The relevance of the CD4+CD26-subset in the identification of circulating Sezary cells

Background The lack of specific markers for the phenotyping of circulating neoplastic T cells in Sezary syndrome (SS) patients makes it difficult both to ascertain the presence of clonal cells and to quantify the tumour burden in the peripheral blood. In previous reports we showed that the lack of CD26 (dipeptidyl-aminopeptidase IV) is a characteristic feature of circulating Sezary cells (SC). Objectives The purpose of this study was to ascertain, by means of high-resolution two-, three- or four-parameter flow cytometry, the relationship between CD26 expression on peripheral blood lymphocytes and peripheral blood involvement in cutaneous T-cell lymphoma patients and to assess its significance in SS diagnosis. Methods The patient population included 52 SS patients, 151 mycosis fungoides (MF) patients at different clinical stages (including 14 with blood involvement, B-1-MF), 88 patients with erythrodermic inflammatory skin diseases (EISD) and 72 healthy donors (HD). CD26+ values were available in all cases, whereas CD4+ CD26- level measurement was performed in 23 SS, 141 MF, 71 EISD and 72 HD. Results CD4+ CD26- percentage values were higher than 30% in all but one B-1-MF and higher than 40% in all SS cases, whereas HD, EISD and B-0-MF patient values were always lower than 30%. A statistically significant difference was found in both CD26- and CD4+ CD26- percentage and absolute values between SS and HD, EISD and B-0-MF patients. The CD26- and CD4+ CD26- percentage values (but not the absolute values) were significantly higher in B-1-MF compared with HD, EISD and B-0-MF patients (P < 0.001). Moreover, CD26- absolute values and CD4+ CD26- percentage and absolute values were significantly higher in SS than in B-1-MF (P < 0.001). A statistically significant direct relationship was found between CD4+ CD26- percentage values and the percentage of circulating SC within the lymphoid population in SS and B-1-MF (r = 0.77; P < 0.001). The lack of CD26 was confirmed on phenotypically clonal cells in patients with an expanded circulating TCRv beta population or a T-cell antigen loss. Sorted CD4+ CD26- cells from both SS patients and HD showed the characteristic cerebriform nuclei of SC. Conclusions We feel that a CD4+ CD26- percentage value higher than 30% of peripheral blood lymphocytes could correctly identify the presence of peripheral blood involvement in SS and MF patients.