Ovarian cancer prediction in adnexal masses using ultrasound‐based logistic regression models: a temporal and external validation study by the IOTA group

• DOI: 10.1002/uog.7636
• Publication Date: 2010-03-30T09:25:04Z
• Authors: D. Timmerman; B. Van Calster; A. C. Testa; S. Guerriero; D. Fischerova; A. A. Lissoni; C. Van Holsbeke; R. Fruscio; A. Czekierdowski; D. Jurkovic; L. Savelli; I. Vergote; T. Bourne; S. Van Huffel; L. Valentin
• Keywords: Adnexal mass; Gold standard (test)

Abstract Objectives The aims of the study were to temporally and externally validate diagnostic performance two logistic regression models containing clinical ultrasound variables in order estimate risk malignancy adnexal masses, compare results with subjective interpretation findings carried out by an experienced examiner (‘subjective assessment’). Methods Patients who put forward 19 centers participating study, underwent a standardized transvaginal examination gynecologist or radiologist specialized ultrasonography. prospectively collected information on variables, classified each mass as benign malignant basis evaluation findings. gold standard was histology local clinicians deciding whether operate picture. models' ability discriminate between masses assessed, together accuracy estimates. Results Of 1938 patients included 1396 had benign, 373 primary invasive, 111 borderline 58 metastatic tumors. On external validation (997 from 12 centers), area under receiver–operating characteristics curve (AUC) for model predictors (LR1) 0.956, reduced six (LR2) 0.949 assessment 0.949. Subjective gave positive likelihood ratio 11.0 negative 0.14. corresponding ratios previously derived probability threshold (0.1) 6.84 0.09 LR1, 6.36 0.10 LR2. temporal (941 seven AUCs 0.945 (LR1), 0.918 0.959 (subjective assessment). Conclusions Both provide excellent discrimination masses. Because objective reasonably accurate estimation, they may improve management women suspected ovarian pathology. Copyright © 2010 ISUOG. Published John Wiley & Sons, Ltd.