Case reportA 10-year-old boy presented with an ulcerated lesion on his lip in 2002. Examination revealed an ulcerated red papule on the lower lip. This was conservatively completely excised and sent for histolopathologic examination.The lesion was diagnosed as a melanoma, Breslow thickness at least 2.7mm, Clarks level V, by three local pathologists as well as by reviewing pathologists at an international melanoma referral service.Histopathologically, the lesion is a compound melanocytic tumour with many features pathologists associate with melanoma, namely: surface ulceration, mitoses, peripheral nerve infiltration, lack of maturation (Figure 1). Figure 1. Histopathologic examination reveals a compound melanocytic tumour with many worrisome features including: surface ulceration, neurotropism, lack of maturation, increased mitoses (H&E stain \u00d710) A wider excision was performed without evidence of residual disease. No clinical or radiologic evidence of metastatic disease was noted at that time. Over the subsequent 10 years he was closely followed, had numerous other benign melanocytic lesions removed, and to date has developed no evidence of metastatic disease. He presented recently for an unrelated naevus. Melanoma is rare but well described in children. Due to the unusual clinical presentation and course the dermatologic surgeon asked for the initial tumour to be reviewed. The archival block was recut and immunohistochemical studies were performed: S100A6 and p16 were diffusely positive; Mib-1 revealed a proliferative rate of 1%; HMB-45 revealed zonal staining with minimal deep staining. None of these features provide absolute diagnostic conclusions but have all been associated with benign melanocytic lesions. In addition, we applied Fluorescence in situ hybridisation (FISH) at IGENZ laboratory in Auckland using Vysis four probe Melanoma FISH probe kit (Abbott) and an additional CDKN2A (9p21) probe. These studies failed to reveal the usual molecular aberrations seen in melanoma. Figure 2a. FISH images with preservation (diploid copies) of the four probes, indicating a negative result Figure 2b. This represents an example of a positive result (not the current case). There are three copies of the red, blue and yellow probes indicating a positive result if the probe set protocol is applied Given the findings of these recently developed diagnostic tools, the lesion was reclassified as an atypical Spitz tumour rather than malignant melanoma. It should be noted that the predicted 10yr survival rate if this was a melanoma (2.7mm thick, ulcerated, axial) is approximately 60% (melanomaprognosis.org), so the patients outcome should not exclude the possibility of a malignant diagnosis. Discussion: Investigation of the molecular basis of disease has provided profound insight into tumour pathogenesis, enhanced our understanding of disease aetiology, provided diagnostic tools, and provided targets for specific therapy. The pathologic evaluation of melanocytic lesions remains one of the most controversial areas of all diagnostic pathology. Though histopathology remains the gold standard in diagnosing melanoma there is a convincing body of literature to suggest that experts may disagree on even the most basic aspects of disease classification.1,2 An impressive increase in incidence of melanoma has been well documented around the world including New Zealand. Rigorous statistical analysis has shown that sun exposure, population trends and other epidemiologic considerations fail to completely explain this phenomenon. At least partially the \"epidemic\" may be attributed to a changing histopathologic diagnostic practice and a tendency to \"over-call\" atypical melanocytic lesions as melanoma due to medico legal, insurability, and even clinical considerations.3 The \"LPLK phenomenon\" illustrates the changing clinical practice of biopsying early suspicious lesions. Lichen planus-like keratoses are common lesions (one of the authors (PE) estimates he diagnoses 5 such lesions per day) were apparently never diagnosed decades ago, as these typically small lesions were seldom biopsied. A similar argument may be made for the steadily rising incidence of melanoma in situ. While diagnostically convincing cases show excellent concordance between reviewers, distinction between unusual forms of melanoma and naevi can be exceedingly difficult. Various immunohistochemical studies, namely, HMB-45, Mib-1, S100A6, p16 have been studied and shown to modestly increased diagnostic concordance and accuracy, particularly with regards to Spitzoid tumours.4,5 