Melanoma of the Ear: Prognostic Factors and Surgical Strategies

V. Jahn, H. Breuninger; C. Garbe; M. Moehrle

Disclosures

The British Journal of Dermatology. 2006;154(2):310-318. 

In This Article

Results

The group studied consisted of 161 patients (stage I ⁄ II melanomas) with a median age of 59·9 years (22–88 years). The clinical and histological risk factors are summarized in Table 2 .

Multiple-step Surgery. Of our patients, 76·4% (n = 123) were treated with multiple-step surgery, for example excisional biopsy followed by definitive surgery with an excision margin or multiple-step 3D histology (i.e. micrographic surgery) in the case of LMM.

Excision Margins. These ranged from 2·0mm to 31·0 mm (median 11 mm, mean 12·61 mm).

Three-dimensional Histology. Thirty-nine melanomas (24·2%) of all histological types were removed with 3D histology. 3D histology with reduced initial excision margins was used for 22 of 42 patients with LMM.

Elective Lymph Node Dissection (ELND). ELND was performed in 5·6% (n = 9) of our patients from April 1979 to November 1991 (results: one positive, eight negative).

Sentinel Lymph Node Biopsy. SLNB was performed in 18·0% (n = 29) of the patients from October 1996 to March 2004 (Fig. 1). Results were positive in two of the 29 SLNBs (6·9%).

Patient with a nodular melanoma on the right ear and sentinel lymph node biopsy. (a) Lymphography with peritumoral intradermal injection of blue dye (patent blue V) and sentinel lymph node biopsy. (b) Situation directly after excision of the melanoma and closure by a helical rim advancement flap. (c) Result after 6 months (the patient refused further scar correction).

Defect Reconstruction. Precise data about defect reconstruction were documented in 146 subjects: locoregional flaps (e.g. advancement ⁄ rotational ⁄ transpositional ⁄ pedicle flaps) were created in 34·2% (n = 50) of these patients, split skin grafts in 20·5% (n = 30), primary closure in 17·8% (n = 26) and wedge excision in 5·5% (n = 8). Partial auriculectomy or a combination of methods was performed in 5·5% (n = 8), full skin graft in 4·8% (n = 7), total auriculectomy in 4·1% (n = 6) and excision with healing by secondary intention occurred in 2·1% (n = 3).

Recurrences appeared in 38 patients (23·6%) during the observation period. These are summarized in Table 3 and were mostly local recurrences (8·7%) or regional lymph node metastases (6·2%) as a first progression of the disease. Satellite ⁄ in-transit metastases (3·7%) appeared more often in conjunction with initial distant metastases (5·0%) or local recurrences (NS).

In particular, preauricular lesions tended to recur. Nine of 19 preauricular lesions developed recurrences, four of these with additional distant metastases. In all, 23 (14·3%) patients developed distant metastases during the course of their follow-up.

There were six preregional recurrences after negative SLNB and one after positive SLNB, but none of the patients died of melanoma-related causes after SLNB during the observation period.

The estimated recurrence-free survival rate was 83% over 3 years and 79% over 5 years.

The following factors were not significant for recurrence-free survival: sex (P = 0·64), ulceration (present vs. absent, P = 0·50), 3D histology (P = 0·26) and one-step vs. multiple-step surgery (P = 0·58).

The following risk factors were significantly correlated with recurrence-free survival in univariate analysis: histological tumour type (SSM, LMM, NM, others: P < 0·0001), level of invasion (level II, level III, ≥ level IV: P < 0·0001), tumour thickness (≤ 1·00mm, 1·01–2·00 mm, > 2·0 mm: P < 0·0001), excision margins (≤ 5 mm, 6–10 mm, 11–15mm, > 15mm: P = 0·035) (Fig. 2), tumour location (front of the ear, preauricular, back of the ear: P = 0·002) (Fig. 3). Additionally, age was a statistically significant predictor (≤ 59·19 vs. > 59·19 years, mean 59·19 years: P = 0·006). Patients younger than 59 years tended to stay free of recurrence for a longer period of time. These factors were evaluated in a Cox proportional hazard model with step-by-step elimination of nonsignificant parameters.

Recurrence-free survival in patients with melanoma of the ear (n = 161) in relation to excision margins (coloured lines) (P = 0·035).

Recurrence-free survival in patients with melanoma of the ear (n = 161) in relation to tumour location (front of the ear, preauricular, back of the ear) (P = 0·002).

Tumour thickness (≤ 1·00 mm, 1·01–2·00 mm, > 2·0 mm: P < 0·0001), tumour location (front of the ear, preauricular, back of the ear: P = 0·011), and the extent of the excision margins (≤ 5 mm, 6–10 mm, 11–15 mm, > 15 mm: P = 0·049) were independent risk factors for recurrence-free survival in multivariate analysis (Cox proportional hazard model, Table 4 ).

The prognosis for patients with stage I and II melanoma of the ear was relatively good. Fourteen of the 161 patients eventually died of melanoma-related causes. The survival rate was 98% over 3 years and 96% over 5 years.

The following factors were not significant for disease-specific survival in univariate analysis: sex (P = 0·75), age (≤ 59·19 vs. > 59·19 years, mean 59·19 years: P = 0·13), ulceration (present vs. absent: P = 0·60), one-step vs. multiple-step surgery (P = 0·66) and the extent of excision margins (≤ 5 mm, 6–10 mm, 11–15 mm, > 15 mm: P = 0·95).

Tumour thickness and Clark invasion level were the only significant risk factors for disease-specific survival in Kaplan-Meier univariate analysis [respectively, P = 0·0076 for thicknesses of ≤ 1·00mm, 1·01–2·00mm, and > 2·0 mm, and P = 0·0082 for levels II, III and IV or more (see Fig. 4)].

Disease-specific survival in patients with melanoma of the ear (n = 161). Tumour thickness groups (< 1·00 mm, 1·01–2·00 mm, > 2·00 mm) (P = 0·0076).

3D histology with initially smaller excision margins was used to treat 22 of 42 patients with LMM (52·4%) (Fig. 5).

(a) Lentigo maligna melanoma at the right ear of a 67-year-old patient (tumour thickness 1·20 mm, level IV). Tumour and safety margin are marked on the skin. (b) Embedding of the tumour margins for 3D histology. (c) Preparation of a pediculed flap from the retroauricular region for (d) reconstruction of the defect on the front of the ear. (e) Result after dissection of the pediculed flap.

Of the 22 LMM patients who underwent micrographic surgery with 3D histology, three died of melanoma-related causes 25, 74 and 93 months after the initial diagnosis of malignant melanoma. In comparison, two of 20 patients who underwent surgery with conventional histology died of melanoma-related causes after 64 and 74 months, respectively.

Twenty-two of the 42 LMM patients underwent surgery with 3D histology. Of these, three developed recurrences (one local, one regional lymph node, one distant). The recurrence-free survival rate remained stable at 90·9% over 3 and 5 years of follow-up. Of the remaining 20 LMM patients who underwent surgery with conventional histology, five developed recurrences (one local, two regional lymph node, two distant). The recurrence-free survival rate was 90% over 3 years and 80% over 5 years (Fig. 6). LMMs removed surgically with 3D histology were thicker (median 0·93mm vs. 0·83 mm), but it was possible to use reduced excision margins (median 5 mm vs. 10 mm) without an increased risk of recurrence.

Recurrence-free survival in patients with lentigo maligna melanoma of the ear (n = 42) in relation to 3D histology (n = 22) vs. conventional histology (n = 20) (P = 0·45). Median tumour thickness 0·93 vs. 0·83 mm. Median excision margin 5 mm vs. 10 mm.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....