Telomerase reverse transcriptase (TRT) is the first bona fide common tumor antigen. While several 9mer peptides of the human TRT have been identified for HLA-A2, little information exists on peptides for the remaining HLA types. Here, we used a multi-step approach to select and characterize a panel of HLA-B7 9mer peptides as candidate immunogens. In sequence, we used algorithm-based predictions, in vivo immunization of HLA-B7 transgenic (Tg) mice, in vitro immunization of human blood lymphocytes from two normal donors and two cancer patients, in vivo processing in HLA-B7 Tg mice and HLA-B7 supertype binding. We found a correlation between the in vivo immunogenicity and the actual HLA-B7 binding avidity of the seven predicted peptides. Furthermore, endogenous processing correlated with in vitro immunogenicity in human PBMC and HLA-B7 supertype binding. Peptide (1123)LPSDFKTIL(1131) (p1123) with the wider spectrum of supertype binding displayed the highest immunogenicity overall and was endogenously processed in several human lymphoblastoid cells. Since no single step of the screening/selection process could substitute for the whole approach, we conclude that the identification of MHC class I-restricted peptides for potential vaccination of cancer patients remains, by and large, an empirical process.