Quantitative immunohistochemical differences in Langerhans cells in dermatitis due to internal versus external antigen sources

Atypical cells in human cutaneous re-excision scars for melanoma express p75NGFR, C56/N-CAM and GAP-43: evidence of early Schwann cell differentiation

BACKGROUND

A common problem in the routine examination of melanoma re-excision scars occurs when a few or rare mildly atypical cells are present within the scar, raising the question of residual disease. Little is known about the derivation of these cells. Because the normal cutaneous wound-healing process is reparative, we hypothesized that these atypical cells may be reactive proliferating Schwann cell precursors.

METHODS

The expression of the Schwann cell differentiation markers p75NGFR, CD56/N-CAM and GAP-43 was examined by immunohistochemistry in scars of wide local re-excisions for melanoma and non-melanoma tumors. Expression of S100, gp100 (with HMB45) and MART1 was also analyzed by immunohistochemistry.

RESULTS

All melanoma and non-melanoma re-excision specimens contained mildly atypical, spindled or epithelioid cells within the scar. They varied in number from case to case and expressed S100, p75NGFR, CD56/N-CAM or GAP-43 but not gp100 (with HMB45) or MART1. Rare epithelioid non-melanoma cells within the superficial dermis expressed MART-1.

CONCLUSIONS

Atypical cells are present in re-excision scars from melanoma and non-melanoma cases. They demonstrate early Schwann cell differentiation and appear to proliferate during the scarring process. The use of anti-MART-1 alone in the examination of melanoma re-excisions specimens may be inadequate as it may label rare, superficially located, non-melanoma cells within the scar.

Quantitation of intraepidermal T-cell subsets in formalin-fixed, paraffin-embedded tissue helps in the diagnosis of mycosis fungoides

Differentiation between mycosis fungoides (MF) and cutaneous inflammatory processes can usually be made on clinical and histologic grounds. In difficult cases, immunohistochemical studies can be helpful since MF infiltrates usually contain a predominance of CD4+ lymphocytes, while most inflammatory lesions usually have a mixture of CD4+ and CD8+ lymphocytes. However, this determination has traditionally required the use of frozen tissue, thus severely limiting its usefulness. Recently, antibodies that differentially label CD4+ and CD8+ lymphocytes in formalin-fixed, paraffin-embedded tissue have become available (OPD4 and C8/144B respectively, DAKO (Carpinteria, CA, USA). This study tests the utility of these antibodies in the pathologic diagnosis of MF and inflammatory lesions with significant exocytosis. In 9 cases of MF for which both frozen and fixed tissues were available for comparison, the OPD4+ cell count in fixed tissue was significantly lower than the Leu-3a+ cell count in frozen tissue. Also, the C8/144B+ cell count in fixed tissue was higher than the Leu-2a+ cell count in frozen tissue, although this difference was not significant statistically. In a larger series for which only fixed tissue was available, epidermal CD4:CD8 ratios were significantly greater in 23 MF cases (mean 4.0+/-4.76) than in 35 inflammatory cases (mean 0.6+/-0.42; p = 0.001). Thus, although the studied antibodies appear to detect different epitopes in frozen versus paraffin-embedded tissue, demonstration of an elevated CD4:CD8 ratio in fixed tissue supports the diagnosis of MF, and is a helpful adjunct to routine histopathology.