The Cutting Edge
January 1999

Mohs Micrographic Excision of Lentigo Maligna Using Mel-5 for Margin Control

Author Affiliations



Copyright 1999 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1999

Arch Dermatol. 1999;135(1):15-17. doi:10.1001/archderm.135.1.15

A 43-year-old man presented with a pigmented lesion on the left earlobe. Histologic evaluation of a biopsy specimen demonstrated lentigo maligna. The patient was referred to our institution for micrographic excision of the lesion.


A 56-year-old man presented with a pigmented lesion on the right preauricular cheek. A biopsy specimen demonstrated lentigo maligna. The lesion was excised in a standard fashion with 0.5-cm margins. Histologic evaluation of the excised specimen showed persistence of the tumor at the surgical margin. A second attempt at excision failed to fully excise the tumor. The patient was referred to our institution for micrographic excision of the residual tumor.


Successful micrographic excision of lentigo maligna is hindered by difficulty interpreting frozen sections stained with hematoxylin-eosin (HE).


The lentigo maligna was excised using Mohs micrographic excision starting with 0.5-cm margins. In addition to HE staining, frozen sections were stained with the immunohistochemical stain Mel-5. To differentiate true lentigo maligna from surrounding sun-damaged skin, a sample was obtained from the contralateral earlobe to serve as a negative control. A sample from the center of the tumor served as a positive control. Representative sections are shown in Figure 1. One stage was required to clear the tumor. The patient remained tumor free for 15 months.

Case 1.
Image not available

Left column, Melanocytes are difficult to distinguish (hematoxylin-eosin, ×200). Right column, Melanocytes are readily identified (Mel-5, ×200). Top row, surgical margin. Middle row, Positive control from center of surgical specimen. Bottom row, Negative control from contralateral body site.


The residual tumor was excised using Mohs micrographic excision, starting with 0.5-cm margins. In addition to HE, frozen sections were stained with Mel-5. Positive and negative controls were also obtained as described above. One stage was required to clear the tumor. The patient has remained tumor free for 16 months.


Opinion is divided among dermatologic surgeons as to whether micrographic excision of melanoma is appropriate management,15 particularly whether interpretation of frozen sections stained with HE is reliable. For micrographic excision of melanoma to be successful and practical, the following 4 criteria must be met: (1) the tumor cells must be visually identifiable in the sections; (2) the tumor must be contiguous to avoid false negative-margins; (3) the mapping and staining component must be technically feasible, preferably in the surgeon's own laboratory; and (4) the total tissue processing time should be short enough to allow for a staged excision and repair on the same day.

Most surgeons who advocate micrographic excision of melanoma use HE staining of frozen sections because the technique is simple and familiar.3,4 However, those who oppose micrographic excision of melanoma feel it is difficult to accurately visualize the tumor using HE, especially at the tumor margins where single cells tend to predominate over nests of cells.2 Some authors advocate standard excision or staged excision with "rush permanent" paraffin-embedded sections.2,6 In paraffin-embedded sections, the clues to identifying melanocytes include the retraction artifact, which leaves a vacuolated space between the melanocyte and the surrounding keratinocytes, and the nuclear structure.

No stain can reliably differentiate between melanoma cells and benign melanocytes; however, one that reliably stains all melanocytes and the epidermal components of melanoma allows for diagnosis based on pattern recognition.

Mel-5 is a murine IgG antibody against a 75-kd protein that may be tyrosine-related protein (TRP-1).79 The antigen appears to be particularly expressed in association with stage III and IV melanosomes.8 Epidermal melanocytes and the epidermal component of nevi and of melanomas are readily stained.10,11 The stain may be performed on frozen sections, as well as paraffin-embedded tissue. Our staining protocol (Table 1) requires 75 minutes, thus allowing for a multistaged excision and repair in the same day. The technique is simple and requires little extra equipment other than the antibody and a detection kit. Melanocytes stained with Mel-5 appear as heterochromatic dendritic cells. Compared with other immunohistochemical stains, Mel-5 is superior for this purpose because it so reliably stains epidermal melanocytes and melanoma cells.10 We feel that the best indication for micrographic excision with Mel-5 is for lentigo maligna, because of the superior ability to visualize the tumor and the irregular melanocytes at the margin. In addition, Mel-5 is indicated for micrographic excision of invasive melanoma to allow narrower margins in functionally or cosmetically critical locations. It is important to note that Mel-5 may not reliably stain amelanotic melanoma, desmoplastic melanomas, or the dermal component of melanomas.10 In our practice, we start with a 5-mm margin around the visible margin of the tumor as seen under examination with a Wood light. We employ standard micrographic mapping techniques to orient and section the tissue. Finally, we stain our sections with Mel-5, and often with HE as well. We use a section of tumor from the center of the lesion as a positive control, and often take a small biopsy specimen from the skin from the contralateral body site to serve as a negative control.

Image not available
Mel-5 Rapid Staining Protocol for Frozen Sections*

In summary, we have presented 2 cases of lentigo maligna micrographically excised using Mel-5 for margin control. The use of Mel-5 on frozen sections is more effective for visualizing residual tumor and normal melanocytes than HE.

Back to top
Article Information

The authors have no financial involvement in any of the materials used herein. Some of the materials in this study (Mel-5 antibody, ultrastreptavidin detection kit) were provided free of charge by the manufacturer (Signet Laboratories, Dedham, Mass).

Dzubow  LM Mohs surgery and malignant melanoma. Dermatol Surg. 1995;21283
Stonecipher  MRLeshin  BPatrick  JWhite  WL Management of lentigo maligna and lentigo maligna melanoma with parafin-embedded tangential sections: utility of immunoperoxidase staining and supplemental vertical sections. J Am Acad Dermatol. 1993;29589- 594Article
Zitelli  JAMoy  RLAbell  E The reliability of frozen sections in the evaluation of surgical margins for melanoma. J Am Acad Dermatol. 1991;24102- 106Article
Robinson  JK Margin control for lentigo maligna. J Am Acad Dermatol. 1994;3179- 85Article
Brown  CDZitelli  JA The prognosis and treatment of true local cutaneous recurrent malignant melanoma. Dermatol Surg. 1995;21285- 290
Dhawan  SSWolf  DJRabinovitz  HSPoulos  E Lentigo maligna. The use of rush permanent sections in therapy.  Arch Dermatol. 1990;126928- 930Article
Slominski  ACostantino  RHowe  JMoellman  G Molecular mechanisms governing melanogenesis in hamster melanomas: relative abundance of tyrosinase and catalase-b (gp-75). Anticancer Res. 1991;11257- 262
Thomson  TMReal  FXMurakami  SCordon-Cardo  COld  L JHoughton  AN Differentiation antigens of melanocytes and melanoma: analysis of melanosome and cell surface markers of human pigmented cells with monoclonal antibodies. J Invest Dermatol. 1988;90459- 466Article
Roux  LLloyd  KO Glycosylation characteristics of pigmentation-associated antigen (GP75): an intracellular glycoprotein of human melanocytes and malignant melanomas. Arch Biochem Biophys. 1986;25187- 96Article
Bhawan  J Mel-5: a novel antibody for differential diagnosis of epidermal pigmented lesions of the skin in paraffin-embedded sections. Melanoma Res. 1997;743- 48Article
Bhawan  JAndersen  WLee  WLabadie  RSolares  G Photoaging versus intrinsic aging: a morphologic assessment of facial skin. J Cutan Pathol. 1995;22154- 159Article