Search

Anorectal melanomas (AMs) are very rare and highly malignant tumors that are often diagnosed in advanced stages. After the differentiation between cutaneous melanoma (CM) and AM on the molecular level based on the presence of BRAF mutations, further modes of differentiation opened up, such as the recently discovered immunohistologically relevant protein deleted in malignant brain tumors 1 (DMBT1). Over the past several years, increasingly specific therapies have been developed on the basis of new therapy principles. Tyrosin kinase receptors such as Her2 and EGFR have been awarded a large role in this context. The goal of this study was to examine AMs for a possible expression or overexpression of these markers. Expression analyses of Her2 and EGFR were performed immunohistologically on 25 primary AMs. An overexpression of Her2 (score: 3+) was found in one AM from a 68-year-old female patient among these samples. In contrast, EGFR expression was not found in any of the AMs. The results presented here show that isolated cases of AM may benefit from an additive Her2-directed therapy, as the overexpression of Her2 was found in one of our AM patients.

Aims: Anorectal melanoma (AM) forms a rare but highly malignant subset of mucosal melanoma with an extremely poor prognosis. Although AMs display histological and immunohistochemical features very similar to cutaneous melanoma (CM), no association exists either with exposure to ultraviolet light or with melanocytic naevi. While AMs are clearly distinguished from CM by displaying few BRAF mutations, they are commonly indistinguishable from CM at the level of gene expression. The aim was to carry out expression analyses of classical immunohistochemical markers and of the protein deleted in malignant brain tumours 1 (DMBT1) in cases of primary anorectal malignant melanoma and CM., Methods and Results: Expression analyses of classical immunohistochemical markers (S100, HMB45, Melan A and MiTF) and of the protein DMBT1 were carried out in 27 cases of primary anorectal malignant melanoma and 26 cases of CM. All AM cases analysed showed expression of at least three of the classical markers for melanoma. However, immunohistochemistry showed 19 out of 27 AM to be positive for DMBT1, which represented a statistically significant difference (P = 0.0009) compared with CM (six out of 26), which more commonly are negative for DMBT1 expression., Conclusion: These results identify DMBT1 as a molecular feature that may allow distinction between AM and CM and support the notion that AM represents an entity molecularly distinct from CM.

Background & Aims: Anorectal melanoma (AM) is a rare but highly malignant tumor, displaying histologic and immunohistochemical features very similar to cutaneous melanoma (CM). Because BRAF mutations were recently identified in the majority of CM and nevi, we investigated AM for BRAF mutations and mutations of NRAS , an additional component of the MAPK-signalling pathway., Methods: DNA from formalin-fixed and paraffin-embedded AM was PCR amplified and sequenced., Results: We detected BRAF mutations in 2 of 19 cases and NRAS mutations in none of the cases. Mutations in exon 15 of BRAF were present in only 1 tumor (1 of 19 cases). The A1800T base exchange represented a novel mutation and resulted in a K600N transition in an AM from a 96-year-old white man who presented with rectal bleeding and painful sitting of a few weeks' duration. The second positive AM case, a 69-year-old white man who presented with painless rectal bleeding and clinical symptoms of an intestinal constipation showed a novel missense mutation (C1327T leading to R443W conversion) in BRAF exon 11. None of the AM cases displayed the oncogenic V599E mutation preponderating in CM., Conclusions: With regard to the frequency of V599E BRAF mutations, AM significantly differs from CM (P < or = .0001), which suggests that BRAF mutations distinguish anorectal from cutaneous melanoma at the molecular level.

Primary cutaneous melanomas and melanoma metastases were examined for differential gene expression using subtractive suppression hybridization in a search for any genes associated with metastasis. Generating a subtracted library of candidate genes up-regulated in melanoma metastases, this library contained 8 different cDNAs, among them a cDNA fragment of the syntenin gene which was overexpressed in further independent melanoma resection specimens on cDNA Southern blots when compared to acquired melanocytic nevi from which melanomas are known to arise. Upon immunohistochemistry, the syntenin protein expression was detected in the cytoplasm of primary cutaneous melanomas and melanoma metastases. Melanoma metastases exhibited higher proportions of tumour cells positive for syntenin immunostaining in comparison to acquired melanocytic nevi or non-metastasizing primary melanomas which was statistically significant. In addition to melanoma, gastric cancer tissues exhibited a higher syntenin mRNA expression on a matched tumour/normal cDNA array than their normal counterparts which was statistically significant. Altogether, we here first describe the detection of the syntenin protein in resection specimens of melanocytic lesions. We conclude that melanoma metastasis is associated with increased expression of the syntenin gene which may participate in signal transduction and cell adhesion via the multifunctional protein-binding properties of its tandem PDZ domains.

