Abstract

Context.—Histologic analysis allows accurate classification of most melanocytic lesions as benign or malignant. Only in a minority of lesions is it necessary to use other techniques as an aid in the diagnosis. Among them, most authors recommend immunohistochemistry.

Objective.—To describe how to apply immunohistochemistry to particular differential diagnoses and the potential pitfalls.

Data Sources.—Personal experience and review of literature.

Conclusions.—here is no single marker, or combination thereof, that establishes an unequivocal diagnosis of melanoma or nevus. Thus it is necessary to carefully analyze the pattern of expression (patchy versus diffuse) and localization (maturation) in the context of morphologic standard features.

ANTIBODIES COMMONLY USED IN DERMATOPATHOLOGY

S100 Protein

More than 95% of primary cutaneous melanomas express S100. However, several conditions may affect its expression, such as too much or too little fixation time, previously frozen tissue, and enzymatic pretreatment with trypsin (see also troubleshooting).

There are 2 types of antibodies, polyclonal and monoclonal. Both label melanocytic lesions, with a cytoplasmic and nuclear pattern. Of the several antigens detected by anti-S100, A6 is expressed by some melanocytic lesions and is also a very helpful marker to detect neurothekeomas.1,2 

gp100 (As Detected With the Antibody HMB-45)

gp100 is fairly specific for melanocytic differentiation; rarely do other lesions express this marker (angiomyolipoma, sugar cell tumor of the lung, and so-called pecoma). HMB-45 is particularly helpful in detecting the pattern of “maturation” of nevi. Thus, superficial, type A melanocytes (epithelioid cells, intraepidermal or close to the epithelium, and mostly pigmented) express neuronal markers and gp100, while the deeply located type C melanocytes (spindle cells) express schwannian markers.3,4 The only exceptions to this rule include blue nevi and related lesions (eg, deep-penetrating nevi) and some Spitz nevi, in which the entire lesion is labeled with HMB-45.

In contrast to nevi, primary cutaneous melanomas usually express gp100 in a patchy pattern, with isolated or clustered cells throughout the dermis; such pattern is also seen in nevoid melanoma.5 By also labeling the intraepidermal component, HMB-45 highlights a single-cell pattern of growth or pagetoid upward migration.

Melanoma Antigen Recognized by T Cells–1

It is one of the most important melanocytic markers is melanoma antigen recognized by T cells–1 (MART1).6 Detected by 2 different antibodies (Melan-A and A-103), this antigen is expressed by most melanocytic lesions, benign and malignant. Therefore, it is very helpful in detecting melanocytic differentiation.7 Conversely, strong and diffuse expression of this marker in a spindle cell melanocytic lesion makes a diagnosis of desmoplastic melanoma unlikely.

Other cells may also express this marker. In particular, steroid-producing tumors may react with A-103. Similarly to gp100, angiomyolipoma, sugar cell tumor of lung, lymphangioleiomyomatosis, and pecoma also express MART1.8 Since the antibody is so sensitive, in sun-exposed skin the labeling of the cell processes may give the appearance that more melanocytes are present than normal, thus raising the consideration of melanoma in situ9 (see below). On occasion, macrophages (particularly pigmented ones) are labeled with anti-MART1.10 

MIB1 (anti–Ki-67)

MIB1 is a proliferation marker expressed in proliferating cells. Its pattern of expression, similar to that of gp100, highlights the presence or absence of “maturation.” Common nevi and dysplastic nevi exhibit reactivity in fewer than 1% of cells, generally disposed at the dermal-epidermal junction or in the more superficial dermal compartment. In contrast, melanomas have a random pattern of immunoreactivity, with a mean proliferative fraction of more than 10%, particularly at the deep edge of the lesion.11 Similarly, desmoplastic melanomas have a much higher proliferation rate, as detected with MIB1, than do desmoplastic nevi.12 

Tyrosinase

Tyrosinase is an enzyme that participates in melanogenesis and is therefore fairly specific for melanocytic differentiation. In our hands, tyrosinase expression is very similar to HMB-45 labeling.

Microphthalmia Transcription Factor

Microphthalmia transcription factor (MiTF) is a nuclear protein involved in the development of melanocytes and the regulation of melanin synthesis in melanocytic lesions13,14; it may be expressed by macrophages, lymphocytes, fibroblasts, Schwann cells, and smooth muscle cells. Owing to its nuclear pattern of expression, we find that anti-MiTF is very helpful when quantifying the number of intraepidermal melanocytes in areas of pigmented epidermis.

