Research news in clinical context

Combining the HPV genotype with epidemiological and clinical parameters markedly improves prediction of cervical cancer Researchers in China explored the added value of HPV genotyping in predicting cervical intraepithelial neoplasia (CIN). Data from 21 720 women (median age 50 years) with high-risk HPV formed a training dataset (n=14 553; 349 with CIN3+; 673 with CIN2+) and a validation dataset (n=7167; 167 with CIN3+; 228 with CIN2+). A machine learning model that added HPV genotypes to epidemiological factors (demographic characteristics, medical history, menstrual status, sexual behaviour factors, family history of cancer) and pelvic examination results showed the best ability to predict CIN3+ and CIN2+, outperforming other models that did not include the HPV genotype. Notably, some predictors were self-reported, smoking and oral contraceptive use were not collected, and women <30 years were not included. Further studies are needed to explore the clinical utility of the model (eg, in guiding colposcopy) and its...
Source: Sexually Transmitted Infections - Category: Sexual Medicine Authors: Tags: Perspectives Source Type: research