Abstract 2237Poster Board II-214 Background:GVHD occurs unpredictably after allogeneic stem cell transplant (ASCT). GVHD is characterized by profound immune dysregulation, which suggests that abnormalities in suppressor lymphocytes known as Tregs may account for the clinical features of this disease. We have shown that the frequency of Tregs or α4β7± Tregs at neutrophil engraftment predict for recurrent grade II-IV GVHD or recurrent gut GVHD during the first 100 days of transplant, respectively. These data indicate that early lymphocyte populations may determine long-term immune-tolerance following ASCT. We hypothesized that Treg subsets present at engraftment would be associated with the occurrence, National Institutes of Health (NIH) phenotype, and severity of GVHD after day±100. Methods:Patients (pts) undergoing T cell replete ASCT were enrolled. GVHD after day±100 was classified prospectively as either: acute GVHD, overlap GVHD, or classic chronic GVHD using consensus criteria derived from the NIH. Clinical severity was determined by the modified Glucksberg criteria for acute GVHD and the NIH global assessment of severity for overlap/classic chronic GVHD. The frequency of CD45RO±CD25±Foxp3±CD127lo Tregs was quantified within the CD4± T-lymphocyte population at the time of neutrophil engraftment using polychromatic flow cytometry. Gut-homing Tregs were identified by the presence of α4β7±, and their frequency was expressed as a percentage of the total Treg population. Results:Treg analysis was performed at a median of 19 days post-transplant (range, 10-31) on 41 pts undergoing ASCT and surviving until day±100. Twenty-four (59%) pts received ablative conditioning and 17 (41%) pts received a reduced intensity regimen followed by matched related [N= 24 (59%)] or unrelated donor [N= 17 (41%)] ASCT. The stem cell source was peripheral blood, bone marrow, or cord blood for 30 (73%), 9 (22%), and 2 (5%) pts, respectively. GVHD prophylaxis consisted of calcineurin inhibitor and methotrexate [N=23 (56%)] or mycophenolate mofetil [N=18 (44%)]. The majority of patients had grade II-IV acute GVHD prior to day±100 [N=34 (83%)]. After a median follow-up of 15 months (range, 5-24) GVHD after day±100 occurred in 32 (78%) pts at a median of 152 days post ASCT (range, 103-533). GVHD phenotype was classified as: acute (N=19), overlap (N=9), or classic chronic (N=4). Grade III-IV acute and moderate to severe overlap/classic chronic GVHD occurred in 3 (7%) and 11 (27%) pts, respectively. Pts with GVHD after day±100 had lower frequencies of Tregs and α4β7± Tregs at engraftment (4.4% vs. 14.3%; P=0.145) and (9.9% vs. 20.7%; P=0.009), respectively. After excluding pts with gut GVHD, the association between α4β7± Tregs and GVHD after day±100 persisted (P=0.034). Median Treg percentages did not differ among GVHD phenotypes [acute (4.4%) vs. overlap (4.2%) vs. classic chronic (5%) vs. none (14.3%); P=0.464]. However, significant differences between α4β7± Treg frequencies and GVHD subtypes were noted [acute (11%) vs. overlap (9.6%) vs. classic chronic (11.6%) vs. none (20.7%); P=0.002]. α4β7± Treg differences were most apparent when pts with any acute GVHD (acute±overlap) were compared to pts without features of acute GVHD (none±classic chronic) (9.9% vs. 19.4%; P=0.009). Increased frequencies of Tregs and α4β7± Tregs at engraftment tended to be associated with lower grades of acute GVHD after day±100 [odds ratio (OR), 0.95; 95% confidence interval (CI), 0.87-1.04; P=0.255 (Tregs) and OR, 0.81; 95% CI, 0.70-0.92; P=0.002 (α4β7± Tregs)] and less severe overlap/classic chronic GVHD [OR, 0.91; 95% CI, 0.79-1.04; P=0.158 (Tregs) and OR, 0.93; 95% CI, 0.86-1.01; P=0.071 (α4β7± Tregs)]. Using Cox Proportional Hazard regression (adjusting for donor type and acute GVHD prior to day ±100), increased frequencies of α4β7± Tregs at engraftment continued to be associated with decreased risk of GVHD after day ±100 (hazard ratio, 0.93; 95%CI, 0.88-0.99; P=0.013). Conclusion:Frequency of α4β7± Tregs at engraftment is associated with the incidence and severity of GVHD after day ±100. These data suggest that early lymphocyte subsets and gut associated mucosal immunity are important in determining long-term graft tolerance. Treg subsets at engraftment could be used to risk stratify pts prior to development of GVHD or as an endpoint in future clinical trials examining interventions aimed at increasing Tregs and preventing GVHD. Disclosures:No relevant conflicts of interest to declare.