Melanoma of the ear: prognostic factors and surgical strategies

Abstract

The ear's specific anatomical and lymphatic characteristics impose special requirements on the treatment of melanoma of the ear. The aim of this prospective study was to define prognostic factors for melanoma of the ear and to evaluate surgical strategies for excision margins, histological evaluation and sentinel lymph node biopsy (SLNB) in order to achieve better cosmetic and functional results. One-hundred and sixty-one patients with stage I/II melanoma of the external ear were treated in the Department of Dermatology, University of Tuebingen, from March 1976 to March 2004 (median follow-up 62 months). Malignant melanoma of the external ear represented 3% of the stage I/II cutaneous melanomas and 20% of the stage I/II head and neck melanomas recorded in the Melanoma Registry of the Department of Dermatology at the University of Tuebingen. Twenty of 42 lentigo maligna melanomas (LMM) underwent conventional histological evaluation, 22 underwent complete three-dimensional histology of excision margins (3D histology) in a paraffin-technique, i.e. micrographic surgery. SLNB was performed in 28 patients with melanomas thicker than 1.0 mm. Clinical, histological and surgical risk factors were evaluated by univariate and multivariate analysis. The median thickness of the tumours in the present study was 1.08 mm (mean 1.51 mm; range 0.18-8.50 mm), and the median excision margins were 11.0 mm (mean 12.61 mm; range 2.0-31.0 mm). The 3-year disease-specific survival rate was 98%, and the 3-year recurrence-free survival rate was 83%. Tumour thickness and invasion level were the only risk factors significant for disease-specific survival. Tumour thickness, location of the tumour and extent of excision margins were independently significant risk factors for recurrence-free survival. LMMs removed surgically with accompanying 3D histology were thicker than those examined by conventional histology (median 0.93 mm vs. 0.83 mm). The use of surgery with 3D histology, i.e. micrographic surgery, made it possible to reduce the excision margins (median 5 mm vs. 10 mm) without an increased risk of recurrence. Two of 29 SLNBs were positive (6.9%). There were six preregional recurrences after negative SLNB and one after positive SLNB. None of the patients who underwent SLNB died of melanoma-related causes during the observation period. This is the largest series of ear melanomas reported so far. The overall survival depended only on the tumour thickness and Clark level of invasion. Local recurrence was more frequent with smaller excision margins, but this did not influence the overall survival. Smaller excision margins under 3D-histological control did not carry an increased risk of local recurrence. Our results do not permit conclusions regarding the prognostic impact of SLNB for patients with melanoma of the ear.