胰岛素样生长因子ⅡmRNA结合蛋白3在恶性黑素瘤及良性痣中的表达

目的 探讨胰岛素样生长因子Ⅱ mRNA结合蛋白3(IMP3)在良性痣及黑素瘤组织中的表达,及其在恶性黑素瘤进展及诊断中的作用.方法 用IMP3抗体对28例恶性黑素瘤、8例Spitz痣、6例发育不良性痣和25例良性痣患者的标本组织进行免疫组化研究.结果 28例恶性黑素瘤组织标本中23例IMP3阳性,8例Spitz痣中4例阳性,6例发育不良性痣中2例阳性,25例良性痣均不表达.IMP3在黑素瘤中的表达明显高于Spitz痣及发育不良性痣(P<0.05),侵袭性黑素瘤表达明显高于原位黑素瘤(P<0.01).结论 IMP3可能是良性痣发展至恶性黑素瘤的一个生物学标志,在鉴别黑素瘤和良性痣之间存在一定的价值.     

Significant melanocytic lesions in infancy, childhood, and adolescence

Malignant melanoma is diagnosed yearly in approximately 300 persons under age 20 in the United States. Relatively recent advances in dermatology include the recognition of lesions felt to be potential precursors of malignant melanoma. Small congenital melanocytic nevi, present in 1 per cent of all newborn infants, may have a small but definite potential for developing malignant melanoma. Furthermore, despite inconclusive data, many leading dermatologists now advocate removal of these small congenital lesions. Giant congenital melanocytic nevi, with their strong predilection for undergoing malignant change, are removed surgically at an early age, often in multistaged procedures. Dermabrasion, once felt to have a role in the treatment of giant congenital nevi, does not remove the malignant potential of these lesions. The dysplastic nevus syndrome, recognized in 1976, identifies individuals at increased risk for developing melanoma. Adolescents who have the dysplastic nevus syndrome or who are members of families with the syndrome require close medical supervision and patient education. The benign Spitz nevus, with its histologic similarity to malignant melanoma, continues to challenge the dermatopathologist and clinician. These lesions--the Spitz nevus, dysplastic nevus, congenital melanocytic nevus, and malignant melanoma--must all be actively considered when regarding the many other benign melanocytic lesions found in infancy, childhood, and adolescence.     

Proliferation, apoptosis, and survivin expression in a spectrum of melanocytic nevi

BACKGROUND: Apoptosis is important for maintenance of tissue homeostasis and often dysregulated in cutaneous neoplasms. The apoptosis inhibitor survivin is expressed in melanoma and non-melanoma skin cancers and benign keratinocytic lesions. Its expression has not been studied in melanocytic nevi. OBJECTIVE: We determined the expression pattern of survivin in benign melanocytic nevi in comparison to markers of proliferation and apoptosis. METHODS: Six cases of each of the following melanocytic nevi were retrieved from a dermatopathology archive: compound dysplastic nevus, intradermal nevus, compound nevus, neurotized intradermal nevus, and Spitz nevus. Survivin expression was evaluated by in situ hybridization. Apoptotic and proliferation indices were calculated by counting immunoreactive cells in terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling and proliferating cell nuclear antigen immunostained sections, respectively. RESULTS: All nevi, regardless of histologic type, expressed survivin. Compound melanocytic lesions expressed survivin in both epidermal and dermal compartments. The apoptotic rate was low for dysplastic, compound, and Spitz nevi, and apoptotic cells were not identified in any neurotized nevus. The proliferative index was highest for Spitz nevi, while all other nevi demonstrated rare positive cells. CONCLUSIONS: Survivin is consistently expressed in benign melanocytic lesions, while apoptotic cells are rarely identified, suggesting the dysregulation of apoptotic pathways with the accumulation of cells in these neoplasms.     

Use of DAPI cytofluorometric analysis of cellular DNA content to differentiate Spitz nevus from malignant melanoma

Cellular DNA content was measured for the purpose of differentiating Spitz nevus from malignant melanoma using the cytofluorometric technique. DNA was stained by 4',6-diamidino-2-phenylindole, and measured by microfluorometer. Among 20 Spitz nevi examined, 18 of them showed a diploid DNA distribution histographic pattern similar to that of acquired pigmented nevi. The other two Spitz nevi had a few polyploid cells with the major population of cells containing diploid DNA content. In contrast, all malignant melanomas showed an aneuploid DNA distribution histographic pattern. The DNA index values of cells from Spitz nevi distributed in the similar range to that of acquired pigmented nevi and separated from those of malignant melanomas distributed in a much higher range. Our results suggest that cytofluorometric analysis of cellular DNA content reflects the biologic behavior more sensitively than do conventional clinical or histologic criteria, and that it serves as a useful aid for the differentiation of Spitz nevus from malignant melanoma.     

