To predict the efficacy of the M-VAC neoadjuvant chemotherapy for invasive bladder cancers, we previously established the method to calculate the prediction score on the basis of expression profiles of 14 predictive genes. This scoring system had clearly distinguished the responder group from the non-responder group. To further validate the clinical significance of the system, we applied it to 22 additional cases of bladder cancer patients and found that the scoring system correctly predicted clinical response for 19 of the 22 test cases. The group of patients with positive predictive scores had significantly longer survival than that with negative scores. When we compared our results with a previous report describing the prognosis of the patients with cystectomy alone, the results imply that patients with positive scores are likely to benefit from M-VAC neoadjuvant chemotherapy, but that the chemotherapy would shorten the lives of patients with negative scores. We are confident that our prediction system to M-VAC therapy should provide opportunities for achieving better prognosis and improving the quality of life of patients. Taken together, our data suggest that the goal of 'personalized medicine', prescribing the appropriate treatment regimen for each patient, may be achievable by selecting specific sets of genes for their predictive values according to the approach shown here.