The use of appropriate antigenic peptides for the most common human major histocompatibility complex (MHC) alleles is required for the amplification of the autologous cytotoxic compartment and the development of cytotoxic T cell-mediated immunity. The human A2 allele of the MHC plays an important role for the identification of peptide-specific cytotoxic T cells (CTL) against tumor and viral epitopes. Computer-based prediction algorithms, which are available on the Internet, have already proved to be applicable for the identification of novel CTL epitopes. Using the bioinformatics approach, the authors have identified the novel influenza matrix protein-derived and HLA-A3-restricted 9-mer peptide RLEDVFAGK capable of inducing peptide specific CTL reactivity. Peripheral blood mononuclear cells (PBMC) from healthy individuals and patients with lung cancer were pulsed with this peptide and with the well-characterized HLA-A2-restricted influenza A virus matrix peptide58–66GILGFVFTL. Using quantitative PCR (TaqMan; Applied Biosystems, Foster City, CA, U.S.A), reactivity for both peptides was determined by measuring the change in type 1 cytokine (IFN-&ggr;) expression upon in vitro stimulation. Peptide-specific reactivity matched well with the subsequently determined MHC-class I alleles of the tested individuals. Results from this study indicate that the use of bioinformatics and the PCR-based screening system for the monitoring of T cell reactivity may allow for the identification of novel CTL epitopes.