IMP‐3 expression in melanocytic lesions

Background: Insulin‐like growth factor‐II mRNA‐binding protein 3 (IMP‐3 ), a member of the insulin‐like growth factor mRNA‐binding protein family, is expressed in several human malignancies, including melanomas. However, the expression of IMP‐3 has not been explored in melanoma in situ, various histologic subtypes of invasive melanomas and atypical Spitz tumors. Methods: IMP‐3 immunostain was performed in 157 melanocytic lesions. Results: Nearly all benign (8/8), dysplastic (8/8) and Spitz nevi (8/9) were negative for IMP‐3. Focal IMP‐3 positivity was observed in 5/12 melanoma in situ and 4/15 superficial melanomas (Breslow depth ≤1 mm). Half (10/20) of deep melanomas (Breslow depth >1 mm) and 25/52 metastatic melanomas demonstrated strong IMP‐3 staining. IMP‐3 expression differs significantly between non‐desmoplastic melanomas (superficial and deep) and benign or dysplastic or Spitz nevi (p = 0.0427, respectively). Four of 23 desmoplastic melanomas expressed IMP‐3 , which was significantly different from deep melanomas (p = 0.0109). IMP‐3 stained 7 of 10 atypical Spitz tumors. The difference between atypical Spitz tumors and Spitz nevi was statistically significant (p = 0.0256). Conclusion: A malignant circumstance, such as non‐desmoplastic melanoma or atypical Spitz tumor, can be inferred when IMP‐3 is expressed, suggesting potential diagnostic value of IMP‐3 in melanocytic lesions. Yu L, Xu H, Wasco MJ, Bourne PA, Ma L. IMP‐3 expression in melanocytic lesions.     

Classic and atypical Spitz nevi: review of the literature

Both classic and atypical Spitz nevi are uncommon melanocytic lesions usually presenting in children and adolescents. The classic Spitz nevus typically is benign and has characteristic clinical and histologic features. In contrast, the atypical Spitz nevus has an unknown clinical prognosis, and its clinical and histologic traits are loosely defined. Melanoma can have similar features to both classic and atypical Spitz nevi and must be ruled out in all cases. We review the literature on classic and atypical Spitz nevi, advances in differentiating both types of nevi from melanoma, and treatment options.     

Agminated Spitz Nevi: Case Report and Review of the Literature

Spitz nevi are benign melanocytic lesions with many histologic similarities to malignant melanoma. A case of agminated Spitz nevi on a 2‐year‐old boy's left cheek is reported and 41 other cases of agminated Spitz nevi are reviewed. In this case, two biopsies were performed on two different‐appearing lesions and the results of both biopsies showed Spitz nevi.     

A comparative study of Spitz nevus and nodular malignant melanoma using image analysis cytometry

Image analysis cytometry can be used to estimate both nuclear DNA content and area in tissue sections. Since nodular malignant melanoma and Spitz nevus can show a remarkably similar light microscopic appearance, but may differ in behavior, we studied typical examples of these neoplasms to determine whether cytometric differences existed. Analysis of relative DNA content alone could not discriminate between these 2 entities in the 13 cases that we examined. However, Spitz nevi and nodular melanoma clearly differed in terms of maturation, which we defined as the difference between the mean nuclear size or mean nuclear DNA content of the uppermost and deepest melanocytes in each lesion. Maturation as defined by a decrease in mean nuclear DNA content proved highly significant (p less than 0.005). Only Spitz nevi showed a lesser DNA content in the deepest dermal cells as compared with upper dermal cells, suggesting that some Spitz nevi have an admixture of diploid and hyperdiploid cells in their upper portions, but mostly diploid cells in their deep portions. Only nodular melanoma showed higher mean DNA content in deep dermal cells as opposed to superficial dermal cells, suggesting that some nodular melanomas may either have clones of cells in their deep portions that have higher levels of ploidy, or more cells in the deep portion of melanomas may be in active phases of the cell cycle. Our study suggests that important cytometric differences exist between Spitz nevi and nodular melanoma, and that these could be exploited to develop cytometry into an adjunctive clinical technique.     

