There are limited data on the prevalence and distribution of human papillomavirus (HPV) genotypes in vaginal intraepithelial neoplasia (VAIN). We sought to clarify this issue in a series of 450 VAIN cases with a confirmed diagnosis between 1990 and 2006. HPV genotyping was performed using paraffin-embedded specimens and polymerase chain reaction (PCR)-based methods. Multiple HPV types were validated by E6 type-specific PCR and direct sequencing. The HPV genotypes of the vaginal and cervical neoplasms were compared for those with incident VAIN and a history of previous/concomitant cervical neoplasms. Ki-67 was performed for supporting diagnosis of VAIN. Of these 450 VAIN cases (median age, 59 years; range, 19-93), two with missing paraffin blocks and 54 with poor DNA quality were excluded. HPV was detected in 273/394 (69.3%) VAIN, and multiple infections were found in 17.9% of HPV-positive samples. The leading types were HPV16 (35.5%), HPV58 (9.9%), HPV52 (9.9%), HPV39 (8.4%), HPV33 (7.3%) and HPV53 (7.0%). Among the 156 cases with a history of previous cervical neoplasia, 29.0% had concordant HPV genotypes, while synchronous VAIN samples (n 5 49) were more likely to harbor concordant genotypes (58.7%) with the concomitant cervical neoplasm (p 5 0.0003). Whether those HPV types in the incident VAIN lesions had existed in the vaginal epithelium at the time of the previous cervical neoplasia or a new acquisition needs to be clarified in prospective follow-up studies.The natural history of invasive vaginal cancer and preinvasive lesions (vaginal intraepithelial neoplasia [VAIN]) has been minimally investigated compared to invasive cervical cancer and cervical intraepithelial neoplasia (CIN). 1,2 Many previous research efforts have indicated that invasive vaginal cancer and VAIN share common risk factors with cervical neoplasms, including the number of lifetime sexual partners, smoking and infection with human papillomavirus (HPV). 1,3,4 Previous radiation exposure, a previous history of cervical neoplasms, diethylstilbestrol exposure and an immune-compromised state were also associated with an increased risk of vaginal cancer and VAIN. 1,[4][5][6][7] Previous studies regarding HPV types in VAIN and vaginal cancer or comparisons of HPV types between incident vaginal and previous cervical lesions were limited by small sample sizes and inconsistent conclusions. 3,4,[8][9][10][11][12][13][14][15] Smith et al. 8 published a systemic review of 725 abstracts, in which the HPV detection rates among 166 cases of VAIN2/3 and 66 cases of VAIN1 were 92.6% and 98.5%, respectively. Another systemic review of 22 U.S. studies found HPV16/18 to be the predominant type in VAIN3 (n ¼ 97, 65.1%). 10 A meta-analysis of 93 studies 9 including 298 cases of VAIN from four continents noted that HPV testing was positive in 100% of VAIN1 and 90.1% of VAIN2/3. In VAIN2/3 (n ¼ 191), the most common HPV types were HPV16 (57.6%), HPV 18 (6.9%) and HPV 58 (5.9%). 9 With the advent of HPV vaccines, cervical neoplasms of the HPV types included in th...