Immunotherapy has become a promising strategy for the treatment of malignancies, as it harnesses the naturally occurring immune defences against cancer. There is an increasing interest in the search for specific anti-tumor targets for cancer immunotherapy. Neo-antigens are mutated proteins that arise from the intrinsic genomic instability of cancer cells and are not expressed in normal cells. Consequently, targeting neo-antigens allows to selectively destroy cancer cells while sparing normal ones. Importantly, neo-antigens are tumor-specific and therefore differ among patients, which means that personalized production protocols are needed. Recent studies published in Nature (Ott et al., 2017; Sahin et al., 2017) demonstrated the feasibility of personalized cancer vaccines against neo-antigens. The treatment was beneficial for a group of advanced melanoma patients (10 and 13 respectively) enrolled in the studies. Nevertheless, clinical trials on a greater number of patients are required to confirm these data. A further reduction of costs and time may be accomplished by optimizing the pipeline of neo-antigen prediction: once amplified and sequenced, tumor DNA is compared with normal samples from the same patient, in order to identify mutated proteins; bioinformatical algorithms are then used to predict which proteins are the most likely to elicit a strong immune response.