Comparison of pHH3, Ki-67, and survivin immunoreactivity in benign and malignant melanocytic lesions

Differentiating malignant melanoma from benign melanocytic lesions can be challenging. We undertook this study to evaluate the use of the immunohistochemical mitosis marker phospho-Histone H3 (pHH3) and the proliferation markers Ki-67 and survivin in separating malignant melanoma from benign nevi. Sixty-six melanocytic lesions (18 malignant melanomas, 8 Spitz nevi, 20 dysplastic nevi, and 20 compound nevi) were stained with antibodies to pHH3, Ki-67, and survivin. No pHH3 expression was detected in the dermis of compound and dysplastic nevi. Rare mitoses were observed in the superficial dermis in 3 of 8 Spitz nevi (37%). Staining for pHH3 was higher in malignant melanomas [average 25 per 10 high-power field (HPF), range 2-75 per 10 HPF] than in Spitz nevi (average 0.5 per 10 HPF, range 0-2 per 10 HPF) and was heterogeneously distributed in the malignant melanomas compared with a superficial dermal location in Spitz nevi. There was no cytoplasmic staining for survivin in any of the 66 melanocytic lesions and no nuclear staining in any of the benign ones. Survivin nuclear staining was present in 12 of 18 cases of malignant melanoma (67%) with an average index of 7% (range 0%-15%). In benign melanocytic lesions, the Ki-67 index was less than 5% (range 0%-4%) and staining was present close to the dermo-epidermal junction compared with an average index of 27% in melanomas (range 5%-50%) and a generally heterogeneous pattern of staining throughout the dermis. pHH3 and Ki-67 can be useful adjuncts to histopathology to separate malignant melanoma from benign nevi. pHH3 is especially useful to highlight mitoses and to rapidly assess the mitotic activity in melanocytic lesions.     

Management of Spitz nevi: a survey of dermatologists in the United States

BACKGROUND: There is no consensus concerning management of Spitz nevi. OBJECTIVE: This study was carried out to ascertain how dermatologists manage Spitz nevi. METHODS: A questionnaire was sent to 997 fellows of the American Academy of Dermatology, 284 pediatric dermatologists, and 27 directors of pigmented-lesion clinics. The results are based on the 381 questionnaires returned. RESULTS: The vast majority of responding dermatologists (93%) recommend biopsies of suspected Spitz nevi. Of this group, 43% recommend total biopsies and 55% recommend partial biopsies; 2% would recommend either total or partial biopsies, depending on the clinical situation. Sixty-nine percent of physicians would completely excise a lesion that was histologically diagnosed as an incompletely removed Spitz nevus. Seventy percent of general dermatologists and 80% of pediatric dermatologists would recommend excision with a 1- to 2-mm margin of normal-appearing skin around a Spitz nevus. Nine percent of general dermatologists would recommend margins of 4 mm or more; however, all pediatric dermatologists surveyed would recommend margins less than 4 mm. Physicians were less likely to monitor patients whose Spitz nevi were completely removed. Three fourths (74%) of respondents believe Spitz nevi are entirely benign, 4% believe they are precursors to melanoma, and 22% are not sure. Seven percent of general dermatologists and 4% of pediatric dermatologists have seen metastatic melanomas arise at sites of lesions initially diagnosed histologically as Spitz nevi; 40% of pigmented-lesion clinic directors have seen such lesions. CONCLUSIONS: We believe that the lack of consensus, both in our survey and in the medical literature, reflects to some extent the lack of certainty in the histologic differentiation of Spitz nevi from melanomas and that concern about melanoma influences management. At the pigmented-lesion clinic of the New York University Skin and Cancer Unit, because of this concern about melanoma, it is usually recommended that Spitz nevi be completely excised.     

Benign melanocytic lesions: Risk markers or precursors of cutaneous melanoma?

