Risk-associated alterations in marrow T cells in pediatric leukemia

Abstract

Current management of childhood leukemia is tailored based on disease risk determined by clinical features at presentation. Whether properties of the host immune response impact disease risk and outcome is not known. Here, we combine mass cytometry, single cell genomics, and functional studies to characterize the BM immune environment in children with B cell acute lymphoblastic leukemia and acute myelogenous leukemia at presentation. T cells in leukemia marrow demonstrate evidence of chronic immune activation and exhaustion/dysfunction, with attrition of naive T cells and TCF1+ stem-like memory T cells and accumulation of terminally differentiated effector T cells. Marrow-infiltrating NK cells also exhibit evidence of dysfunction, particularly in myeloid leukemia. Properties of immune cells identified distinct immune phenotype-based clusters correlating with disease risk in acute lymphoblastic leukemia. High-risk immune signatures were associated with expression of stem-like genes on tumor cells. These data provide a comprehensive assessment of the immune landscape of childhood leukemias and identify targets potentially amenable to therapeutic intervention. These studies also suggest that properties of the host response with depletion of naive T cells and accumulation of terminal-effector T cells may contribute to the biologic basis of disease risk. Properties of immune microenvironment identified here may also impact optimal application of immune therapies, including T cell-redirection approaches in childhood leukemia.

Keywords: Immunology; Leukemias; T cells.

Figure 1. Differences in BM T cells in children with B-ALL and AML at diagnosis.

BM mononuclear cells (BMMNCs) from patients with B-ALL (n = 36), AML (n = 28), and healthy donors (n = 11; n = 5 for 4-1BB) were characterized using single cell mass cytometry. (A) Figure shows percent naive (CCR7⁺CD45RO^–), central memory (TCM; CCR7⁺CD45RO⁺), effector memory (TEM; CCR7^–CD45RO⁺), and terminal effector (TERM Eff; CCR7^–CD45RO^–) CD8⁺ T cells in B-ALL and HD BM. (B) Percent naive, central memory, effector memory, and terminal effector CD8⁺ T cells in AML and HD BM. (C) Expression of CD69 on memory CD8⁺ T cells from B-ALL, AML, and HD marrow. (D) CD8⁺ T cells expressing 4-1BB in B-ALL, AML, and HD BM. (E) Figure shows expression of inhibitory immune checkpoints PD-1, TIGIT, and LAG3 on CD4⁺ and CD8⁺ T cells in B-ALL and HD BMMNCs. (F) Expression of inhibitory immune checkpoints PD-1, TIGIT, and LAG3 in CD4⁺ and CD8⁺ T cells from AML and HD BM. All graphs show mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 by Mann-Whitney U test.

Figure 2. Changes in BM T cell function in children with B-ALL and AML at diagnosis.

BMMNCs from B-ALL (n = 13), AML (n = 17), or HD (n = 5) were cultured alone or with PMA/ionomycin in the presence of GolgiStop. After 4 hours of culture, cells were stained with dead cell exclusion dye as well as antibodies to detect surface CD3, CD4, CD8, PD-1, TIGIT, intracellular IFN-γ, and IL-2 and analyzed using flow cytometry. (A) Proportion of CD8⁺ and CD4⁺ T cells secreting IFN-γ and IL-2. (B) IFN-γ secretion by cells expressing PD-1 and/or TIGIT. Figure shows a representative plot from patient with AML. All graphs show mean ± SEM. *P < 0.05 by Mann-Whitney U test with Bonferroni’s correction for multiple comparisons.

Figure 3. Changes in BM T cell transcription factors in children with B-ALL and AML at diagnosis.

BMMNCs from B-ALL (n = 13), AML (n = 17), or HD (n = 5) were cultured alone or with PMA/ionomycin in the presence of GolgiStop. After 4 hours of culture, cells were stained with dead cell exclusion dye as well as antibodies to detect surface CD3, CD4, CD8, PD-1, TIGIT, intracellular IFN-γ, and IL-2 and analyzed using flow cytometry. (A) Expression of TCF1, T-bet, GATA3, and EOMES transcription factors in memory CD4⁺ (left) and CD8⁺ (right) T cells in BMMNCs from HD and patients with B-ALL. (B) Expression of TCF1, T-bet, GATA3, and EOMES transcription factors in memory CD4⁺ (left) and CD8⁺ (right) T cells in BMMNCs from HD and patients with AML. (C) Tregs (CD3⁺CD4⁺CD25⁺CD127^–FOXP3^hi) as percentage of total T cells in BM from HD (n = 11) and patients with B-ALL (n = 36) and AML (n = 28). (D) Heatmaps showing characteristics of the TCF1^hi and TCF1^– CD8⁺ memory T cells from HD, B-ALL, and AML. All graphs show mean ± SEM. **P < 0.01, ***P < 0.001 by Mann-Whitney U test with Bonferroni’s correction for multiple comparisons.

Figure 4. Changes in BM NK cells.

