Changes in protein glycosylation of tumour cells, as detected by lectin histochemistry, have been associated with metastasis formation in several human malignancies. This study analysed the association between lectin binding and metastasis in cutaneous malignant melanoma. In a 10-year retrospective study, sections of 100 primary cutaneous malignant melanomas were histochemically stained for the following 5 lectins: HPA, SNA-I, MAA, WGA and PHA-L, differing in their carbohydrate specificity. Since differences in the results of HPA binding depending on methodology have been reported, an indirect and a biotinylated method were employed for HPA. Kaplan-Meier analysis of time to first metastasis revealed a positive correlation between HPA binding and metastasis for both methods, with the biotinylated HPA method (P<0.0001) being superior to the 'indirect' method (P = 0.0006). Cox regression analysis demonstrated that even after adjustment for stage, HPA positivity is an independent predictor for metastasis. The results of the present study indicate that N -acetyl-galactosamine/-glucosamine residues, recognized by HPA, are linked to metastasis in malignant melanoma. In contrast, beta1-6 branched oligosaccharides or sialic acid residues, both of which were correlated with metastasis in other malignancies, are of no functional importance for metastasis formation in malignant melanoma. Thus, HPA proved to be a useful and independent prognostic marker for the metastatic phenotype of melanoma.
Changes in protein glycosylation of tumour cells, as detected by lectin histochemistry, have been associated with metastasis formation in several human malignancies. This study analysed the association between lectin binding and metastasis in cutaneous malignant melanoma. In a 10-year retrospective study, sections of 100 primary cutaneous malignant melanomas were histochemically stained for the following 5 lectins: HPA, SNA-I, MAA, WGA and PHA-L, differing in their carbohydrate specificity. Since differences in the results of HPA binding depending on methodology have been reported, an indirect and a biotinylated method were employed for HPA. Kaplan-Meier analysis of time to first metastasis revealed a positive correlation between HPA binding and metastasis for both methods, with the biotinylated HPA method (P<0.0001) being superior to the 'indirect' method (P = 0.0006). Cox regression analysis demonstrated that even after adjustment for stage, HPA positivity is an independent predictor for metastasis. The results of the present study indicate that N -acetyl-galactosamine/-glucosamine residues, recognized by HPA, are linked to metastasis in malignant melanoma. In contrast, beta1-6 branched oligosaccharides or sialic acid residues, both of which were correlated with metastasis in other malignancies, are of no functional importance for metastasis formation in malignant melanoma. Thus, HPA proved to be a useful and independent prognostic marker for the metastatic phenotype of melanoma.