To increase the sensitivity of immunohistochemistry, several cocktails have been developed to include more than 1 antibody. A popular combination is designated “pan-melanocytic cocktail” and consists of HMB-45, anti-MART1, and anti-tyrosinase. Analogously, at our institution, we have developed a cocktail including anti-MART1 and anti–Ki-67. Since MART1 is a cytoplasmic marker and Ki-67 is a nuclear marker, by using 2 different systems (eg, ABC and alkaline phosphatase) with 2 different chromogens (eg, diaminobenzidine [brown] and aminoethylcarbazole [red]), it is relatively easy to identify which cells express both markers, thus representing the fraction of melanocytes that are proliferating.

Other antibodies/antigens that may become more popular in the future include NKI-C3,15 p16,16 galectin-3,17 COX-2,18 TRP,19 survivin,20 and claudin-1.21 

USE OF IMMUNOHISTOCHEMISTRY IN SPECIFIC DIFFERENTIAL DIAGNOSES

For the analysis of sentinel lymph nodes, please see the corresponding section.

1. Distinction Between Melanoma and Nevus

The distinction between melanoma and nevus is shown in Figures 1, A through J; 2, A through K; 3, A through H; and 4, A through C.

Most nevi show a pattern of maturation, that is, a change in expression of several immunohistochemical markers from the top to the bottom of the lesion (with the exception of Spitz nevi and blue nevi and related nevi). In our experience, the 2 most helpful markers to detect the pattern of maturation appear to be HBM-45 antigen (gp100) and Ki-67. Both markers are predominantly expressed in those melanocytes located in the epidermis/adnexa and the periepithelial dermis (papillary and adventitial). Therefore, a pattern in which HMB-45 antigen and Ki-67 are expressed in the intraepithelial and periepithelial components but are almost completely absent from the deep areas of the lesion is more consistent with a nevus than with a melanoma (Figure 1, G through J). In our practice, rather than actually counting the number of cells that are positive for Ki-67, we prefer to compare the top and the bottom of the lesion, since all nevi should have many fewer cells at the base of the lesion than the top areas (intraepithelial and periepithelial), regardless of the amount of positive cells.

The 2 types of nevi that diverge from this pattern are blue nevi (including cellular blue, plexiform, and “deep-penetrating” nevi) and Spitz nevi, which may show diffuse labeling with HMB-45 throughout the lesion (Figure 2, F through H). However, as with common nevi, these lesions show a very low proliferation rate with anti–Ki-67 (Figure 2, I through K).

The use of these 2 immunohistochemical markers may help also in the distinction between the invasive component of melanoma and an associated nevus. In contrast with the associated dermal nevus, the invasive component of melanoma is usually positive with HMB-45 and has a higher rate of Ki-67–positive cells. Furthermore, by examining the pattern of expression of gp100 in a blue nevus–type lesion, it may be possible to distinguish a primary blue nevus–type melanoma from a metastatic melanoma. This type of melanoma resembling blue nevus (so-called malignant blue nevus) sometimes arises in association with a blue nevus. In such lesions, gp100 is strongly and diffusely expressed in the benign, preexistent blue nevus, while its expression becomes patchy in the malignant areas (Figure 2, F through H).

2. Analysis of Melanocytic Desmoplastic Lesions

Melanocytic desmoplastic lesions are shown in Figures 3 and 4.

Desmoplastic melanomas usually display high numbers of Ki-67–positive cells and may show rare cells labeled with HMB-4512 (Figure 3, E through H). Another marker that may be helpful in distinguishing desmoplastic nevus from desmoplastic melanoma is MART1. As mentioned above, most melanocytic lesions express this marker, with the notable exception of spindle cell melanoma. Therefore, a spindle cell melanocytic lesion that does not express MART1 is more likely to be a melanoma than a nevus.

Immunohistochemistry is also helpful in the diagnosis of desmoplastic melanoma, since the tumor cells will express S100 protein. Furthermore, anti-S100 may help in delineating the extent of the lesion to determine the depth of invasion. In the differential diagnosis, even though scar tissue shows the presence of scattered S100-positive cells,10,22 when compared with scars, desmoplastic melanomas will have many more positive cells for this marker (Figure 4, B and C). Another marker typically expressed in desmoplastic melanoma and not in scars is p75.23 

3. Diagnosis of Lentigo Maligna

Lentigo maligna is shown in Figure 5, A through F.

