CMV reactivation influences relative frequencies and absolute counts of NKG2C⁺ and NKG2A⁺ NK cells in recipients of TCD HCT. (A) Dot plots of NKG2A vs NKG2C NK cells from representative recipients of TCD HCT with and without CMV reactivation. (B) Percentages of NKG2C⁺ (left) and NKG2A⁺ (right) NK cells of total NK cells and (C) absolute counts of NKG2C⁺ (left) and NKG2A⁺ (right) CD56⁺CD16⁺ NK cells of recipients of TCD HCT are shown at sequential time points (mean ± standard error of the mean [SEM]). Two-way analysis of variance (ANOVA) with multiple comparisons was used for statistical analysis, with ∗∗P ≤ .01; ∗∗∗P ≤ .001; ∗∗∗∗P ≤ .0001. (D) Percentages of NKG2C⁺ NK cells of total NK cells at late time points (≥ 180 days after HCT) in recipients of TCD HCT with (n = 16) or without (n = 32) persistent CMV infection. Unpaired/nonparametric t test was used for statistical analysis, with ∗P ≤ .05. (E) Percentage of NKG2C⁺ NK cells (left) and patients with first detection of NKG2C⁺ NK cells (percentages >4% were considered positive for NKG2C expression) at indicated time points relative to CMV detection (T0) (right) (n = 16). Data from each recipient of TCD HCT with CMV reactivation are plotted individually over time (left). T0 is the first immune-monitoring time point after CMV detection. Immune-monitoring time points before CMV detection were, therefore, denoted reverse chronologically as T−1 and T−2, and immune-monitoring time points after CMV detection were denoted from T1 to T4.