The present investigation was conducted to examine the effect of neoadjuvant PVB and CAP regimens for locally invasive bladder cancer and consisted of two studies: (1) a retrospective nonrandomized study of neoadjuvant PVB therapy, and (2) a well-controlled randomized study of neoadjuvant CAP therapy. A total of 25 patients with primary locally invasive bladder cancer were entered into the PVB study between January 1981 and December 1985. Since 1986, 31 patients have been randomized into the CAP study. In the PVB-treated group, a 71.4% complete response (CR) plus partial response (PR) rate and a 71.4% downstaging were noted. On the other hand, in the CAP-treated group, a 50.0% CR plus PR rate and a 88.9% downstaging were noted. The 2- and 5-year survival rates of neoadjuvant PVB were 78.6 and 60.6%, respectively. In contrast, the 2-year survival rate of the neoadjuvant CAP-treated group was 100% at a mean follow-up of 15.8 months. No statistical significance was achieved in the survival rates. These results indicated that neoadjuvant PVB and CAP would be useful in the management of invasive bladder cancer.