Background: The role of benign melanocytic lesions as precursors and not only as risk markers for the development of cutaneous melanoma is controversial.Objective: The purpose of the study was to assess the frequency of the histologic association of benign melanocytic lesions with cutaneous melanoma of a maximum thickness of 1.00 mm. The possibility that the spatial association of benign lesions with melanoma may be co-incidental was also investigated.Methods: The study subjects representing 289 cases of cutaneous melanoma of maximum thickness 1.00 mm (or less) were examined histologically for the presence of an associated benign melanocytic lesion(s), including lentiginous melanocytic proliferation; junctional, compound, or intradermal nevus; dysplastic nevus; and congenital nevus contiguous with or adjacent to the melanoma. The effects of age, tumor thickness, level of invasion, histologic type, and anatomic site on the association of benign melanocytic lesions with melanoma were assessed. In the control subjects 40 basal cell carcinomas and 38 compound nevi (not dysplastic) randomly chosen and matched for age (±1 year) and site (head/neck, trunk, upper and lower limbs) with a melanoma case were examined to assess the proportion of these cases associated with benign lesions compared with the matched melanoma cases.Results: A nevus was associated with melanoma in 51% of cases (n = 147). Of these, 82 (56%) were dysplastic nevi, 61 (41%) were common acquired nevi, and 4 (3%) were congenital nevi. Lentiginous melanocytic proliferation was present in the epidermis adjacent to 219 melanomas (75%) and in 44% of these cases (n = 97) a coexisting nevus was also present.Conclusion: The results of this study lend further support to the concept of common acquired nevi and dysplastic nevi as precursors of cutaneous melanoma. In addition, lesions diagnosed clinically as simple lentigo and solar lentigo may be important as potential precursors of melanoma, particularly in the elderly.     

Cyclin D1 overexpression in Spitz nevi: an immunohistochemical study

The morphologic distinction between Spitz nevus and malignant melanoma can be difficult. Because cyclin D1 has been reported to be overexpressed in malignant melanomas, but not in common acquired nevi, we hypothesized that cyclin D1 might be a useful marker to distinguish Spitz nevi from malignant melanoma. Thus, we assessed for cyclin D1 expression in 11 Spitz nevi (10 compound and 1 intradermal) and 9 malignant melanomas (4 Clark stages I-III and 5 Clark stages IV-V) using an immunohistochemical method and routinely fixed and processed tissues. The cyclin D1 results were arbitrarily divided into three groups: 0% to 10%, >10% to 25%, and >25%. We confirmed the observations reported previously by others that cyclin D1 is expressed in malignant melanomas but not in common acquired nevi. Unexpectedly, a relatively high number of cyclin D1-positive cells (i.e., >10%) was also found in all cases of Spitz nevus. However, unlike malignant melanoma, the cyclin D1 positivity in Spitz nevi was present in a zonal pattern. In other words, the number of cyclin D1-positive cells decreased as the lesion extended more deeply, with the number of positive cells in the reticular dermis being less than that in the papillary dermis. Fluorescence in situ hybridization methods were used to assess amplification of 11q13, the locus harboring the cyclin D1 gene, in four cases of Spitz nevus; all were disomic. Using the antibody MIB-1, we compared cyclin D1 expression to the proliferation rate in Spitz nevi. Despite the high cyclin D1 positivity, all Spitz nevi had a relatively low number of MIB-1-positive cells (mean=3.2%), which was significantly lower than that of malignant melanomas (mean=15.3%) (p < 0.001). Thus, unlike malignant melanoma, there appears to be a dissociation between cyclin D1 overexpression and cell proliferation in Spitz nevi.     

Clinical diagnosis of dysplastic melanocytic nevi: A clinicopathologic correlation

A prospective, community practice-based, clinicopathologic correlation was undertaken in 165 melanocytic nevi excised from a group of forty-three patients, each patient having previously had at least one clinically suspected and histologically confirmed dysplastic melanocytic nevus. Eighty-two percent of seventy-two lesions with histologic evidence of mild dysplasia had been diagnosed correctly as such clinically. The accuracy of clinical diagnosis of moderate dysplasia was low (20%); however, all cases of severe dysplasia with or without in situ melanoma were diagnosed correctly. In 75% of all cases in which dysplasia of any degree was diagnosed clinically, histologic evidence of dysplasia was found. In order to investigate further the clinical features of these nevi, 175 color enlargements of histologically confirmed dysplastic melanocytic nevi were examined. The following clinical features were found to be most common: ill-defined border (90%), irregularly distributed pigmentation (84%), maximum diameter greater than 5.0 mm (72%), erythema (64%), and accentuated skin markings (63%). Increasing darkness and confluence of pigmentation in these dysplastic melanocytic nevi correlated with increasing severity of dysplasia. We conclude that careful clinical examination of individual melanocytic nevi will separate severe dysplasia with or without in situ melanoma from low-grade (mild or moderate) dysplasia in a high percentage of nevi from patients with the dysplastic nevus syndrome. Clinical examination will yield a diagnosis of dysplasia in approximately 75% of nevi from such patients in whom histologic evidence of dysplasia is present. Clinical examination constitutes a practical and sufficiently reliable method for the assessment of melanocytic nevi in patients with the dysplastic nevus syndrome.     

