Rapid In Vivo Acquisition of Effector Tc1 Phenotype Even before Myeloid Engraftment Predicts Opportunistic Infections (OI) in Unrelated Cord Blood Transplant Recipients

Unrelated cord blood transplantation (UCBT) is a viable option for patients who lack HLA-matched sibling donors. However, opportunistic infections (OI), occurring in the first 100 days, remain the major cause of morbidity and mortality. Viral infections are the primary cause of OI death. Despite the large number of T cells in UCB grafts there is no established antiviral immunity in nave UCB lymphocytes available for adoptive transfer. UCB T cells also lack expression of Th1/Tc1 cytokines, Granzymes A, B, and Perforin which are prerequisite cytotoxic effector molecules for controlling viral pathogens. We postulated that following transplant the infused naïve T cells will be critically influenced by the lymphopenic environment and emerging de novo acquired or latent viruses. Phenotypic alterations induced by these pathogens may identify those at risk for OI even before virological/clinical diagnosis is evident. Since our previous analysis of d+ 50 immune profile could not fully separate true risk factors from immunological correlates/consequences of OI we designed a new study. Methods: 70 pediatric recipients of HLA-mismatched UCBT were studied prospectively as soon as the WBC exceeded 0.3/ul, at a median of day+18, typically before myeloid engraftment and prior to the onset of clinical infections. Unprocessed whole blood was stained in “Trucount” (B-D) tubes for absolute T cell and dendritic cell (DC1, DC2) counts, while T cells subsets were similarly analyzed by 4 color FACS from Ficolled specimens for lymphocyte subsets, their activation state, tissue homing molecules. Intracellular (ic) staining identified cytotoxic effectors and inducible cytokine secretion besides characterizing homeostatic proliferation and apoptosis. 2-tailed student’s t-test identified significant variables (p Results: 39/70 patients developed OI at a median of day +35 (37 viral cases, 1 case of Legionella and 1 Candida), in average 2weeks after immune studies. While only 2/31 non-OI patients died (relapse), 18/39 from the OI group died at a median of 82 days post UCBT, p=0.001. Median f/up of survivors is 28 ± 17 months. Age, malignancy, cell dose had no impact on OI presence. The median values for major leukocytes were; WBC: 0.7×10e3/ul, T cells: 66/ul, NK cells: 46/ul, B cells: 0/ul, CD4+T cells: 42/ul, %CD4+ T cells: 64%. Both the OI+ and OI negative group had comparable WBC, CD3+, CD4+ T cells, NK lymphocytes, or DC1, DC2 subsets. Strikingly, 44% of circulating T cells were in cell cycle (KI-67+) regardless of OI status reflecting vigorous homeostatic expansion with only 16% preserving the CD45RA+/CD62L+ phenotype. Notably, ~9% of the circulating T cells were entering apoptosis (activated Caspase-3+), regardless of OI status. In those who eventually developed OI significantly more CD8+ cells circulated (40% vs 28%, p=0.03), more expressed CCR-5 (82% vs 55%, p=0.009), and more T cells were secreting IFNγ (35% vs 12%, p=0.006). In patients who developed OI significantly more Perforin+/CD8+ T cells circulated (49% vs 26%, p=0.01) and more of them expressed Granzyme B (p=0.05). There was higher BCL-2 expression in T cells from patients developing OI (p=0.04), and greater presence of CD28−/CD57+/CD8+ ‘effector’ CTLs (p=0.014). We conclude that in lymphopenic UCBT recipients subclinical viral infections may skew T cell maturation allowing early identification those at risk for clinical OI.