Skip to main content
. 2019 Dec 10;3(23):4117–4130. doi: 10.1182/bloodadvances.2019000835

Figure 4.

Figure 4.

NK cell diversity correlates with risk for relapse after CBT. (A) Lack of correlation between the PB-NK cell repertoire diversity and absolute PB-NK cell count (103/μL) in matched samples collected from CBT recipients. Red circles denote patients who relapsed in the first year post-CBT (n = 9), and blue circles those who remained in complete remission (CR; n = 34). The Pearson’s correlation test was used to calculate the correlation coefficient (r) and 2-sided P value. (B) Comparison of the phenotypic complexity of NK cells in PB samples collected in the first 100 days after CBT in patients in CR and those who relapse in the first year posttransplant, using t-SNE-based dimensionality reduction. t-SNE analysis distributed NK cells from patients in CR vs relapse along 2 t-SNE axes based on normalized expression of 32 NK cell markers per single cell. NK cells from patients in CR are on the left side of the map, and relapsed patients on the right side. Inset values indicate the frequency of cells that fall within the CR or relapse gate (top 2 t-SNE maps). The composite t-SNE maps include NK cells from patients who remained in CR (top left blue t-SNE map) vs those who relapsed in the first year posttransplant (top right blue t-SNE map) show normalized expression levels of the indicated markers from a representative individual. Cells were colored according to normalized expression intensity. Color scales indicate signal intensity, ranging from low (blue) to high (red) after arcsine transformation. Bars represent median with interquartile range and summarize the frequencies of expression of each marker of NK cells from patients who remained in CR (blue) and those who relapsed (red). *P ≤ .05; **P ≤ .01; ***P ≤ .001; ****P ≤ .0001.