Microsoft Word - DRM166BF

Dr. Rino Cerio, Prof. E. Wilson Jones, Institute of Dermatology, St. John’s Hospital for Diseases of the Skin, London (UK) Dear Sir, We read with interest the rapid communication by Arrese Estrada and Piérard entitled ‘factorXIIIa-positive dendrocytes and the dermal microvascular unit’ [1]. The authors labelled a variety of benign and malignant skin conditions with factor XΠIa (F-XIIIa) and proposed that positive cells may be resident perivascular phagocytic cells. This observation confirms our original suggestion that these cells were indeed phagocytic [1 ]. Moreover, our work on dermatofibroma [2] and fibrous papules of the nose [3] demonstrates that the histogenesis of these skin lesions is probably different from most other fibrocollagen-ous conditions such as scar and keloid. Furthermore, the intense posi-tivity observed in dermatofibroma can be useful in distinguishing this lesion from negatively labelled dermatofibrosarcoma [2]. Since then, one of us [R.C.] in collaboration with Headington and colleagues has characterized F-XIIIa-positive dermal dendritic cells in normal neonatal and adult skin [4] by employing double immunofluorescence techniques together with a large panel of mononuclear cell markers. Our findings indicate that these cells may represent a specific population of bone-marrow-derived dermal dendritic cells, distinct from LangerLetters to the Editor 167 hans cells, that indeed share some epitopes with mononuclear phagocytes (monocytes/macrophages). In addition, the detection of HLA-DQ on 48% of F-XIIIa-positive cells and the lack of OKM1 in combination with high OKM5 expression suggests an antigenpresenting cell phenotype. Parallel studies by Sontheimer et al. [5, 6] have independently identified this cell population using class II antigens introducing the term perivascular dendritic macrophage. A recent communication has confirmed that most of these cells are F-XIIIa-positive [Sontheimer, pers. commun.]. We agree ‚ therefore ‚ with the proposals of Arrese Estrada and Pié-rard that in normal skin these cells are resident phagocytic cells and, furthermore, it is not surprising that there should be a modest increase in such cells in many inflammatory, non-inflammatory, fibrohistiocytic and granulomatous dermatoses [7]. We cannot agree with Arrese Estrada and Piérard that the hyper-plasia of dendrocytes is not a specific feature of histiocytoma cutis. Unlike inflammatory and non-inflammatory skin conditions, F-XIIIa labelling is observed in the majority of cells that comprise dermatofi-broma, especially at the periphery of early lesions with maximum immunoreactivity adjacent to the dermal papillae. This consistent labelling pattern differs from that found in most other