Dysplastic nevus in histologic contiguity with acquired nonfamilial melanoma. Clinicopathologic experience in a 100-bed hospital

In the face of alarming rates of increase in melanoma worldwide, dysplastic nevi, especially any that are clinically changing in size, color, or borders, may be regarded as playing a potential role in the progression to a tumor stage. Dysplastic nevi are known to occur in multiples in family members of heritable malignant melanoma. Intraepidermal atypical melanocytes fulfilling the criteria of the dysplastic nevus were seen in histologic contiguity with superficial malignant nonfamilial melanomas in six of 13 patients. With one exception, all melanomas in this study that were associated with histologically contiguous dysplastic nevi were relatively thin, allowing identification of the melanoma at a potentially curable stage.     

Spitz nevus is relatively frequent in adults: a clinico-pathologic study of 247 cases related to patient's age

Spitz nevus is a clinico-pathologic entity that can cause diagnostic concern, particularly in adults. Many studies have been performed to establish reliable histologic criteria, in the attempt to differentiate this lesion from melanoma. A series of 247 Spitz nevi, 6 of which were formerly classified as melanomas, were reviewed for clinical and histopathological parameters. Patients older than 20 comprised 66% of cases, with a predominance of women. The lower extremity was more affected in females of any age, whereas the trunk was more frequently involved in men over 40. Histopathologic examination showed the following differences among Spitz nevi related to age: acanthosis, parakeratosis, pagetoid infiltration, and Kamino bodies were more frequent in young people, whereas multinucleated melanocytes were more frequent in adults. The latter also had lesions that were less pigmented, with less maturation and more desmoplasia. At a mean follow-up of 94 months (range 52-172), recurrence at the site of biopsy or metastases were absent. In our study, a greater proportion of Spitz nevi occurred in adults than in previous series. Moreover, the relative incidence of Spitz nevus compared with melanoma in our population was higher than in other studies. Histopathologic criteria elaborated to diagnose Spitz nevus, applied to our cases, appeared reliable, allowing a correct diagnosis, even in adults.     

Cdc7 expression in melanomas, Spitz tumors and melanocytic nevi

Background:  Cdc7 is a serine-threonine kinase required for initiation of DNA replication that may play a role in the development and progression of melanoma.
Materials and Methods:  Tissue microarrays containing 40 melanomas, 40 Spitz tumors and 30 nevi were constructed. Staining for Cdc7 was scored semiquantitatively according to intensity and extent, and the values were converted into composite scores.
Results:  Nodular melanomas, atypical Spitz tumors and superficial spreading melanomas had the highest scores (nodular melanomas, 3.67; atypical Spitz tumors, 2.78 and superficial spreading melanomas, 2.44). Typical Spitz nevi, dysplastic nevi and ordinary nevi had the lowest scores. Cdc7 expression in melanomas differed significantly from non-Spitz nevi (p < 0.001). The difference was also significant when invasive melanomas were compared with dysplastic nevi (p < 0.005) and when invasive melanomas were compared with non-atypical Spitz nevi (p < 0.001). However, there was no significant difference between invasive melanomas and atypical Spitz tumors (p = 0.69) or between dysplastic nevi and ordinary nevi (p = 0.73).
Conclusion:  Cdc7 expression differs significantly among cutaneous melanocytic neoplasms and can be evaluated by routine immunohistochemical methods. The results suggest that differences in Cdc7 expression may account for some of the differences between malignant melanomas and benign melanocytic nevi.     

Multiple agminated Spitz nevi arising on a café au lait macule: review of the literature with contribution of another case

The majority of Spitz nevi are acquired solitary lesions. Multiple Spitz nevi are rare and may develop on hyperpigmented skin. We report a 16-year-old girl with multiple Spitz nevi arranged on a café au lait macule. Immunohistochemistry showed positivity for S-100 and HMB-45. Of interest, expression of Polo-like kinase (PLK), a novel proliferation marker that recently proved to be positive in up to 98% of malignant melanoma cells, showed positivity in 40% of the nevus cells. The clinical development of multiple Spitz nevi is not yet clear, as they are preferentially excised. Reviewing 70 cases in the literature we found that multiple agminated Spitz nevi occur more frequently than reported previously. In about one-third of these cases Spitz nevi arose on congenitally hyperpigmented skin.     

