A histopathological and immunohistochemical evaluation of scar basal cell carcinomas.

Abstract

Infiltrative morphological mimicry at sites of biopsy-proven nodular basal cell carcinoma has been described. The immunoprofile of scar BCCs (scar BCCs,SBCCs) has not been documented. The aim of this study was to assess the histopathological spectrum, stromal (fibronectin, laminin, actin, desmin and vimentin) response and proliferation (bcl-2, MIB1 and p53) status of SBCCs. Twenty nine BCCs occurring in scars, unrelated to previous malignancy (de novo scar BCCS, DN-SBCCs), 27 BCCs that were incompletely excised and regrew at the same site (regrowth scar BCCs, RG-SBCCs) and 25 BCCs that were completely excised with tumour free margins, but recurred at the same site (recurrent scar BCCs, R-SBCCs) were accessed from the files of the Department of Pathology and Plastic and Reconstructive Surgery of the Faculty of Medicine, University of KwaZulu Natal, and formed the basis of this study. The morphological features of DN-SBCCs was pure (3%), predominantly nodular (79%), micronodular (7%) and infiltrative (11 %). RG-SBCCs were predominantly nodular (82%), micronodular (7%) and infiltrative (11%). RSBCCs were predominantly nodular (80%), micronodular (4%) and infiltrative (16%). The majority of DN-SBCCs, RG-SBCCs and R-SBCCs showed intact basement membrane laminin staining, while two (7%) DN-SBCCs showed 1 + and 2+ loss of basement membrane laminin staining. Three (11 %) and two (8%) RG-SBCCs and R-SBCCs,respectively, showed 2+ or 3+ basement membrane laminin discontinuity. The majority of DN-SBCCs (83%), RGSBCCs (75%) and R-SBCCs (88%) were actin negative. No desmin immunopositivity was demonstrated in the epithelial or stromal components of DN-SBCCs, RG-SBCCs and R-SBCCs. All BCC groups showed high 3+ or 4+ vimentin immunopositivity. The majority (>50%) of the SBCCs showed low (2+) bcl-2 immunopositivity. There was no significant difference in p53 immunopositivity in all SBCCs. SBCCs demonstrate phenotypic and immunophenotypic heterogeneity. That DN-SBCCs with the infiltrative and micronodular patterns have not recurred implies that the histomorphology is a pseudo-aggressive pattern. A similar view could pertain to RG-SBCCs, but because the scar did not cicatrise the incompletely excised BCC implies that the histomorphology of RG-BCC may be a potentially more aggressive phenotype. The recurrence of a completely excised basal cell carcinoma may be viewed as a feature of an aggressive tumour, especially when the recurrent BCC contains micronodular and infiltrative components. However, as most R-SBCCs occurred at head and neck sites that are exposed to ultraviolet light, it is also possible that these are simply new BCCs occurring within scars in head and neck sites prone to BCCs. Furthermore, these R-SBCCs were not destructive tumours. CONCLUSION: None of the infiltrative foci of DN-SBCCs demonstrated laminin loss. Three of 5 with intra-epithelial actin immunopositivity also demonstrated low bcl-2 and high p53 staining, immunoprofiling these with an aggressive infiltrative component. Of 11 RG-SBCCs with high p53 staining, 4 had high p53 staining in the infiltrative component, but only one had a low bcl-2 composite score and low bcl-2 score in the infiltrative focus. In addition, these infiltrative foci demonstrated intraepithelial MSA positivity and a "VA" immunophenotype of the stromal cells, indicating one RG-SBCC with an established, aggressive immunophenotype. Those positive with one or more, but not all, aggressive immunostains, are hypothesised to be RG-SBCCs evolving/developing an aggressive immunophenotype. Only one R-SBCC, with a predominantly infiltrative pattern, had a "full-house" of aggressive immunostaining in the infiltrative foci: low bcl-2, high p53, 2+ laminin discontinuity and intra-epithelial and stromal MSA positivity. Of significance is that 7 with a predominant nodular pattern had a high p53 score. Of these, 5 had high bcl-2 scores. Hence, while high p53 may be a feature of aggressive growth, it is important that this staining be complemented with that of bcl-2, laminin and MSA.