing in 74 patients (37%) of whom 62 (31%) were disease-free at the primary site.

Conclusions: Excellent local control rates were obtained in this population using surgery and postoperative radiotherapy and we recommend this combined approach for most patients with adenoid cystic carcinomas of the head and neck. Perineural invasion was an adverse prognostic factor only when a major (named) nerve was involved. Microscopic positive margins was also an adverse prognostic factor, but even when present, local control was achieved in over 80% of our patients. We recommend a dose of 60 Gy to the tumor bed, supplemented to 66 Gy for patients with positive margins. Despite effective local therapy, one-third of patients fail systemically, and good treatment to address this problem is lacking.