Anti-MART1 or HMB-45 will label the atypical melanocytes of lentigo maligna (melanoma in situ) in the epidermis, highlighting a confluent pattern, in contrast to the scattered, atypical melanocytes that can be seen in actinically damaged skin/pigmented actinic keratosis. Since anti-MART1 extensively labels the melanocyte dendrites, thus mimicking a confluent pattern of growth (Figure 5, C and D), some authors discourage the use of anti-MART1.24 However, addition of other antibodies, such as HMB-4525 or anti-MiTF, may be helpful since these markers label the cytoplasm of the melanocyte body or only the nuclei, respectively (Figure 5, E and F).

SUMMARY

It is our opinion that immunohistochemistry has an important role in the diagnosis of melanocytic lesions. However, there is no single marker, or combination thereof, that establishes an unequivocal diagnosis of melanoma or nevus. Thus it is necessary to carefully analyze the pattern of expression (patchy versus diffuse) and localization (maturation) in the context of morphologic standard features.

References

References
1
McNutt
,
N. S.
The S100 family of multipurpose calcium-binding proteins.
J Cutan Pathol
1998
.
25
(
10
):
521
529
.
2
Fullen
,
D. R.
,
L.
Lowe
, and
L. D.
Su
.
Antibody to S100a6 protein is a sensitive immunohistochemical marker for neurothekeoma.
J Cutan Pathol
2003
.
30
(
2
):
118
122
.
3
Prieto
,
V. G.
,
N. S.
McNutt
,
J.
Lugo
, and
J. A.
Reed
.
The intermediate filament peripherin is expressed in cutaneous melanocytic lesions.
J Cutan Pathol
1997
.
24
(
3
):
145
150
.
4
Huttenbach
,
Y.
,
V. G.
Prieto
, and
J. A.
Reed
.
Desmoplastic and spindle cell melanomas express protein markers of the neural crest but not of later committed stages of Schwann cell differentiation.
J Cutan Pathol
2002
.
29
(
9
):
562
568
.
5
McNutt
,
N. S.
,
C.
Urmacher
,
J.
Hakimian
,
D. M.
Hoss
, and
J.
Lugo
.
Nevoid malignant melanoma: morphologic patterns and immunohistochemical reactivity.
J Cutan Pathol
1995
.
22
:
502
517
.
6
Busam
,
K. J.
,
Y. T.
Chen
,
L. J.
Old
, et al
.
Expression of melan-A (MART1) in benign melanocytic nevi and primary cutaneous malignant melanoma.
Am J Surg Pathol
1998
.
22
(
8
):
976
982
.
7
Jungbluth
,
A. A.
,
K. J.
Busam
,
W. L.
Gerald
, et al
.
A103: an anti-melan-a monoclonal antibody for the detection of malignant melanoma in paraffin-embedded tissues.
Am J Surg Pathol
1998
.
22
(
5
):
595
602
.
8
Fetsch
,
P. A.
,
J. F.
Fetsch
,
F. M.
Marincola
,
W.
Travis
,
K. P.
Batts
, and
A.
Abati
.
Comparison of melanoma antigen recognized by T cells (MART-1) to HMB-45: additional evidence to support a common lineage for angiomyolipoma, lymphangiomyomatosis, and clear cell sugar tumor.
Mod Pathol
1998
.
11
(
8
):
699
703
.
9
El Shabrawi-Caelen
,
L.
,
H.
Kerl
, and
L.
Cerroni
.
Melan-A: not a helpful marker in distinction between melanoma in situ on sun-damaged skin and pigmented actinic keratosis.
Am J Dermatopathol
2004
.
26
(
5
):
364
366
.
10
Trejo
,
O.
,
J. A.
Reed
, and
V. G.
Prieto
.
Atypical cells in human cutaneous re-excision scars for melanoma express p75NGFR, C56/N-CAM and GAP-43: evidence of early Schwann cell differentiation.
J Cutan Pathol
2002
.
29
(
7
):
397
406
.
11
Rudolph
,
P.
,
C.
Schubert
,
B.
Schubert
, and
R.
Parwaresch
.
Proliferation marker Ki-S5 as a diagnostic tool in melanocytic lesions.
J Am Acad Dermatol
1997
.
37
(
2, pt 1
):
