Impact of Umbilical Cord Blood (UCB) T Regulatory Cells (Tregs) On Infection Risk Early After UCB Transplant

Abstract 4188

We have previously shown that the infusion of ex vivo expanded UCB-derived Tregs at a dose of ≥3 × 10⁶/kg immediately after UCB transplantation is associated with a reduced incidence of acute GVHD as compared to historical controls (p=0.05, Blood 2011; 117:1061). Together with a trend toward increased dual chimerism, these data support the conclusion that UCB Tregs are functional in vivo. We therefore asked the question whether Treg increased the increase risk of opportunistic infection (OI), particularly in the early transplant period when Treg are detectable in the peripheral blood. Opportunistic infections were defined as viral or fungal infections that were either documented or sufficiently supported by radiographic studies. For this analysis, patients receiving UCB-Tregs (n=24) were compared to historical controls (n=65) treated the same but without Tregs. All patients received a nonmyeloablative conditioning consisting of cyclophosphamide 50mg/kg/x1day, fludarabine 40 mg/m²/x5d, and total body irradiation 200 cGy/x1d and posttransplant immunosuppression consisting of mycophenate mofetil in combination with cyclosporine A (CsA, n=17 Treg patients and all controls) or sirolimus (n=7 Treg patients). Tregs were initially enriched by CD25+ selection (Miltenyi CliniMACS) from a partially HLA matched 3^rd UCB unit Tregs were activated and expanded for 18+1 days in the presence of anti-CD3/CD28 mAb coated beads and rhIL-2. 90% of patients received one or two doses of 3 × 10⁶/kg on d+1 (dose 1) and d+15 (dose 2) after UCB transplant. The median follow up of survivors is 2.6 (2.0–3.3) years and 4.8 (2.0–6.9) years for Treg and control patients, respectively. For non-Treg and Treg groups, the incidence of OI at day +100 was 58% (95%CI, 38–79%) and 57% (95%CI, 43–70%), respectively (p=0.74) at a median of 28 (21–87) and 41 (1–82) days after UCBT. Viral infections represented the majority of OI observed for Treg (58% [95%CI, 38–79%]) and non-Treg (55% [95%CI, 42–69%]) patient populations (p=.67). The overall day 0 to day +180 cumulative density of OI was 2.10 infections/1000 patient-days in Treg versus 1.52 infections/1000 patient-days in non-Treg patients (RR 1.39, p=.08). Notably, there was increased density of OI through day +30 with 18.06 infections/1000 patient-days for Treg patients versus 7.71 infections/1000 patient-days (RR 5.35, p=.02). In contrast, between day +31 and +180, the density of infection in Treg vs non-Treg patients was 4.22/1000 patient-days versus 7.22/1000 patient-days (RR 0.58, p=.07), respectively. The predominant viral infection observed in Treg patients through day +30 was HHSV-6 (n=9) with only two cases of CMV viremia and one case of parainflunenza upper respiratory infection). Despite the increased risk of early viral infections among Treg recipients, treatment related mortality at 1 year (18% vs. 10 %, p=0.36) and disease free survival at 2 years (33% vs. 34%, p=.95) were similar compared to historical controls. Interestingly, of the 7 recipients of Treg in the presence of sirolimus, none developed OI and 5/7 are alive. In conclusion, there is an increased risk of OI early after transplant in recipients of UCB Tregs and posttransplant pharmacological immunosuppression. Whether higher or multiple Treg doses may be more effective in suppressing acute GVHD, OI risk may be increased especially when used in combination with CsA. A dose escalation study UCB-derived Tregs expanded with KT86/64 artificial antigen presenting cells in the presence of sirolimus/MMF immune suppression is underway.