Primary acquired melanosis of the conjunctiva

Forty-one cases of conjunctival primary acquired melanosis (PAM) were studied to determine the frequency of progression to malignant melanoma and to establish prognostic parameters for progression to melanoma. Two subdivisions were identified: lesions with cytologically atypical melanocytes (PAM with atypia, 28 lesions, 68.3 per cent) and those lacking cytologically atypical melanocytes (PAM without atypia, 13 lesions, 31.7 per cent). None of the lesions of PAM without atypia progressed to melanoma. Thirteen of the 28 lesions of PAM with atypia (46.4 per cent) progressed to melanoma. Progression to melanoma was more frequent in the lesions of PAM with atypia if basilar hyperplasia was not the dominant histologic pattern (90 per cent progression, P = 0.02) or if any epithelioid cells were present (75 per cent progression, P = 0.02). It was not possible to determine which lesions were atypical on the basis of clinical appearance. Lesions at risk for the development of melanoma should be totally extirpated.     

Malignant melanoma of the conjunctiva

One hundred thirty-one cases of conjunctival melanoma in which biopsies had been performed were studied to determine potential factors that might affect outcome in patients with these lesions. Two groups of lesions were identified: those associated with primary acquired melanosis (melanoma with PAM, 98 cases, 74.8 per cent) and those without primary acquired melanosis (melanoma without PAM, 33 cases, 25.2 per cent). The overall mortality rate in the 131 cases was 26 per cent (34 of 131); the mortality rate due to melanoma with PAM was 25.5 per cent (25 of 98), and that due to melanoma without PAM was 27.3 per cent (9 of 33). If PAM was associated with the lesion, the presence of atypical melanocytes within the epithelium (pagetoid invasion) was a sensitive indicator of subsequent metastasis. Tumor thickness may also be useful for predicting subsequent metastases. None of the histologic parameters studied proved useful for predicting outcome in patients who had melanomas without PAM. The presence or absence of nevi had no effect on prognosis.     

Spitz nevi and atypical Spitz nevi/tumors: a histologic and immunohistochemical analysis.

A subset of Spitz nevi poses substantial diagnostic difficulty, even among experts, due to its resemblance to malignant melanoma. These lesions are termed atypical Spitz nevi/tumors and there is currently a lack of objective criteria for predicting their biologic behavior. We compared the expression of Ki-67, p21, and fatty acid synthase by immunohistochemistry in 10 atypical Spitz nevi, 28 typical Spitz nevi, 19 compound melanocytic nevi and 18 invasive malignant melanomas. There was a progressive increase in fatty acid synthase cytoplasmic expression with statistically significant differences observed between Spitz nevi and atypical Spitz nevi (P=0.003) and between atypical Spitz nevi and malignant melanoma (P<0.050). Ki-67 nuclear staining was lower in both typical and atypical forms of Spitz lesions than in malignant melanoma (P<0.001). The degree of P21 nuclear expression in atypical Spitz nevi was not significantly different than in Spitz nevi, but was significantly greater than expression in conventional nevi and approached significance after multiple comparisons corrections for malignant melanoma. Thus, a high level of P21 expression makes a tumor more likely to be a typical or atypical Spitz nevus than a malignant melanoma, especially when coupled with a low Ki-67 index and weak expression of fatty acid synthase. These immunohistochemical observations support the concept that atypical Spitz nevi are distinct lesions of borderline biologic behavior residing between Spitz nevi and malignant melanoma. The study also compared a large array of histologic features of 16 cases of typical Spitz nevi in children with 12 typical Spitz nevi in adults. The adult lesions were significantly more likely to be intradermal and to display dermal fibroplasia, but were histologically similar to their pediatric counterparts in all other respects.     

Immunohistochemical profiling including beta-catenin in conjunctival melanocytic lesions

Conjunctival melanocytic lesions encompass a group of clinically diverse, benign to malignant, neoplasms that may contain overlapping histopathological features, making definitive diagnosis challenging in some cases. In this series, we compared multiple immunohistochemical (IHC) markers in 11 conjunctival nevi, 10 primary acquired melanosis (PAM) lesions, and 11 conjunctival melanomas. Immunostains included the melanocytic markers HMB-45 and Melan-A, as well as the proliferative marker Ki-67. Loss of beta-catenin expression has been associated with more aggressive clinical disease in cutaneous melanoma, but its status in conjunctival melanocytic lesions is not known, therefore we incorporated beta-catenin immunohistochemical staining in our study. In this series, conjunctival melanomas had a higher Ki-67 proliferative index and HMB-45 immunoreactivity than did PAM lesions and conjunctival nevi (P < 0.001). Melan-A was highly expressed in all 3 groups. Beta-catenin was more strongly expressed in melanomas and nevi than in PAM (P < 0.001). There was high inter-grader reliability (Kappa = 0.53). Overall, IHC labeling of HMB-45 and Ki-67 is increased in conjunctival melanomas compared to PAM or conjunctival nevi. Beta-catenin, an IHC marker previously unstudied in conjunctival melanocytic lesions, is not preferentially expressed in benign lesions and may play a different role in conjunctival atypia than it does in cutaneous melanoma.     

p16ink4a expression in benign and malignant melanocytic conjunctival lesions

Acquired conjunctival melanocytic lesions include nevi, primary acquired melanoses (PAMs), and melanomas. Conjunctival melanoma is a malignant melanocytic neoplasm with a high metastasis and mortality rate. Usually, the diagnosis can be achieved only with routine microscopic analysis, but in some cases, the samples are small or have artifacts. In these cases, complementary studies will be helpful, but currently, there are no well-understood or studied complementary methods. Objective. To analyze the immunohistochemical expression of p16 in conjunctival melanocytic lesions and to assess its potential for differentiating between benign and malignant melanocytic lesions. Methods. Immunohistochemical study against p16ink4a (p16) was performed on paraffin-embedded sections on 45 melanocytic lesions (9 melanomas, 19 nevi, and 2 PAMs with atypia and 15 without atypia). Expression was scored according to the German immunoreactive score (IRS). Results. Expression of p16 IRS differed between nevi, PAMs, and melanomas. The mean IRS for melanomas was 3.3 ± 1.8 and was lower than those for nevi (7.63 ± 3.24; P < .05), PAM with atypia (12 ± 0; P < .05), and PAM without atypia (11 ± 1.69; P < .05). Lesions with infiltration depths lower than 2 mm showed higher levels of p16. There were no differences between favorable and unfavorable locations. Conclusion. p16 Expression in conjunctival melanocytic lesions showed an expression similar to that in skin and seems to be a good marker to differentiate nevi and PAMs from melanomas. However, additional studies of larger series and follow-up are needed to confirm these findings.     

Vascular endothelial growth factor expression in malignant melanoma: prognostic versus diagnostic usefulness.

Vascular endothelial growth factor (VEGF), an endothelial cell mitogen, plays a role in angiogenesis and progression in malignant melanoma. VEGF expression was examined in 62 biopsy specimens of melanocytic proliferations, including 45 malignant melanomas, 3 cellular blue nevi, 12 atypical compound nevi, and 2 Spitz nevi. The cases of malignant melanoma included 11 in situ melanomas, 18 Clark Level II, 9 Clark Level III, and 7 Clark Level IV tissue samples. All of the specimens were fixed in formalin and embedded in paraffin. Cytoplasmic immunoreactivity for VEGF was demonstrated in 19 (42%) of 45 melanoma samples, but there was no immunoreactivity for VEGF exhibited by any of the atypical compound melanocytic nevi, cellular blue nevi, or Spitz nevi (P < .009). Immunoreactivity for VEGF was found to be related to tumor thickness (as evidenced by Clark level [P < .03]) and to absence of regression (P < .04). Although VEGF is not a useful prognostic indicator for malignant melanoma, it may be useful as a discriminating factor between malignant melanoma and benign melanocytic lesions, and it may offer some insight into tumor growth.     

Diagnosis and management of premalignant melanocytic proliferations

The histopathology of melanocytic proliferations in human skin can be defined in a way which allows a rational approach to their management. Early and/or premalignant lesions such as melanocytic hypertrophy, hyperplasia, dysplasia, and atypical hyperplasias are correlated with clinical lesions such as lentigo, compound nevoid lentigo, changes in nevi during pregnancy, and unusual moles seen in patients with the dysplastic nevus syndrome. Clinical management of such lesions may be determined from the pathological process. Hypertrophic and hyperplastic lesions need not be re-excised, although partially removed moles showing junctional hyperplasia may recur clinically. The mildly and moderately dysplastic nevus need only be narrowly removed. Severe dysplasia and melanoma in situ may recur locally as invasive melanoma, and consideration for conservative reexcision is warranted. Dysplastic nevi should be considered to be markers of patients who may develop melanoma. Patients with dysplastic nevi or a family history of unusual moles or melanoma should have continued follow-up, preferably with standardized clinical photographs.     

Atypical Spitz nevi/tumors: lack of consensus for diagnosis, discrimination from melanoma, and prediction of outcome.

The biological nature of Spitz nevi/tumors and their diagnostic distinction from, or relationship to, melanoma remain unresolved issues. In this report, a series of 30 melanocytic lesions removed from 28 patients, including atypical Spitz nevi/tumors and metastasizing Spitzoid tumors/melanomas, were evaluated by a panel of dermatopathologists to evaluate interobserver diagnostic concordance and to assess the prognostic power of histological criteria. For inclusion in the study, each lesion had to display some criteria for the Spitz nevus, and in addition one of the following was required: (1) definitive clinical outcome such as metastasis or death of disease, or (2) long-term follow-up if the patient remained disease free. Each lesion was reviewed independently and blinded as to the clinical data by 10 pathologists, who categorized them as (1) typical Spitz nevus/tumor, (2) atypical Spitz nevus/tumor, (3) melanoma, (4) tumor with unknown biological potential, or (5) other melanocytic lesion. There was limited discussion of criteria before the review. Evaluation of 17 Spitzoid lesions yielded no clear consensus as to diagnosis; in only one case did six or more pathologists agree on a single category, regardless of clinical outcome. Notably, however, some lesions that proved fatal were categorized by most observers as either Spitz nevi or atypical Spitz tumors. Conversely, seven or more pathologists scored 13 lesions as melanoma. These results illustrate (1) substantial diagnostic difficulties posed by many Spitz tumors, especially those with atypical features, even among experts, and (2) the lack of objective criteria for their distinction from melanoma and for gauging their malignant potential. Nevertheless, our observations do suggest that a biological relationship exists between the Spitz nevus/tumor and melanoma.     

Pigmented melanocytic lesions of the conjunctiva--a new approach to their classification

We report a series of 23 malignant melanomas and 24 cases of benign and premalignant lesions of the conjunctiva and propose a new classification of non-nevoid pigmented benign, premalignant and malignant melanocytic lesions. The question of prognostic features in conjunctival melanoma is briefly discussed.     

Proliferative activity of primary cutaneous melanocytic tumours

The expression of proliferating cell nuclear antigen (PCNA) was examined in formalin-fixed paraffin-embedded tissue sections from 41 lesions (27 melanocytic nevi, 3 atypical nevi and 11 malignant melanomas) to determine the proliferative activity of primary cutaneous melanocytic tumours. Most of the malignant melanomas had more than 7% PCNA-positive cells (9.2±0.5%), while the melanocytic nevi manifested less than 1% PCNA-positive cells (0.4±0.1%). Atypical nevi exhibited an intermediate, but still significantly lower, labelling ratio when compared with malignant melanomas (0.8±0.2%). The proliferative activity of the lesions was compared between portions at different depths in the skin (epidermal, upper dermal and lower dermal location). In cases of malignant melanoma, the proliferative activity was higher in the deeper portion of dermis whereas PCNA-positive cells in melanocytic nevi were located in the upper dermis predominantly. Thus the PCNA labelling ratio of malignant melanoma and/or melanocytic nevus cells located in the epidermodermal junction was not necessarily higher than that of malignant melanoma and/or melanocytic nevus cells in the dermis. These results indicate that staining with PCNA would be very useful in the differentiation of malignant melanoma from melanocytic nevi manifesting cellular and/or structural atypia by virtue of a significant difference in the proportion of PCNA-positive cells. Although malignant melanomas have higher proliferative activity than melanocytic nevi in the deeper dermis, junctional activity in melanocytic tumours does not indicate cell proliferation.     

