Abstract
Although it is well known that tumor site- or bone marrow-infiltrating regulatory T cells (Tregs) might be correlated with worse outcomes in solid tumors and acute leukemias by promoting immune surveillance escape, their contribution to the immediate post-allogeneic transplantation phase by peripheral blood (PB) allografts remains unclear. Moreover, the Treg content in stem cells harvested from PB has been suggested to be correlated with acute graft versus-host-disease (aGVHD) and immunologic recovery after allogeneic PB stem cell transplantation (allo-PBSCT). This study aimed to investigate the impact of the graft content of Tregs, as graft CD3+/Tregs ratio (gCD3/TregsR), on acute GVHD and post-allo-PBSCT outcomes. We prospectively enrolled 94 consecutive patients at 9 Italian centers of the Gruppo Italiano Trapianto di Midollo Osseo (GITMO) with acute myelogenous (n = 71; 75%) or lymphoblastic (n = 23; 25%) leukemia in complete remission who underwent matched related donor (n = 35; 37%) or unrelated donor (n = 59; 63%) allo-PBSCT. The median graft CD3+ cell, Treg, and gCD3/TregsR values were 196 × 106/kg body weight (range, 17 to 666 × 106/kg), 3 × 106/kg (range, 0.1 to 35 × 106/kg), and 71 (range, 1 to 1883), respectively. The discriminatory power of the gCD3/TregsR value to predict grade ≥II aGVHD was assessed by estimating the area under the receiver operating characteristic (ROC) curve (AUC). Any grade and grade ≥II aGVHD occurred in 24 (26%) and 17 (18%) allo-PBSCT recipients, respectively. By ROC analysis, AUC (0.74; 95% confidence interval [CI], 0.608 to 0.866; P = .002) identified 70 as the optimal gCD3/TregsR cutoff value predicting the appearance of grade ≥II aGVHD with 76% sensitivity and 71% specificity. Patients were subdivided into a high (ROC curve value ≥70) gCD3/TregsR group (HR; n = 48) and a low (ROC curve value <70) gCD3/TregsR group (LR; n = 46). The incidence of grade II-IV aGVHD was lower in the LR group compared with the HR group (9% [4 of 46] versus 27% [13 of 48]) in both univariate analysis (odds ratio [OR], 4.8; 95% CI, 1.44 to 16.17; P = .015) and multivariate analysis (OR, 5.0; 95% CI, 1.34 to 18.93; P = .017), whereas no differences were documented taking into account aGVHD of any grade. The overall survival, disease-free survival, nonrelapse mortality, and relapse rates at 2 and 3 years were 61% and 54%, 62% and 55%, 15% and 23%, and 27% and 30%, respectively. Of note, gCD3/TregsR did not significantly correlate with relapse (P = .135). Taken together, our data from this prospective multicenter study confirm the value of Tregs in preventing aGVHD while maintaining the graft-versus-leukemia effect. © 2021 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc.