The Wilms' tumor 1 (WT1) antigen is expressed in solid and hematological malignancies, but not healthy tissues, making it a promising target for cancer immunotherapies. Immunodominant WT1 epitopes, the native HLA-A2/WT1126-134 (RMFPNAPYL) (HLA-A2/RMFPNAPYL epitope (WT1A)) and its modified variant YMFPNAPYL (HLA-A2/YMFPNAPYL epitope (WT1B)), can induce WT1-specific CD8+ T cells, although WT1B is more stably bound to HLA-A*02:01. Here, to further determine the benefits of those two targets, we assessed the naive precursor frequencies; immunogenicity and cross-reactivity of CD8+ T cells directed toward these two WT1 epitopes. Ex vivo naive WT1A- and WT1B-specific CD8+ T cells were detected in healthy HLA-A*02:01+ individuals with comparable precursor frequencies (1 in 105-106) to other naive CD8+ T-cell pools (for example, A2/HIV-Gag77-85), but as expected, ~100 × lower than those found in memory populations (influenza, A2/M158-66; EBV, A2/BMLF1280-288). Importantly, only WT1A-specific naive precursors were detected in HLA-A2.1 mice. To further assess the immunogenicity and recruitment of CD8+ T cells responding to WT1A and WT1B, we immunized HLA-A2.1 mice with either peptide. WT1A immunization elicited numerically higher CD8+ T-cell responses to the native tumor epitope following re-stimulation, although both regimens produced functionally similar responses toward WT1A via cytokine analysis and CD107a expression. Interestingly, however, WT1B immunization generated cross-reactive CD8+ T-cell responses to WT1A and could be further expanded by WT1A peptide revealing two distinct populations of single- and cross-reactive WT1A+CD8+ T cells with unique T-cell receptor-αβ gene signatures. Therefore, although both epitopes are immunogenic, the clinical benefits of WT1B vaccination remains debatable and perhaps both peptides may have separate clinical benefits as treatment targets.