This retrospective study was conducted to examine whether the presence of Foxp3+ cells in biopsies of kidney transplants displaying early acute rejection (AR) predicted the outcome of the episode. Seventeen biopsies showing AR included in this study were obtained at 42 +/- 30 days after transplantation. Lesions were graded according to the Banff classification. Foxp3 staining was performed on paraffin-embedded sections with a monoclonal antibody after antigen retrieval. We evaluated relationships between the number and the location of Foxp3+ cells, the type of rejection, and the serum creatinine value at 1 year. Foxp3+ cells were detected in 11 of 17 biopsies with AR (9.5 +/- 13.3 cells/mm(2)). These elements were mixed with other interstitial inflammatory cells. Intraepithelial tubular Foxp3+ cells were seen in 9 biopsies (1.5 +/- 2.5 cells/mm(2)). Foxp3+ cells were associated with borderline lesions (25.5 +/- 22.4/mm(2)); type 1 AR (7.18 +/- 9/mm(2)) and type 2 AR (1.99 +/- 3.46/mm(2)). The average number of cells per field was not different in C4d(+) and C4d(-) AR (6 +/- 8.35 vs 8.5 +/- 14.7/mm(2)). Graft loss within the first year was higher among the group of recipients without Foxp3+ cells (3/6) than those with Foxp3+ cells (0/11). All AR with intraepithelial tubular Foxp3 cells had favorable outcomes. Foxp3 has been proposed as a relevant marker of CD4(+)CD25(+) regulatory T cells. This study showed that Foxp3+ cells can be detected in kidney transplant biopsies with AR. The absence of Foxp3+ cells, especially in epithelial tubular cells, might indicate a poor prognosis following an AR episode.