3). On the

other hand, cyclin D1 expression was <25% in Groups 1, 2, and 3, but >50% in Group 4 (70.6% of the samples). Group 2 showed no cases with >75% of the cells expressing cyclin D1. A significant negative correlation was observed between ROC1 and cyclin D1 expression levels regardless of neoplasia type (benign or malignant) (p = 0.0008985). Comparisons between ROC1 and cyclin D1 expression in melanomas and melanocytic nevi are shown in Table 1 and Table 2, respectively. In some cases of melanoma, areas with >75% of the cells expressing ROC1 and <25% of cells expressing cyclin D1 were observed adjacent to areas wherein ROC1 was positive in <25% of the cells, and cyclin D1 was expressed in >75% of the cells. This was found to be independent of increased gene expression (Fig. 4). The ROC1/cyclin D1 relationship did not vary with age, gender, or lesion site in either melanomas or melanocytic nevi (p > 0.05).

Increased C646 cost ROC1 protein expression, as compared with cyclin D1 expression, RGFP966 ic50 predominated in all samples (65% of cases; n = 78). In the melanocytic nevus group, the ROC1 expression increase was remarkably predominant in relation to cyclin D1 expression (86.2% of the cases). In melanomas, this ROC1 expression predominance was also observed, but in only 45.2% of the cases (p < 0.001) ( Table 3). Although ROC1 and cyclin D1 expression levels were predominantly proportional in melanomas with thickness >2 mm, and although a great number of cases with melanomas >4 mm (35.3%) showed increased cyclin D1 expression in comparison with ROC1 levels, no statistically significant difference was seen among the groups (p = 0.166). Only in the acral lentiginous melanoma group was cyclin D1 expression greater than that of ROC1 in a large number of cases (40%). On the other hand, this group also showed the largest number of cases with increased ROC1 expression as compared TCL to cyclin

D1 expression (50%). No statistically significant difference in the ROC1/cyclin D1 relationship was observed in relation to melanoma histological type (p = 0.605). Six cases (five melanomas and one melanocytic nevus) exhibited CCND1 gene amplification. In two amplified cases, one was acral lentiginous melanoma and the other was nodular melanoma with Breslow thickness of >4 mm. Cyclin D1 was expressed in 51–75% of the acral lentiginous melanoma cells and in >75% of the nodular melanoma cells. In both the acral lentiginous and nodular melanomas, ROC1 expression was present in <25% of the cells. In the other amplified melanomas (2 SSM and 1 LMM), in one case, the Breslow's thickness was <1 mm, in another it was 1.01–2 mm, and in the other it was 2.01–4 mm. Of these three amplified melanomas, two showed cyclin D1 and ROC1 expression in 51–75% of the cells, while in the other case, cyclin D1 positivity was <25%, and ROC1 was expressed in >75% of the cells.