Bone Marrow T Cell Subpopulations in Patients with Newly Diagnosed B-Cell Precursor Acute Lymphoblastic Leukemia (ALL).

Abstract

Besides its role in hemato- and lymphopoiesis, bone marrow (BM) has emerged as a secondary lymphoid organ with important roles in both T cell priming and memory responses. Due to these properties, non-malignant T cells persisting within the BM of patients with acute leukemias may be involved in the immune response to leukemia and the control of minimal residual disease. Here, we investigated the phenotypic signature of residual T cells present at diagnosis in 25 pediatric patients (age 2–16 years) with B cell precursor ALL. Patients with high risk disease including Philadelphia chromosome-positive or MLL-rearranged leukemias were excluded from this analysis. Mononuclear cells were isolated from freshly aspirated BM by density gradient centrifugation and analyzed by six-color-flow cytometry using monoclonal antibodies directed towards various T-cell associated surface and intracellular markers, including CD3/CD56/TCRαβ/TCRγδ/CD86/HLA-DR (Panel 1); CD3/CD4/CD8/CCR7/CD45RA/ CD45RO (Panel 2); CD4/FoxP3/CD25/CD45RA/CD45RO/CCR7 (Panel 3). For each sample, ≥15,000 CD3+ cells (Panel 1, 2) or ≥8,000 CD4+ cells (Panel 3) were analyzed with FACS Canto and Diva Software. The Student's t test was used to determine statistical significances between individual subgroups, and correlations were performed using the Pearson test. Consistent with published data on BM T cell subsets in healthy donors, the CD4+/CD8+ T cell ratio was 1.32±0.41. The predominant subset among CD8+ T cells (55.2±17.6%) had a naïve T cell phenotype (CD45RA+CCR7+), while 20.7±11.5% were effector memory T cells (TEM; CD45RA-CCR7-), and 7.1±6.2% were central memory T cells (TCM; CD45RA-CCR7+). No differences were found between TEL/AML1 positive or negative leukemias, or between patients stratified into standard (SR) vs. medium risk (MR) groups according to the criteria of the ALL-BFM 2000 study group. T cells bearing γδ T cell receptors have been attributed important roles in the primary immune defense against microbes and in immune control of cancer. We found that 6.9±3.0% (1.8 to 11.8%) of BM T cells were γδTCR+ (Vγ9Vδ2). A statistically significant (p <0.02) difference in the percentage of γδ T cells was found in patients stratified within risk groups MR and SR, respectively (7.0 vs. 4.5%). BM was further found to be a significant reservoir for regulatory T cells (Tregs). The presence of Tregs in the tumor microenvironment has been correlated with an unfavorable prognosis in many types of cancer, indicating a role in tumor immune escape. We found a proportion of 3.6±1.6% FoxP3+CD4+CD25high Tregs among BM CD4+ T cells from all ALL patients. Again, no significant differences were found according to TEL/AML1 status or risk stratification. Recent attention was drawn on the existence of both naïve and memory Treg subpopulations, differing by their expression of CD45RA and CCR7, and conflicting results were reported regarding the relative frequency of either subpopulation. We found a slight predominance of TEM Tregs (47.4±11.4%) above naïve (36.0±10.1%) and TCM Tregs (9.8±4.4%). In summary, while γδ T cells were found at significantly distinct numbers between patients stratified into standard and medium risk groups, no major phenotypic differences were found in residual BM αβ T cell subsets between individual subgroups in non-high-risk pediatric ALL. Future experiments will establish the functional role of the T cell subpopulations in immune control or escape in pediatric ALL.