General functionality of T cells in CMLTKIpatients and immunogenicity of HLA class II–restricted CML–associated antigens. (A) CD8+ T-cell counts for CML patients under TKI treatment (CMLTKI patients, n = 7) compared with HVs (n = 10) and CLL patients (n = 10). Retrospective analysis of preexisting immune responses directed against HLA class I–restricted (B) and HLA class II–restricted (C) viral T-cell epitopes (supplemental Table 3) analyzed in IFN-γ ELISPOT assays after a 12-day recall stimulation of PBMCs from CMLTKI patients (HLA class I, n = 10; HLA class II, n = 12), HVs (HLA class I, n = 14; HLA class II, n = 6), and CLL patients (HLA class I, n = 31; HLA class II, n = 24). (D) HLA class II–restricted CML-associated peptides with their corresponding source proteins and frequencies of preexisting immune recognition by CD4+ T cells from CML patients or HVs in IFN-γ ELISPOT assays after a 12-day stimulation. (E) Examples of CML-associated HLA class II–restricted peptides evaluated in IFN-γ ELISPOT assays using PBMCs from CML patients. Results are shown for immunoreactive peptides only. Phytohemagglutinin was used as positive control and the HLA class II–restricted FLNA_HUMAN1669-1683 peptide ETVITVDTKAAGKGK served as negative control. Because of low cell numbers, the results for UPN41 and UPN49 are shown as pool read-outs of all 6 HLA class II–restricted CML-associated peptides. Data are expressed as mean ± standard deviation of 2 independent replicates. ***P < .001. ID, identification; neg., negative; pos., positive; SFU, spot-forming unit; UPN, uniform patient number.