Differences in tumor thickness between hematoxylin and eosin and Melan-A immunohistochemically stained primary cutaneous melanomas.

Tumor thickness (Breslow thickness) represents the main prognostic factor in primary melanoma. Potential differences in melanoma tumor thickness measurements between conventional hematoxylin and eosin (H&E) and Melan-A immunohistochemical staining were evaluated. Ninety-nine excisional biopsies were included in the study. From each sample, 2 consecutive histological sections were stained with H&E and Melan-A, respectively. Tumor thickness was measured from both sections by 2 independent observers. In 59 biopsy specimens (59.6%), higher tumor thickness measurements were recorded in Melan-A-stained than in H&E-stained sections. In 42.4% of such cases (25 biopsies), the observed differences were ≥0.2 mm. After Melan-A evaluation, 33% of in situ melanoma cases were reclassified as invasive melanoma, with thickness measurements ranging from 0.15 to 0.35 mm. In 23 biopsies, identical values were recorded with both techniques, whereas in 17 cases, measurements obtained with H&E staining were slightly higher (from 0.01 to 0.18 mm) than those obtained with Melan-A staining. A high rate of interobserver agreement was noted, and significant intertechnique measurement differences were detected. Significant discrepancies (≥0.2 mm) in thickness measurements between the 2 techniques were mainly attributed to the presence of individual or small clusters of melanocytic cells in the papillary dermis. These melanocytic cells could be easily overlooked in H&E-stained sections, especially in sections showing dense lymphohistiocytic inflammatory infiltrates, numerous melanin-containing histiocytic cells in the upper dermis, or extensive fibrotic changes or regression phenomena. This study confirms the practical interest of immunohistochemical staining with Melan-A in evaluating primary melanoma and, specifically, in situ melanoma cases.