• cellular angiofibroma;
  • differential diagnosis;
  • immunohistochemistry;
  • morphology;
  • vulva

Aims : Cellular angiofibroma (CA) is a rare benign mesenchymal lesion with a predilection for the vulval region. In this report we aim to describe the clinical, pathological and immunohistochemical features of a series of vulval mesenchymal lesions, some of which have the classically described histological appearance of CA while others exhibit atypical features. We believe these lesions fall within the broad spectrum of fibromatous lesions of the vulva.

Methods and results : Seven cases were included. Histological sections were examined and immunohistochemical staining with vimentin, desmin, α smooth muscle actin, h-caldesmon, S100, EMA, AE1/3, CD34, CD10, ER, PR and MIB1 was performed. The patients' ages ranged from 20 to 65 years and the lesions ranged in size from 10 to 50 mm. All lesions were well circumscribed, moderately cellular lesions and were composed of bland spindle-shaped cells set in a fibrous stroma. Many blood vessels with thick hyalinized walls were present in four cases, in one case occasional such blood vessels were present and in two cases vessels with thick hyalinized walls were not present. In five cases the vessels were at least focally dilated resulting in a haemangiopericytomatous pattern. Histological features identified in a variable numbers of cases included peripheral adipose tissue (four cases), adipose tissue within the centre of the lesion (one case), stromal mast cells (six cases), stromal lymphoid aggregates (five cases), scattered multinucleate cells (five cases), hypocellular hyalinized areas (two cases), myxoid areas (four cases) and focal areas of marked cellular atypia reminiscent of symplastic change within a uterine leiomyoma (one case). Mitotic figures were identified in four cases, all with a mitotic count of <1 per 10 high-power fields. Immunohistochemically all neoplasms were positive with vimentin and all but one with ER and PR (PR staining was not performed in one tumour). In all cases desmin, α smooth muscle actin, h-caldesmon, S100 and AE1/3 were negative (h-caldesmon and AE1/3 staining were not performed in one case). Three cases were positive with CD34, one with EMA and two with CD10. All exhibited a low MIB1 proliferation index of approximately 1%. One lesion recurred locally 6 months following initial removal.

Conclusions : CA is a rare benign vulval mesenchymal lesion with limited potential for local recurrence. We describe several hitherto unreported histological features which add to the morphological spectrum. Although not all lesions exhibit the classically described histological features of CA, we believe all fall within the broad spectrum of benign vulval fibromatous lesions. These cases are characterized by vimentin positivity but negative staining with smooth muscle markers which assists in excluding many of the other vulvovaginal mesenchymal lesions which enter into the differential diagnosis. The immunophenotype indicates that CA probably exhibits fibroblastic rather than myofibroblastic differentiation. These lesions are almost always positive with ER and PR, suggesting that they probably arise from the hormone receptor-positive subepithelial mesenchymal layer within the lower female genital tract.