Melanocytic lesions associated with dermatofibromas: a spectrum of lesions ranging from junctional nevus to malignant melanoma in situ.

Dermatofibromas are common lesions that are often associated with epidermal hyperplasia and basal layer hyperpigmentation. A single case of lentiginous melanocytic hyperplasia overlying a dermatofibroma has been reported, however, nevi and melanoma have to the best of our knowledge, not been previously reported. We present 14 cases of melanocytic lesions associated with dermatofibromas. The clinical data and hematoxylin- and eosin- stained sections were obtained and formalin-fixed, paraffin-embedded tissue was immunostained with antibodies against S-100, Mart-1, Factor XIIIa, and CD117. There were nine females and five males ranging in age from 30 to 64 years and anatomic sites included back (five), arm (six), flank (two), and leg (one). The clinical diagnosis ranged from dermatofibroma to desmoplastic melanoma. Histologically, the melanocytic lesions included junctional, compound, and dermal nevi, and malignant melanoma in situ. In four cases the dermal component appeared to merge with the dermatofibroma. In the case of the melanoma in situ, the dermatofibroma abutted the epidermis. Immunohistochemically, the melanocytic lesions were S-100/ Mart-1+, FXIIIa-, and the dermatofibromas were S-100/Mart-1-, FXIIIa+. Melanocytic neoplasia may appear in association with dermatofibromas. The fibrohistiocytic proliferation may be misinterpreted as a spindle or pleomorphic melanocytic process. Awareness of this association will aid in the correct diagnosis, and immunohistochemical studies will help in the differentiation of these two cell populations.     

Malignant melanoma in situ

There is general agreement that there are lesions which should be diagnosed as malignant melanoma in situ. The biologic behavior of the intraepidermal component of superficial spreading melanoma (the radial growth phase) has been demonstrated to have significantly different properties than those of the cells in the vertical growth phase. The central controversies regarding malignant melanoma in situ relate to the criteria for diagnosing melanoma and whether there are atypical melanocytic lesions which are neither nevi nor melanoma. I propose that reliable and reproducible objective criteria for melanoma have not been developed; rather, dermatopathologists use differing subjective criteria which result in diagnostic agreement in the vast majority of lesions. The controversy is over those lesions which are classified as melanoma by one set of criteria and not by others. I am also a proponent of the diagnosis of atypical melanocytic lesions, and have suggested an analogy to keratinocytic lesions in which this concept is well established. Studies of the sensitivity and specificity of the correlation between clinical and histologic features and the biologic behavior of melanocytic lesions are necessary to resolve these issues.     

Congenital melanocytic nevi of the small and garment type. Clinical, histologic, and ultrastructural studies

The clinical and histologic features of 60 congenital compound and dermalnevi were reviewed and compared to those of 60 acquired compound and dermal nevi. Clinically the two groups could be distinguished by size, most congenital nevi being larger than 1.5 cm. and most acquired nevi smaller than 1.5 cm. in diameter. In addition, congenital nevi commonly are different in coloration and surface topography from acquired nevi. Histologically the congenital nevi had three diagnostic features that separated them from acquired nevi: (1) Nevus cells were present in the lower two thirds of the reticular layer of the dermis in 59 of the 60 cases and in the subcutis as well in 35 cases. (2) Nevus cells were disposed between collagen bundles singly or in Indian files, or both, in all cases. (3) Nevus cells commonly involved appendages, nerves, and vessels in the lower two thirds of the reticular dermis or subcutis and occasionally involved more than one of these structures in the same nevus. Of the 60 compound and dermal nevi present at birth, six were garment (giant hairy) nevi, and their histologic characteristics were similar to those of the other congenital compound and dermal nevi. In light of the increased incidence of malignant melanoma in garment nevi, the distinctive nature of congenital nevi should be borne in mind in order to determine whether other congenital nevi of a smaller size may share in the propensity for malignant degeneration. Although it was not a formal part of the series of cases analyzed in this report, we have observed one case of malignant melanoma that arose in a relatively small congenital nevus.     

