Complete regression of primary cutaneous malignant melanoma associated with distant lymph node metastasis: a teaching case mimicking blue nevus.

BACKGROUND: Malignant melanoma (MM) tends to be spontaneously regressed, however, complete regression of primary cutaneous MM is an extremely rare phenomenon. Our aim is to be aware that pathologists and/or dermatologists can readily misinterpret it as the other benign or malignant lesions. CASE PRESENTATION: A gradually growing and verrucous hypopigmented macule had been noticed in the right sole of a 65-year-old Japanese male since 2 years before, and it turned to be a solitary bluish to black patch with surrounding depigmentation and was recently decreased in size. In parallel, the patient had a rapidly growing black-pigmented mass lesion at the right inguen. The cutaneous specimen from the sole showed an aggregation of many melanophages predominantly in the middle to deep layer of dermis, associated with surrounding fibrosis, reactive vascular proliferation and CD8-positive T-lymphocytic infiltrate, covered by attenuated epidermis with absence of rete ridge. However, no remnant MM cells were completely seen in the step-serial sections. We first interpreted it as blue nevus. By contrast, the inguinal mass revealed a diffuse proliferation of highly atypical mono- to multi-nucleated large cells having abundant eosinophilic cytoplasm in the enlarged lymph node tissue. Immunohistochemical findings demonstrated that these atypical cells were specifically positive for HMB45 and Melan A. Therefore, we finally made a diagnosis of complete regression of primary cutaneous MM associated with distant lymph node metastasis of MM. CONCLUSION: Careful, not only general/cutaneous but histopathological, examinations should be necessary and adjunctive aids for reaching the correct diagnosis of complete regression of cutaneous MM.

Oral melanocytic lesions: differences in expression of HMB-45 and S-100 antigens in round and spindle cells of malignant and benign lesions.

Immunohistochemistry and melanin bleaching were used to assess the expression of antigens identified by anti-S-100 and anti-HMB-45 antibodies on melanomas and intramucosal and blue nevi from the oral mucosa of 18 patients. Both antibodies reacted with cells in all three types of lesions, but there were differences in the expression of these antigens between the round and spindle cells within the lesions. In melanomas composed of round cells, the intensity and distribution of staining with HMB-45 was greater than with S-100. The opposite was true in melanomas composed of spindle-shaped cells, and one spindle-cell melanoma was HMB-45-negative. The round cells of intramucosal nevi expressed S-100 more intensely and more frequently than HMB-45. The spindle-shaped cells of blue nevi strongly expressed both S-100 and HMB-45. Whereas intradermal nevi from the skin do not express HMB-45, intramucosal nevi consistently express this antigen in the lesion and overlying mucosa. Oral melanomas composed of round and spindle-shaped cells show differences in their expression of S-100 and HMB-45 antigens, making the use of both antibodies complementary in the diagnosis of undifferentiated tumors.

Proliferative activity in the malignant cellular blue nevus.

The proliferative activity of four malignant cellular blue nevi (MCBN) was assessed in routinely fixed, paraffin-embedded material using staining for the argyrophilic nucleolar organizer regions (AgNORs), immunohistochemical staining for proliferating cell nuclear antigen (PCNA [PC10]), and DNA flow cytometry. The objective was to determine whether the evaluation of proliferative activity could represent a useful diagnostic parameter. Four cellular blue nevi (CBN), 10 melanocytic nevi (MN), four common blue nevi (BN), and 10 conventional malignant melanomas (MMs) were selected as controls. In the MCBN the mean AgNOR number, evaluated on the basis of 100 tumor cells, was 8.33 +/- 0.83; NORs were small and dispersed throughout the nucleus; the mean PCNA score was 31.93% +/- 4.4; and two of the cases were aneuploid and two diploid. In the CBN the AgNOR count was 3.69 +/- 0.56; NORs were large and mainly grouped in a central cluster; the mean PCNA score was 3.53% +/- 1.28; and three of the cases were diploid and one aneuploid. The AgNOR counts in the MCBN were significantly different from those in the CBN (P = .0002), MN (3.04; P = .00001), and BN (2.93; P = .00006), whereas they were not significantly different from those in the conventional MMs (7.64; P = .58). The PCNA (PC10) scores in the MCBN were significantly different from those in the CBN (P = .00003), MN (2.05%; P = .00001), and BN (5.06%; P = .00002), whereas they were not significantly different from those in the conventional MMs (28.9%; P = .49). In all the cases a linear relationship between AgNOR counts and PCNA scores was observed (r = .94, P = .00001). Our results indicate that AgNOR analysis and PCNA immunostaining can be regarded as useful additional parameters for the diagnosis of MCBN.