The “dysplastic” nevus

Dysplastic nevi have become an increasing focus clinically, with evidence that they are associated with a higher risk of developing melanoma. However, there still is contention regarding the significance of dysplastic nevi. This contribution provides an overview of the history, epidemiology, genetics, clinical and histologic features, and procedures for clinical management of dysplastic nevi. Since dysplastic nevi were described originally in 1978, a great deal of research has examined the epidemiology of these lesions and the genetic factors related to the development of dysplastic nevi. However, there is disagreement regarding the clinical management of dysplastic nevi and the histologic definition of dysplastic nevi. Current recommendations include preventative measures, such as sun protection and careful surveillance and biopsies of suspicious lesions as needed. The advent of new technologies, such as computer-vision systems, have the potential to significantly change treatment of dysplastic nevi in the future.     

WT1 marker is not sufficient for distinguishing between melanoma and melanocytic nevi

Background: The heterogeneous histological features of melanoma may often overlap with melanocytic nevi. For this reason, pathologists have sought after immunohistochemistry to assist with difficult cases. Recently, Wilms' tumor 1 protein (WT1) has been suggested to differentiate between melanoma and melanocytic nevi.
Objective: Our objective was to determine whether immunohistochemistry analysis of WT1 expression is a reliable tool in differentiating cutaneous melanoma from melanocytic nevi.
Methods: Forty-five melanoma and 43 melanocytic nevi were immunostained with anti-WT1 monoclonal antibody (clone 6F-H2).
Results: Forty of the 45 cutaneous melanoma (89%) and 22 of the 43 melanocytic nevi (51%) stained (> 10% cells) for WT1. The highest sensitivity for WT1 was expressed by nodular melanoma (19/20), superficial spreading melanoma (8/10) and Spitz nevi (9/11). At the threshold of above 75% WT1-stained cells, the specificity for melanoma was 95% but the sensitivity was only 31%. At the threshold of 10%, the sensitivity increased to 89% but the specificity decreased to only 49%. Finally, at the threshold of 25% and 50%, the sensitivity and specificity were 71%, 61% and 64%, 77%, respectively.
Conclusions: Our data suggest that melanoma is associated with increased WT1 expression. However, as a single immunostaining marker, WT1 is not sufficient for distinguishing melanoma from melanocytic nevi.     

Consumption of the epidermis: a criterion in the differential diagnosis of melanoma and dysplastic nevi that is associated with increasing breslow depth and ulceration

Consumption of the epidermis (COE), defined as thinning of the epidermis with attenuation of basal and suprabasal layers and loss of rete ridges adjacent to collections of melanocytes, is a recently coined term encompassing changes of the epidermal architecture associated with melanoma. To evaluate this feature as an additional diagnostic criterion for melanoma, we examined COE in 453 melanocytic lesions, including 213 invasive melanomas from a population-based series and 240 suspicious pigmented lesions from a clinic-based series, excluding halo and Spitz nevi. In the population-based series, COE was identified in 92/213 (43%) invasive melanomas and became progressively more frequent with increasing Breslow depth (P < 0.0001) and Clark level (P = 0.0002). COE was more frequent when mitotic figures (P < 0.0001), ulceration (P = 0.005), or vertical growth phase (P = 0.009) were present, but it was not significantly associated with age, gender, site, regression, or tumor-infiltrating lymphocytes. In the clinic-based series of pigmented lesions, COE was present in 2/25 (8%) in situ melanomas, 1/29 (3%) lesions classified as melanoma in situ/high-grade dysplastic nevi, and 1/40 (2.5%) high-grade dysplastic nevi. COE was not identified in 146 low-grade dysplastic, congenital, or common nevi. In the combined datasets, 94/96 (98%) lesions exhibiting COE were classified as melanoma. This study demonstrates that COE is frequently present in invasive melanomas, is associated with more aggressive histopathologic features (including increased Breslow depth and ulceration) and may be a useful supplementary diagnostic criterion for melanoma. Furthermore, the process leading to COE may be the first step in a progression to ulceration.