A randomized double-blind placebo-controlled phase II trial was conducted to evaluate the efficacy of a prophylactic quadrivalent vaccine targeting the human papillomavirus (HPV) types most frequently associated with cervical cancer (types 16/18) and genital warts (types 6/11) in Japanese women aged 18–26 years. Participants were randomly assigned to either quadrivalent HPV (types 6/11/16/18) L1 virus-like particle vaccine (GARDASIL) (n = 509) or placebo (n = 512). Participants underwent regular gynecological examinations, cervicovaginal sampling for HPV DNA, testing for serum neutralizing antibodies to HPV and Papanicolau testing. The primary end-point was the combined incidence of persistent infection with HPV types 6, 11, 16 or 18 and cervical or external genital disease (i.e. cervical intraepithelial neoplasia, cervical cancer or external genital lesions related to HPV 6, 11, 16 or 18. Primary analyses were done per protocol. Combined incidence of persistent infection or disease with HPV 6, 11, 16 or 18 fell by 87.6% (95% confidence interval [CI], 59.2–97.6; P < 0.001), with HPV 6 or 11 by 73.1% (95% CI, −1.1–97.3; P = 0.0756) and with HPV 16 or 18 by 94.5% (95% CI, 65.2–99.9; P < 0.001) in those assigned vaccine compared with those assigned placebo. The median duration of follow up after month 7 in subjects was 23 months. In addition, the vaccine was well tolerated in Japanese women aged 18–26 years. Quadrivalent HPV vaccine could significantly reduce the acquisition of infection and clinical disease caused by HPV types 6, 11, 16 and 18.