Chromosomal instability has long been recognised as a hallmark of cancer. Interest in applying molecular diagnostics techniques to melanoma can be largely attributed to the work of Bastian and colleagues who applied comparative genomic hybridisation (CGH) to melanocytic lesions. This technique essentially compares a tumour karyotype with that of benign tissue to assess gross gains or losses in chromosomal regions. By evaluated 132 melanomas and 54 benign nevi they showed that 96% of the melanomas had some type of chromosomal copy number aberration, while chromosomal copy number aberrations were only rarely seen in nevi. 6,7 FISH analysis has been applied to melanoma and purported to be able to detect genetic abnormalities discriminating melanoma from nevi. Importantly, the technique allows in-situ assessment of chromosome abnormalities by directly applying probes to tissue sections and thus allowing the pathologist to be certain the correct cells are being assessed. This is particularly significant for thin lesions with a low tumour volume. It also offers an advantage over other techniques which extract cells from tumour sections in a manner which loses the context of the tissue section and may allow contamination with other cell types such as inflammatory cells or cells from an antecedent naevus. In addition, FISH can be performing on archival material, is rapid (can be completed in 2 days), fits into existing pathology laboratory workflows, and is relatively inexpensive. It has already gained some acceptance as a technique in prognosticating uveal melanoma.8 From the CGH studies, 4 chromosome regions (the most frequently aberrant in melanoma) were chosen and appropriate FISH probes were produced as a multicolour 4 probe set by Vysis Melanoma Probe kit (Abbott Molecular). The 4-probe multicolour FISH probe panel targeting chromosomes 6 and 11. Specifically, the panel consisted of ras responsive element binding protein 1 (RREB1, 6p25), v-myb myeloblastosis viral oncogene homologue (MYB, 6q23), CEN6 (centromere 6), and cyclin D1 (CCND1, 11q13). Utilising the Abbott Molecular probe set, Gerami et al,9 studied a series of 169 melanocytic lesions including 86 unequivocal benign nevi and 83 malignant melanomas, showed a sensitivity of 86.7% and a specificity of 95.4%. The four cases of benign nevi showing chromosomal alterations on FISH were reviewed and reclassified as Spitz nevi or dysplastic nevi. The technique has been refined and a variety of institutions have shown results similar to this. In a large study of diagnostically ambiguous tumours, FISH was compared with expert histopathologic review. The sensitivity and specificity of histopathological review were 95 and 52%, and the sensitivity and specificity of FISH were 43 and 80% respectively. Interestingly, by combining the histopathological diagnosis with FISH results, the diagnosis was optimized, especially by increasing specificity (76% instead of 52% for expert diagnosis alone) and by improving sensitivity compared with FISH alone (90 vs 43% for FISH result alone).10 Practically speaking, the relatively high specificity (i.e. low false positive result) of FISH when compared with expert review itself may help justify applying this test to ambiguous tumours. As illustrated in the case described herein, Spitzoid tumours remain diagnostically difficult. A recent study of 43 unequivocal Spitzoid melanomas, standard FISH studies were only 70% sensitive. The Abbott four probe methodology has been further refined to include a 9p21 FISH assay, which revealed a combined sensitivity of 85% and specificity of 100%. The results suggested that the 9p21 assay may be highly complementary to the standard melanoma FISH assay.11 In a recent study, 64 atypical Spitz tumours with 5 years of uneventful follow-up and 11 atypical Spitz tumours resulting in advanced disease or death were evaluated by FISH. Cases with homozygous 9p21 deletions had the greatest risk. Cases with 6p25 or 11q13 gains also had higher risk for aggressive clinical behaviour than FISH-negative atypical Spitz tumours or cases with 6q23 deletions.12 In summary, we present a case of a lesion diagnosed as a thick malignant melanoma in a child which was reclassified as an atypical Spitz tumour on the basis of recently developed diagnostic tools. FISH does not always low precise classification of melanocytic lesions. However, its relatively high specificity (when compared with expert histology review) and utility in stratifying risk of metastasis suggest it may become routine practice in approaching difficult melanocytic lesions.

