The practical value of the detection of clonality within the T-cell receptor gamma locus by polymerase chain reaction for the diagnosis of cutaneous T-cell lymphomas is well known. However, studies dealing with this subject so far, with special emphasis on the sensitivity of the technique in comparison to, for example, Southern blotting have used mixtures of DNA in various concentrations instead of using mixtures of the cells involved, which would reflect the in vivo situation in a more realistic scope. The purpose of this study was therefore to determine the sensitivity and the limitations of the PCR assay by dilution experiments, using mixtures of cells. Furthermore we studied its applicability to cutaneous T-cell proliferative disorders. Two clonal T-cell lines (MyLa and Jurkat) served as positive control. Dilutions of MyLa cells, cultured normal human keratinocytes and peripheral blood mononuclear cells from lymphoma negative volunteers were used to assess the sensitivity of the PCR-DGGE assay. Skin samples from 4 patients with cutaneous T-cell lymphoma, 1 lesional lymph node, 2 blood samples from a patient with Sézary syndrome and 4 lymphoma-negative tissue samples were analysed. Two samples were uncertain for diagnosis of lymphoma. The PCR-DGGE assay consisted of a 2-round nested PCR with consensus primers within the TCR-gamma locus followed by electrophoretic separation of the product along a denaturing urea/formamide gradient gel. PCR-DGGE sensitivity was, to our knowledge, for the first time investigated for mixtures of lymphocytes (clonal and polyclonal) and keratinocytes. Clonal T-cells were detected in a concentration between 1-0.1% in keratinocytes, whereas the sensitivity was generally lower upon dilution in peripheral blood mononuclear cells or in a mixture of keratinocytes and peripheral blood mononuclear cells. Nevertheless, T-cell clonality was detected in 2 blood samples of a patient with Sézary syndrome, which were negative by Southern blot analysis. The crucial point of this work was the new approach to establish the sensitivity of the PCR-DGGE, in a way which more closely mimics the condition of clinical specimens. Instead of mixing and amplifying DNA extracted from clonal T-cell lines and polyclonal bone marrow cells, we amplified DNA from clonal and polyclonal cells which had been mixed in various ratios before DNA extraction. Polymerase chain reaction in conjunction with denaturing gradient gel electrophoresis is a sensitive and versatile molecular tool for the assessment of clonality of suspect cutaneous lesions. The determination of sensitivity using DNA extracted from premixed cells more closely corresponds to the actual test situation when testing skin samples.