Seeking to identify melanoma-associated genes by comparing gene expression in uncultured primary melanoma specimens with those in nevi, from which melanomas often are known to arise, we applied subtractive suppression hybridization. Generating a subtracted library of candidate genes up-regulated in primary melanomas, this library contained cDNA fragments of the genes encoding heat shock cognate protein (HSP73) and major histocompatibility complex (HLA-DR) which were overexpressed in further 19 independent melanoma resection specimens on cDNA Southern blots when compared to 19 acquired melanocytic nevi. Upon immunohistochemistry, HSP73 protein expression was detected in the cytoplasm of melanoma cells in primary tumours and metastases. In primary melanomas, the proportion of HSP73 protein expressing cells correlated with tumour thickness according to Breslow which was statistically significant. HSP73 immunostaining was stronger in melanoma metastases when compared with acquired melanocytic nevi which was statistically significant. In addition to melanoma, gastric and uterus cancer tissues exhibited higher HSP73 mRNA expression on a matched tumour/normal cDNA array than their normal counterparts which was statistically significant. Participating in the regulation of folding, assembly and degradation of proteins and protecting cellular proteins from the damage caused by cellular stress like hypoxia or changes in cellular pH, elevated HSP73 expression possibly confers proliferative advantage on melanoma cells.

Anorectal melanomas represent a very small group of mucosal melanoma with unknown etiology and poor prognosis. In view of their location, a history of sun exposure is not likely to have had an impact on their development. Recent epidemiologic data indicate a bimodal age distribution. To date there is no information whether an infection with the human papilloma virus plays a role in the tumorigenesis of anorectal melanoma. The lesions can be misdiagnosed as hemorrhoids on clinical examination. On histological examination amelanotic types have been misdiagnosed as lymphoma, sarcoma,and undifferentiated carcinoma. Useful immunohistochemical markers are S 100 protein, HMB-45, Melan A, and MiTF (microphthalmia-transcription-factor). Therapy includes local excision or abdominoperineal resection followed by optional inguinal and parailiac lymph node dissection, and consecutive chemo- and immunotherapy. The poor long-term prognosis of anorectal melanomas correlates with their advanced tumor size and depth of infiltration at diagnosis. The overall 5-year survival rates range between <5 and 22% in different series.

The history of gallbladder involvement by a malignant melanoma in a 65-year-old woman is reported. The gallbladder, clinically resected for cholecystitis, harboured a polypoid dark pigmented tumour. The tumour was identified as a malignant melanoma immunohistochemically by positive reactions for gp100 (HMB45), melan A, and MiTF. Clinically, the patient was treated for cutaneous malignant melanoma by local excision 10 years earlier. The literature of pigmented lesions of the gallbladder is reviewed. In conclusion, the most important differential diagnosis of pigmented lesions of the gallbladder is the secondary gallbladder melanoma.

Anorectal melanomas are similar to cutaneous melanomas with regard to the mode of spread and to the immunophenotype. When compared with patients with cutaneous melanoma, those suffering from anorectal melanoma have a much worse outcome. The etiology of anorectal melanomas is as yet completely unknown. For anatomical reasons, ultra-violet (UV-B) radiation can not cause anorectal melanomas as in cutaneous tumours, that are associated with exposure of the skin to UV-B radiation. As the cytokine interleukin-6 (IL-6) is known to stimulate melanoma tumour cell proliferation and a functional homologue of human IL-6 has been identified recently in the HHV-8 genome, this tumorigenic virus might be involved in the pathogenesis of anorectal melanomas. Twelve formalin fixed and paraffin embedded primary anorectal melanomas from seven female and five male patients with a mean age at diagnosis of 71 years (range 38-88 years) were investigated for the presence of HHV-8 DNA. Using a specific and highly sensitive polymerase chain reaction protocol, this tumorigenic gamma-herpesvirus was not detectable in any tumour. This data indicates that HHV-8 is not involved in the development of anorectal melanomas., (Copyright 2001 Wiley-Liss, Inc.)