Langerhans cells, immunomodulation and skin lesions. A quantitative, morphological and clinical study

Abstract

The epidermal Langerhans cell (LC) plays an important role in contact hypersensitivity reactions by presenting the antigens to T lymphocytes. LCs may also play a role in defence mechanisms against neo-antigens in skin tumours. Some studies have indicated that the LC population declines with age. Ultraviolet radiation induces a significant reduction in the number of epidermal LCs and most immunosuppressive drugs decrease the number and function of LCs as well. Such alterations in LCs might predispose to the development of skin tumours. To evaluate the importance of LCs in immunosurveillance of skin tumours, the number and the morphology of LCs was investigated in unaffected skin of patients with cutaneous tumours and in immunosuppressed patients. LCs within basal cell carcinoma (BCC) were examined as well. ATPase and CD1 a staining was used to visualise LCs. The inflammatory response around BCC was estimated by the expression of HLA-DR+, CD3+ and ICAM-1+ cells. The prevalence of skin tumours was studied in renal transient recipients on different immunosuppressive treatments such as azathioprine (Aza) and prednisolone (P), cyclosporin (CyA), azathioprine and prednisolone or cyclosporin and prednisolone. We found no difference in LC populations in patients treated with PUVA (psoralen and UVA-radiation) or in patients with skin tumours as compared with controls and no age-related reduction in LC numbers. However, immunosuppressed patients showed a reduced number of LCs, especially those who had received triple drug therapy (CyA+Aza+P). Patients treated with azathioprine and prednisolone (10-25 years) had a high prevalence of multiple warts (40%) and skin tumours (29%). In contrast, warts and skin tumours were not common in patients treated for 5 years with Cya+Aza+P or with CyA+P. Thus, the duration of immunosuppressive treatment seems crucial for the development of warts and skin tumours. However, the reduction in LC numbers was not more pronounced with time or in patients with skin lesions as compared with those without lesions. In the epidermis overlying basal cell carcinoma (BCC) the number of LCs was decreased and their morphology changed as compared with LCs in perilesional skin. These alterations wee documented in horizontal sheets as well as in vertical sections of the epidermis analysed by light microscopy and with confocal laser scanning microscopy (CLSM). The latter technique permits a quantitative and morphological analysis of LCs in the same tissue volume. In vertical sections, numerous LCs were observed in the dermis surrounding BCC nests. LCs were seldom seen in the dermis of normal skin. No change of the LC populations was found in skin perilesional to BCC. It therefore seems unlikely that the development of skin tumours is due to a primary defect in LCs. Instead, the sparse distribution of LCs overlying BCC might be due to LCs being trapped in the dermis in response to tumour growth. In an attempt to study the regulation of LC migration in tumours, we transplanted UV-induced tumour cell-lines into mice. The alterations found in the epidermal LC populations overlying tumours were not related to tumour anti-genicity or to the immunological status of the host. Cytokines produced by the tumour or by the inflammatory infiltrate observed around BCC nests might instead be responsible for the alterations in LC distribution. In conclusion, long-term sun-exposure of PUVA treatment is not associated with a permanent depletion of LCs. The LC numbers in the skin cannot be used to predict the risk of development of skin tumours in patients on immunosuppressive drugs. The observed changes in LC numbers and morphology do not seem to be primary but are most likely secondary to tumour growth. To clarify the possible role of LCs in tumour immunosurveillance additional functional studies will be required.