After allogeneic hematopoietic stem cell transplantation (HSCT), immune recovery is important to protect the patient from relapse and co-morbidities such as graft-versus-host disease (GVHD) and infection. Various numbers of low-frequency immunocompetent cells are known to exist among T cells and each subset shows different immunological action. Among them, γδ T cells were reported to facilitate a graft-versus-leukemia (GVL) effect and regulatory T cells (Tregs) were reported to prevent acute GVHD. In this study, we focused on the clinical relevance of γδ T cells and Tregs in peripheral blood (PB) after allogeneic HSCT in patients with hematological neoplasm to outcome.

We retrospectively analyzed 33 adult patients with hematological neoplasms who underwent allogeneic HSCT between July 2011 and February 2015 at Niigata University Medical and Dental Hospital, including 17 with acute myeloid leukemia, 8 with acute lymphoblastic leukemia, 4 with myelodysplastic syndromes, 2 with Epstein-Barr virus-associated lymphoproliferative disorder, 1 with adult T-cell leukemia/lymphoma and 1 with primary myelofibrosis. Circulating γδ T cells and Tregs were analyzed by flow cytometry within 30-100 days after allogeneic HSCT. γδ T cells were identified as CD3+/γδTCR+ cells. Tregs were identified as CD4+/Foxp3+/CD25+ cells. The percentage of γδ T cells was calculated by dividing by CD3+ cells. The percentage of Tregs was calculated by dividing by CD4+ cells. The Kaplan-Meier method was used to estimate the probability of disease-free survival (DFS). The Mann-Whitney U test was used to compare the percentage of γδ T cells in PB or Tregs in PB and any grade of acute GVHD or grade II-IV acute GVHD. Cumulative relapse rate and non-relapse mortality (NRM) were based on Gray's estimates. Fine-Gray proportional hazards models were used for assessment by multivariate analysis of relapse rate. Factor adjustment was performed for age, conditioning regimen, disease status and HLA compatibility.

The median percentage of γδ T cells divided by CD3+ cells in PB was 3.3% (0-28.4%). The median percentage of Tregs divided by CD4+ cells in PB was 1.9% (0-17.3%). The percentage of γδ T cells in PB was not associated with the incidence of acute GVHD. In addition, the percentage of Tregs in PB was not associated with the incidence of acute GVHD. Next, we established an immune scoring system according to the percentage of γδ T cells and Tregs in PB. Less than 4% of γδ T cells as a proportion of CD3+ cells in PB was scored as 1 point and more than 4% of Tregs as a proportion of CD4+ cells in PB was scored as 1 point. The patients with 1 point for γδ T cells did not show a significant difference to the patients with 0 points in terms of cumulative relapse rate or NRM. In addition, the patients with 1 point for Tregs did not show a significant difference to the patients with 0 points in terms of cumulative relapse rate or NRM. We classified the patients into score 0-1 and score 2 upon adding the points. Patients with score 2 showed a higher relapse rate in univariate analysis (p=0.002) and multivariate analysis (hazard ratio 3.65, p=0.017) than patients with score 0-1. Moreover, patients with score 2 showed higher DFS in univariate analysis (p=0.001) and multivariate analysis (hazard ratio 3.50, p=0.027) than patients with score 0-1.

Our study suggests that the combination of a low rate of γδ T cells and a high rate of Tregs in PB after allogeneic HSCT is a poor prognostic factor for patients with hematological neoplasm. In addition, it suggests that the balance of immunosuppression and immunoactivation may be important for the outcome of patients after allogeneic HSCT.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution