Fig. 2

Patients who received PTCy showed a distinct T cell immune signature post-allo-HSCT. A Data of sixty-one nonredundant variables, including the frequencies of immune cell subsets as well as T-cell phenotypes, transcription factors and functions were collected via flow cytometry and analyzed by PCA algorithms. Two components, PC1 and PC2, capture the most and second most variation of the parameters, respectively. Each dot represents the corresponding value from one timepoint of a patient and was colored according to its group and timepoint. The circles denote the confidence intervals of specific groups at the level of 0.68. The arrow represents each variable, and the direction displays its contribution to the principal components. P: PTCy group; NP: non-PTCy group. B Volcano plot of the above-mentioned 61 immune parameters analyzed in PTCy relative to non-PTCy samples. Red and green dots denote the statistically significant (adjusted P < 0.05) parameters that are twofold higher or ½ fold lower than non-PTCy samples, respectively. The expression of surface inhibitory molecules C Ki67 D and IFN-γ production E of CD4⁺/CD8⁺ T cells are shown through the box-and-whiskers plots. P value of the comparison between the PTCy versus non-PTCy group was calculated using Wilcoxon signed-rank test and was corrected for multiple comparisons using the Benjamini–Hochberg adjustment. F Immune cell components, phenotypes and functions at 30 days after allo-HSCT were compared between patients who were relapsed post-transplant (R, n = 5) or patients who had no relapse (NR, n = 17). Data that have significant differences between the 2 groups (Granzyme B and Perforin intracellular expression in CD8 T cells) are shown here. G Immune cell components, phenotypes and functions 30 days after allo-HSCT were compared between two groups of patients: no clinically significant GVHD (grade 0–1 aGVHD and mild/moderate cGVHD, n = 17); clinically significant GVHD (grade 2–4 aGVHD and severe cGVHD, n = 5). Parameters that have statistical significance or trend are shown. The value of each parameter is normalized to a mean of 0 and standard deviation of 1. Each column represents an individual patient, and each row represents an immune marker. Relative over-expressed and under-expressed values are denoted as red and blue, respectively. The dendrograms were constructed via hierarchical clustering, and patient GVHD stages are separated as indicated by the bars at the top. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001