169
178
.
12
Harris
,
G. R.
,
C. R.
Shea
,
M. G.
Horenstein
,
J. A.
Reed
,
J. L.
Burchette
Jr
, and
V. G.
Prieto
.
Desmoplastic (sclerotic) nevus: an underrecognized entity that resembles dermatofibroma and desmoplastic melanoma.
Am J Surg Pathol
1999
.
23
(
7
):
786
794
.
13
King
,
R.
,
P. B.
Googe
,
K. N.
Weilbaecher
,
M. C.
Mihm
Jr
, and
D. E.
Fisher
.
Microphthalmia transcription factor expression in cutaneous benign, malignant melanocytic, and nonmelanocytic tumors.
Am J Surg Pathol
2001
.
25
(
1
):
51
57
.
14
Koch
,
M. B.
,
I. M.
Shih
,
S. W.
Weiss
, and
A. L.
Folpe
.
Microphthalmia transcription factor and melanoma cell adhesion molecule expression distinguish desmoplastic/spindle cell melanoma from morphologic mimics.
Am J Surg Pathol
2001
.
25
(
1
):
58
64
.
15
Paul
,
E.
,
A. J.
Cochran
, and
D. R.
Wen
.
Immunohistochemical demonstration of S-100 protein and melanoma-associated antigens in melanocytic nevi.
J Cutan Pathol
1988
.
15
(
3
):
161
165
.
16
Reed
,
J. A.
,
F.
Loganzo
Jr
,
C. R.
Shea
, et al
.
Loss of expression of the p16/cyclin-dependent kinase inhibitor 2 tumor suppressor gene in melanocytic lesions correlates with invasive stage of tumor progression.
Cancer Res
1995
.
55
(
13
):
2713
2718
.
17
Prieto
,
V. G.
,
A. A.
Mourad-Zeidan
,
V.
Melnikova
, et al
.
Galectin-3 expression is associated with tumor progression and pattern of sun exposure in melanoma.
Clin Cancer Res
2006
.
12
(
22
):
6709
6715
.
18
Chwirot
,
B. W.
and
L.
Kuzbicki
.
Cyclooxygenase-2 (COX-2): first immunohistochemical marker distinguishing early cutaneous melanomas from benign melanocytic skin tumours.
Melanoma Res
2007
.
17
(
3
):
139
145
.
19
Hashimoto
,
Y.
,
Y.
Ito
,
T.
Kato
,
T.
Motokawa
,
T.
Katagiri
, and
M.
Itoh
.
Expression profiles of melanogenesis-related genes and proteins in acquired melanocytic nevus.
J Cutan Pathol
2006
.
33
(
3
):
207
215
.
20
Ding
,
Y.
,
V. G.
Prieto
, and
P. S.
Zhang
.
Nuclear expression of the antiapoptotic protein survivin in malignant melanoma.
Cancer
2006
.
105
(
5
):
1123
1129
.
21
Cohn
,
M. L.
,
V. N.
Goncharuk
,
A. H.
Diwan
,
P. S.
Zhang
,
S. S.
Shen
, and
V. G.
Prieto
.
Loss of claudin-1 expression in tumor-associated vessels correlates with acquisition of metastatic phenotype in melanocytic neoplasms.
J Cutan Pathol
2005
.
32
(
8
):
533
536
.
22
Chorny
,
J. A.
and
R. J.
Barr
.
S100-positive spindle cells in scars: a diagnostic pitfall in the re-excision of desmoplastic melanoma.
Am J Dermatopathol
2002
.
24
(
4
):
309
312
.
23
Kanik
,
A. B.
,
M.
Yaar
, and
J.
Bhawan
.
p75 nerve growth factor receptor staining helps identify desmoplastic and neurotropic melanoma.
J Cutan Pathol
1996
.
23
(
3
):
205
210
.
24
Yan
,
S.
and
J. B.
Brennick
.
False-positive rate of the immunoperoxidase stains for MART1/MelanA in lymph nodes.
Am J Surg Pathol
2004
.
28
(
5
):
596
600
.
25
Wiltz
,
K. L.
,
H.
Qureshi
,
J. W.
Patterson
,
D. C.
Mayes
, and
M. R.
Wick
.
Immunostaining for MART-1 in the interpretation of problematic intra-epidermal pigmented lesions.
J Cutan Pathol
2007
.
34
(
8
):
601
605
.

Author notes

From the Departments of Pathology and Dermatology, University of Texas M. D. Anderson Cancer Center, Houston (Dr Prieto); and Section of Dermatology, University of Chicago, Chicago, Illinois (Dr Shea).

The authors have no relevant financial interest in the products or companies described in this article.