Data were analyzed from 23 patients with Sézary syndrome (defined by erythroderma, more than 10% circulating atypical mononuclear cells, and peripheral blood T-cell clone) undergoing monthly extracorporeal photopheresis as the sole therapy for up to 1 year. The cohort showed a significant reduction of skin scores during treatment (P =.001). Thirteen patients (57%) achieved a reduction in skin score greater than 25% from baseline at 3, 6, 9, or 12 months (responders). Reduction in skin score correlated with reduction in the Sézary cell count as a percentage of total white cell count (P =.03). Responders and nonresponders were compared. None of the measured parameters was significantly different between the 2 groups. It was assessed whether any of the baseline parameters predicted outcome. A higher baseline lymphocyte count was significantly associated with a decrease in skin score at 6 months (P <.05). A higher baseline Sézary cell count as a percentage of total white cell count predicted a subject was more likely to be a responder after 6 months of treatment (P =.021). No other parameters predicted responder status. These data show that the modest falls in CD4, CD8, and Sézary cell counts were seen in all patients and might have resulted from lymphocyte apoptosis. This mechanism could explain the more favorable response seen in patients with higher percentages of Sézary cells in the peripheral blood. Alternatively, minimum tumor burden might be required for the induction of a cytotoxic response. Analysis of tumor-specific cytotoxic T cells is needed to investigate these possibilities further.