Differentiation of malignant melanoma from benign nevus using a novel genomic microarray with low specimen requirements

Context.-Histologic examination of clinically suspicious melanocytic lesions is very sensitive and specific for the detection of malignant melanoma. Yet, the malignant potential of a small percentage of melanocytic lesions remains histologically uncertain. Molecular testing offers the potential to detect the genetic alterations that lead to malignant behavior without overt histologic evidence of malignancy. Objective.-To differentiate benign melanocytic nevi from malignant melanoma and to predict the clinical course of melanocytic lesions with ambiguous histology using a novel genomic microarray. Design.-We applied a newly developed single-nucleotide polymorphism genomic microarray to formalin-fixed, paraffin-embedded melanocytic lesions to differentiate benign nevi (n = 23) from malignant melanoma (n = 30) and to predict the clinical course of a set of histologically ambiguous melanocytic lesions (n = 11). Results.-For cases with unambiguous histology, there was excellent sensitivity and specificity for identifying malignant melanoma with this genomic microarray (89% sensitivity, 100% specificity). For cases with ambiguous histology, the performance of this genomic microarray was less impressive. Conclusions.-Without microdissection and with quantities of DNA one-tenth what is required for more commonly used microarrays, this microarray can differentiate between malignant melanoma and benign melanocytic nevi. For histologically ambiguous lesions, longer clinical follow-up is needed to confidently determine the sensitivity and specificity of this microarray. Some of the previous technical hurdles to the clinical application of genomic microarray technology are being overcome, and the advantages over targeted fluorescence in situ hybridization assays currently in clinical use are becoming apparent.     

Melanozytäre Tumoren Klassische und neue diagnostische Möglichkeiten

Melanocytic tumors are one of the major problems in diagnostic dermatopathology as they comprise benign nevi, malignant melanomas and borderline cases. Apart from a proportion of congenital lesions, most benign nevi are acquired tumors that arise during early adulthood and eventually may undergo regressive change. Histologically, they present as so-called common nevi, so-called dysplastic nevi, Spitz's nevi, blue nevi and their variants, and combined nevi. In typical cases, the distinction from a malignant melanoma is not difficult. However, benign simulators of malignancy exist as much as deceptively bland appearing melanomas and in some cases the diagnosis remains dubious despite careful weighting of criteria. Indeed, morphological features may not always suffice to disclose the nature of a melanocytic tumor. Ancillary techniques including immunohistochemistry and measurement of telomerase activity may be of assistance in this respect. One should nevertheless be aware that the biological behavior of certain borderline cases cannot be predicted with certainty.     

On the limited value of fine-needle aspiration for the diagnosis of benign melanocytic proliferations of the skin

This a retrospective study of 39 patients with pigmented cutaneous lesions with a subsequent histologic diagnosis of melanocytic neoplasia. The most important cytologic features seen in the fine-needle aspirates were assessed in the 26 cases deemed satisfactory for evaluation. Though cytology, along with essential clinical data, could enable a general diagnosis of melanocytic lesion and could differentiate it from other nonmelanocytic pigmented lesions, cytology could not provide a precise diagnosis of the different histologic types of benign melanocytic nevi, nor could it enable their differentiation from dysplastic melanocytic nevi or incipient malignant melanoma. Diagn. Cytopathol. 1998;19:441–445. © 1998 Wiley-Liss, Inc.     

