Abstract / Summary
Abstract Introduction Vaginal intraepithelial neoplasia 2 or worse (VaIN2+) has long remained in a largely passive diagnostic pattern due to the lack of specific clinical manifestations and routine screening strategies. This study aimed to explore its clinical characteristics and develop a prediction model for individualized risk assessment. Material and Methods We retrospectively analyzed the clinical characteristics and human papillomavirus (HPV) genotype distribution in 3259 women undergoing colposcopy at a single center between September 2012 and December 2024. A nomogram was developed based on predictors initially screened by least absolute shrinkage and selection operator regression and finalized through multivariable logistic regression. The model's discriminative ability, calibration, and clinical utility were assessed in both training and validation sets. Results Among 1192 women with VaIN2+, older age, HPV infection, cytology worse than atypical squamous cells of undetermined significance (ASCUS), and concurrent cervical/vulvar lesions were more prevalent than in 2067 women without vaginal lesions. HPV16, HPV58, HPV52, HPV33, and HPV18 were the most common genotypes. A nomogram was developed based on 10 independent predictors identified by multivariable analysis, including age, hysterectomy history, specific HPV genotypes (16, 58, 56, 35, 39), cytological results (ASCUS and worse), concurrent cervical lesions (cervical intraepithelial neoplasia 1, cervical intraepithelial neoplasia 2 or 3, cervical squamous cell carcinoma), and concurrent vulvar lesions. The nomogram demonstrated good discrimination, with an area under the receiver operating characteristic curve of 0.771 in the training set and 0.791 in the validation set. Calibration was satisfactory, and decision curve analysis confirmed superior clinical net benefit across a wide threshold probability range. Conclusions By profiling VaIN2+ characteristics, this study developed the first prediction model integrating specific HPV genotypes with multiple clinical factors. This model enables clinicians to proactively identify high‐risk individuals, thereby facilitating earlier diagnosis and intervention for VaIN2+.