(A–E) Expression of granzyme (GZM) (A), CD16 (B), CD57 (C), NKG2D (D), and TIM3 (E) on BM NK cells from HD (n = 11), as well as patients with B-ALL (n = 35) and AML (n = 26). (F) Histograms show change in surface expression of CD107a as a marker of NK cell degranulation upon NK cell culture alone (control) or with K562 cells. Left panels are representative patients, and the graph on the right shows data from several patients (B-ALL, n = 8; AML, n = 9). All graphs show mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 by Mann-Whitney U test with Bonferroni’s correction for multiple comparisons.

Figure 5. Immune correlates of clinical disease risk in B-ALL.

BMMNCs from patients with standard-risk (n = 22) and high-risk (n = 16) B-ALL were characterized using single cell mass cytometry. (A) Distribution of CD4⁺ and CD8⁺ T cells by risk status. (B) Frequency of naive, central memory (TCM), effector memory (TEM), and terminal effector (TERM Eff) CD8⁺ T cells by risk status. (C) viSNE density plots for patient groups (standard risk, n = 17; high risk, n = 9) visualized by concatenating FCS files for patients within each risk group. (D) viSNE plots showing expression of CD4, CD8, CCR7, CD45RO, T-bet, granzyme (GZM), and CD57 from a representative patient. (E) Expression of T-bet in memory CD4⁺ and CD8⁺ T cells from B-ALL patients with standard-risk or high-risk disease. All graphs show mean ± SEM. *P < 0.05, **P < 0.01 by Mann-Whitney U test.

Figure 6. Immune correlates of clinical disease risk in AML.

BMMNCs from patients with low-risk (n = 14) and high-risk (n = 14) AML were characterized using single cell mass cytometry. (A) Distribution of CD4⁺ and CD8⁺ T cells by risk status. (B) Frequency of naive, central memory (TCM), effector memory (TEM), and terminal effector (TERM Eff) CD8⁺ T cells by risk status. (C) viSNE density plots for patient groups (low risk, n = 10; high risk, n = 10) visualized by concatenating FCS files for all patients within each disease risk group. (D) Expression of T-bet in memory CD4⁺ and CD8⁺ T cells from AML patients with low-risk or high-risk disease. All graphs show mean ± SEM. *P < 0.05, **P < 0.01 by Mann-Whitney U test.

Figure 7. Distinct immune clusters associate with disease risk and outcome in childhood B-ALL.

BMMNCs from patients with standard-risk (SR, n = 22) and high-risk (HR, n = 16) B-ALL were characterized using single cell mass cytometry. (A) Hierarchical cluster analysis based on immune markers in B-ALL. (B) Mosaic plot showing distribution of HR and SR B-ALL patients in the 2 clusters (relative frequency in cluster 1 is in red and cluster 2 is in yellow). P value corresponds to Wald’s test. (C) Bar graph showing distribution of TEL-AML⁺ B-ALL patients in the immune clusters. P value corresponds to Wald’s test.

Figure 8. Distinct immune clusters associate with disease risk and outcome in childhood AML.

BMMNCs from patients with AML (n = 28) were characterized using single cell mass cytometry. (A) Forest plot showing Cox regression analysis of immune markers with progression-free survival in AML. (B and C) Progression-free survival of AML patients with above- versus below-median proportions of CD27⁺CD4⁺ T cells of total CD4⁺ T cells (B) and CD27⁺CD8⁺ T cells (C) of total CD8⁺ T cells. P values correspond to log-rank test.

Figure 9. Single cell transcriptome analysis of BM mononuclear cells in childhood B-ALL and AML.

A total of 24,081 BM mononuclear immune cells from 19 samples (4 healthy donor, 8 AML, and 7 B-ALL) was characterized using single cell mRNA sequencing. (A) Uniform manifold approximation and projection (UMAP) plot with 29 distinct cell populations determined by unsupervised clustering (T, T cells; NK, NK cells; B, B cells; EP, erythroid progenitor cells; HSC, hematopoietic stem cells). (B) UMAP plot distinguishing cells by disease state.

Figure 10. Single cell transcriptome analysis of BM mononuclear cells in childhood B-ALL and AML.

A total of 24,081 BM mononuclear immune cells from 19 samples (4 healthy donor, 8 AML, and 7 B-ALL) were characterized using single cell mRNA sequencing. (A) Volcano plot of differential gene expression between T cell cluster T1 (naive/stem like enriched in HD) and T cell cluster T2 (effector-like enriched in malignancy). Adjusted P value (p_val_adj) corresponds to Wilcoxon rank-sum test with Bonferroni’s correction. Positive log-fold change (avg_logFC) corresponds to higher expression in T1 relative to T2. (B) Cells in each T/NK cell cluster as a percentage of total T/NK cells by disease state. (C) Distribution of T cells in cluster T1 and T2 in B-ALL patients based on NCI disease risk (standard risk; high risk). (D) Expression of granzyme (GZMA) and perforin (PRF1) in NK cells from healthy donors (HD) and AML patients (AML). *P < 0.01, **P < 0.0001 by Wilcoxon rank-sum test with Bonferroni’s correction. (E) Volcano plot of genes differently regulated between tumor cells from patients with T cells enriched for naive/stem like phenotype (T1 T cells) versus tumor cells from B-ALL patients with BM enriched for terminally differentiated effector T cells or T2 cluster.