Case reportA 10-year-old boy presented with an ulcerated lesion on his lip in 2002. Examination revealed an ulcerated red papule on the lower lip. This was conservatively completely excised and sent for histolopathologic examination.The lesion was diagnosed as a melanoma, Breslow thickness at least 2.7mm, Clarks level V, by three local pathologists as well as by reviewing pathologists at an international melanoma referral service.Histopathologically, the lesion is a compound melanocytic tumour with many features pathologists associate with melanoma, namely: surface ulceration, mitoses, peripheral nerve infiltration, lack of maturation (Figure 1). Figure 1. Histopathologic examination reveals a compound melanocytic tumour with many worrisome features including: surface ulceration, neurotropism, lack of maturation, increased mitoses (H&E stain \u00d710) A wider excision was performed without evidence of residual disease. No clinical or radiologic evidence of metastatic disease was noted at that time. Over the subsequent 10 years he was closely followed, had numerous other benign melanocytic lesions removed, and to date has developed no evidence of metastatic disease. He presented recently for an unrelated naevus. Melanoma is rare but well described in children. Due to the unusual clinical presentation and course the dermatologic surgeon asked for the initial tumour to be reviewed. The archival block was recut and immunohistochemical studies were performed: S100A6 and p16 were diffusely positive; Mib-1 revealed a proliferative rate of 1%; HMB-45 revealed zonal staining with minimal deep staining. None of these features provide absolute diagnostic conclusions but have all been associated with benign melanocytic lesions. In addition, we applied Fluorescence in situ hybridisation (FISH) at IGENZ laboratory in Auckland using Vysis four probe Melanoma FISH probe kit (Abbott) and an additional CDKN2A (9p21) probe. These studies failed to reveal the usual molecular aberrations seen in melanoma. Figure 2a. FISH images with preservation (diploid copies) of the four probes, indicating a negative result Figure 2b. This represents an example of a positive result (not the current case). There are three copies of the red, blue and yellow probes indicating a positive result if the probe set protocol is applied Given the findings of these recently developed diagnostic tools, the lesion was reclassified as an atypical Spitz tumour rather than malignant melanoma. It should be noted that the predicted 10yr survival rate if this was a melanoma (2.7mm thick, ulcerated, axial) is approximately 60% (melanomaprognosis.org), so the patients outcome should not exclude the possibility of a malignant diagnosis. Discussion: Investigation of the molecular basis of disease has provided profound insight into tumour pathogenesis, enhanced our understanding of disease aetiology, provided diagnostic tools, and provided targets for specific therapy. The pathologic evaluation of melanocytic lesions remains one of the most controversial areas of all diagnostic pathology. Though histopathology remains the gold standard in diagnosing melanoma there is a convincing body of literature to suggest that experts may disagree on even the most basic aspects of disease classification.1,2 An impressive increase in incidence of melanoma has been well documented around the world including New Zealand. Rigorous statistical analysis has shown that sun exposure, population trends and other epidemiologic considerations fail to completely explain this phenomenon. At least partially the \"epidemic\" may be attributed to a changing histopathologic diagnostic practice and a tendency to \"over-call\" atypical melanocytic lesions as melanoma due to medico legal, insurability, and even clinical considerations.3 The \"LPLK phenomenon\" illustrates the changing clinical practice of biopsying early suspicious lesions. Lichen planus-like keratoses are common lesions (one of the authors (PE) estimates he diagnoses 5 such lesions per day) were apparently never diagnosed decades ago, as these typically small lesions were seldom biopsied. A similar argument may be made for the steadily rising incidence of melanoma in situ. While diagnostically convincing cases show excellent concordance between reviewers, distinction between unusual forms of melanoma and naevi can be exceedingly difficult. Various immunohistochemical studies, namely, HMB-45, Mib-1, S100A6, p16 have been studied and shown to modestly increased diagnostic concordance and accuracy, particularly with regards to Spitzoid tumours.4,5 Chromosomal instability has long been recognised as a hallmark of cancer. Interest in applying molecular diagnostics techniques to melanoma can be largely attributed to the work of Bastian and colleagues who applied comparative genomic hybridisation (CGH) to melanocytic lesions. This technique essentially compares a tumour karyotype with that of benign tissue to assess gross gains or losses in chromosomal regions. By evaluated 132 melanomas and 54 benign nevi they showed that 96% of the melanomas had some type of chromosomal copy number aberration, while chromosomal copy number aberrations were only rarely seen in nevi. 6,7 FISH analysis has been applied to melanoma and purported to be able to detect genetic abnormalities discriminating melanoma from nevi. Importantly, the technique allows in-situ assessment of chromosome abnormalities by directly applying probes to tissue sections and thus allowing the pathologist to be certain the correct cells are being assessed. This is particularly significant for thin lesions with a low tumour volume. It also offers an advantage over other techniques which extract cells from tumour sections in a manner which loses the context of the tissue section and may allow contamination with other cell types such as inflammatory cells or cells from an antecedent naevus. In addition, FISH can be performing on archival material, is rapid (can be completed in 2 days), fits into existing pathology laboratory workflows, and is relatively inexpensive. It has already gained some acceptance as a technique in prognosticating uveal melanoma.8 From the CGH studies, 4 chromosome regions (the most frequently aberrant in melanoma) were chosen and appropriate FISH probes were produced as a multicolour 4 probe set by Vysis Melanoma Probe kit (Abbott Molecular). The 4-probe multicolour FISH probe panel targeting chromosomes 6 and 11. Specifically, the panel consisted of ras responsive element binding protein 1 (RREB1, 6p25), v-myb myeloblastosis viral oncogene homologue (MYB, 6q23), CEN6 (centromere 6), and cyclin D1 (CCND1, 11q13). Utilising the Abbott Molecular probe set, Gerami et al,9 studied a series of 169 melanocytic lesions including 86 unequivocal benign nevi and 83 malignant melanomas, showed a sensitivity of 86.7% and a specificity of 95.4%. The four cases of benign nevi showing chromosomal alterations on FISH were reviewed and reclassified as Spitz nevi or dysplastic nevi. The technique has been refined and a variety of institutions have shown results similar to this. In a large study of diagnostically ambiguous tumours, FISH was compared with expert histopathologic review. The sensitivity and specificity of histopathological review were 95 and 52%, and the sensitivity and specificity of FISH were 43 and 80% respectively. Interestingly, by combining the histopathological diagnosis with FISH results, the diagnosis was optimized, especially by increasing specificity (76% instead of 52% for expert diagnosis alone) and by improving sensitivity compared with FISH alone (90 vs 43% for FISH result alone).10 Practically speaking, the relatively high specificity (i.e. low false positive result) of FISH when compared with expert review itself may help justify applying this test to ambiguous tumours. As illustrated in the case described herein, Spitzoid tumours remain diagnostically difficult. A recent study of 43 unequivocal Spitzoid melanomas, standard FISH studies were only 70% sensitive. The Abbott four probe methodology has been further refined to include a 9p21 FISH assay, which revealed a combined sensitivity of 85% and specificity of 100%. The results suggested that the 9p21 assay may be highly complementary to the standard melanoma FISH assay.11 In a recent study, 64 atypical Spitz tumours with 5 years of uneventful follow-up and 11 atypical Spitz tumours resulting in advanced disease or death were evaluated by FISH. Cases with homozygous 9p21 deletions had the greatest risk. Cases with 6p25 or 11q13 gains also had higher risk for aggressive clinical behaviour than FISH-negative atypical Spitz tumours or cases with 6q23 deletions.12 In summary, we present a case of a lesion diagnosed as a thick malignant melanoma in a child which was reclassified as an atypical Spitz tumour on the basis of recently developed diagnostic tools. FISH does not always low precise classification of melanocytic lesions. However, its relatively high specificity (when compared with expert histology review) and utility in stratifying risk of metastasis suggest it may become routine practice in approaching difficult melanocytic lesions.