Melanocytic nevi in children

Melanocytic nevi in children can be acquired or congenital. The vast majority are benign but diagnosis can be difficult. Nevi in newborns and nodular proliferations in giant congenital nevi can display histological features that lend to their misdiagnosis as melanoma. Acquired melanocytic nevi in children are often Spitz nevi or related nevi. These lesions are too often misdiagnosed as melanoma. On the other hand, clinicians and pathologists must be aware that melanoma does occur in childhood, albeit very rarely. It is exceptionally rare in prepubescent individuals. It can be associated with the following risk factors: giant congenital nevus, family history of melanoma, xeroderma pigmentosum, and immunosuppression. Melanoma in children younger than 10 years of age is extremely rare and has different clinical and histopathological features than those that arise in postpubescents, often being confused with a Spitz's nevus. Consequently, the diagnosis is often made too late. It should be emphasized that thickness is the main prognostic parameter in childhood melanoma and that early diagnosis is also crucial in this age group. The lesion must therefore be examined in a complete excision and not in a partial biopsy, and if atypical features are present it must be reviewed by an expert. On the other hand extreme caution should be exercised when diagnosing melanoma in children, as some benign lesions exhibit similar histologic features.     

Nuclear parameters in the superficial and deep portion of melanocytic lesions--a morphometrical investigation

Morphometric techniques in diagnostic pathology of pigmented skin lesions are not yet well established. The so-called maturation of melanocytes (nevus cells) with progressive descent into the dermis is one important criterion to differentiate benign from malignant melanocytic lesions in conventional histology. We therefore examined morphometrically the nuclear parameters of melanocytes (nevus cells) in the superficial and deep portion of 120 pigmented skin tumors (40 cases each of malignant melanoma. Spitz's nevus and benign dermal nevus) with special emphasis on the "maturation to the depth". A "maturation parameter" (MP) was assessed in each individual case, calculating the ratio of the nuclear area in the deep portion and in the superficial portion. The mean values of MP were 0.720 +/- 0.11 for dermal nevi, 0.725 +/- 0.10 for Spitz's nevi and 1.125 +/- 0.11 for malignant melanoma. The difference between malignant melanoma and both groups of benign nevi was statistically significant (p less than or equal to 0.01). The efficiency of the maturation parameter was 0.95 for the distinction of dermal nevi and malignant melanomas, and 0.97 for the distinction of Spitz's nevi and malignant melanoma. Measurements of the superficial portion only revealed comparatively low efficiency values (0.70 and 0.56, respectively). Our results indicate that morphometric evaluations of the nuclear area of melanocytes (nevus cells) in the deep portion of the infiltrate are more discriminative than those in the superficial portions. The establishment of the MP allows a better differentiation between benign and malignant melanocytic lesions and therefore morphometric methods may obviously be a helpful tool in the diagnosis of melanocytic skin tumors.     

Incidence of melanocytic lesions of the conjunctiva in a review of 10 675 ophthalmic specimens

During the study period, 10,675 human ophthalmic specimens were received at The Henry C. Witelson Ophthalmic Pathology Laboratory and Registry, McGill University, Montreal, Canada. Of those, 271 were conjunctival lesions (2.5%), with 101 being classified as melanocytic: 50 (49.5%) nevi, 36 (35.6%) primary acquired melanoses, and 15 (14.9%) melanomas. After exclusion of referred cases, 85 lesions were included in the study: 44 (51.7%) nevi, 33 (38.8%) primary acquired melanoses, and 8 (9.4%) melanomas. The most prevalent location was the bulbar conjunctiva. Conjunctival melanomas were most commonly found in an older age group than primary acquired melanosis or nevi. Conjunctival nevi were subdivided into compound (32.9%), subepithelial (16.4%), and junctional (2.3%). Primary acquired melanosis were further classified into primary acquired melanosis with atypia (8.2%) and primary acquired melanosis without atypia (30.5%). Primary acquired melanoses was the predisposing lesion in 75% of the cases of melanoma. In our sample, referral bias could alter the distribution of conjunctival pigmented lesions, with a shift toward the malignant end.     

Der dysplastische Nävus Ein Ariadnefaden durch ein konzeptuelles Labyrinth

The concept of the dysplastic nevus is flawed by inconsistencies regarding the clinical presentation, histological features and biological considerations. With regard to basic mechanisms of malignancy, dysplastic nevi exhibit neither increased proliferation indices nor elevated telomerase activity. These traits denote a benign lesion that is highly prevalent among Caucasians. In rare cases, the distinction from malignant melanoma may pose problems.