OC05.02: External validation of ultrasound‐based prediction models for discrimination between benign and malignant adnexal masses in Italy: IOTA phase 6 study

To prospectively evaluate the performance of RMI, Simple Rules risk model, ADNEX, and the IOTA two-step approach to estimate the risk of malignancy in different types of ultrasound centres in Italy. This is a multicentre prospective observational study including a mix of regional referral centres and district Italian hospitals. Consecutive patients with an adnexal mass examined with ultrasound by an IOTA certified ultrasound examiner (different levels of experience) were included. Discrimination (area under receiver operating characteristic curve, AUROC), calibration, and clinical utility were assessed to illustrate the ability of the four methods to discriminate between benign and malignant adnexal masses. 1431 patients were included from 21 Italian centres (12 oncological and 9 non-oncological). The outcome was benign for 995 (69.5%) patients and malignant for 436 (30.5%). The overall AUROC was 0.93 for ADNEX with CA125, the two-step strategy with CA125, and SRRisk, 0.92 for ADNEX without CA125 and the two-step strategy without CA125, and 0.87 for RMI. Calibration was best for SRRisk. RMI had the lowest utility to select patients for referral to a gynecological oncology centre. The other methods had similar utility for risk thresholds up to 0.2, for higher thresholds SRRisk had slightly higher utility. Subjective assessment had an overall sensitivity of 0.93 and specificity of 0.87. For subgroups based on menopausal status, type of centre (oncology vs non-oncology), and examiner experience, RMI had AUROCs between 0.84 and 0.90 whereas all other models had AUROCs between 0.90 and 0.96. Two step strategy with or without CA125, ADNEX with or without CA125 and SRRisk had similar performance to distinguish benign from malignant adnexal tumours in patients examined by both expert and non-expert operators.

Authors

Institutions

Publication Details

Journal
Ultrasound in Obstetrics and Gynecology
Published
2024-09-01
DOI
https://doi.org/10.1002/uog.27749
Primary Topic
Ovarian cancer diagnosis and treatment
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

OC05.02: External validation of ultrasound‐based prediction models for discrimination between benign and malignant adnexal masses in Italy: IOTA phase 6 study

D. Timmerman, Lucimara Fabiana Fornari, Roberta Granese, Tiziana Tagliaferri et al.
Ultrasound in Obstetrics and Gynecology
Ovarian cancer diagnosis and treatment
article

OC05.02: External validation of ultrasound‐based prediction models for discrimination between benign and malignant adnexal masses in Italy: IOTA phase 6 study

D. Timmerman, Lucimara Fabiana Fornari, Roberta Granese, Tiziana Tagliaferri, Francesca Ciccarone, Tom Bourne, Giovanni Scambia, F. Mascilini, F. Moro, E. Mor, Federica Pozzati, Tiziana De Grandis, Saverio Tateo, L. Valentin, Davide Sturla, Valentina Bertoldo, Fabio Ghezzi, R. Giacchello, Filippo Piccinini, Patrícia da Cruz Fávaro, I. Pezzani, B. Virgilio, Antonella Vimercati, E. Fruscella, Antonella Boschi, Daniele Garbin, Jolien Ceusters, Wouter Froyman, Ben Van Calster, Lorenzo Falchi, Olga Petruccelli, M. Momi, Patrizia Ragno, Annibale Mazzocco, Alison Ledger, A.C. Testa
article en

Abstract

To prospectively evaluate the performance of RMI, Simple Rules risk model, ADNEX, and the IOTA two-step approach to estimate the risk of malignancy in different types of ultrasound centres in Italy. This is a multicentre prospective observational study including a mix of regional referral centres and district Italian hospitals. Consecutive patients with an adnexal mass examined with ultrasound by an IOTA certified ultrasound examiner (different levels of experience) were included. Discrimination (area under receiver operating characteristic curve, AUROC), calibration, and clinical utility were assessed to illustrate the ability of the four methods to discriminate between benign and malignant adnexal masses. 1431 patients were included from 21 Italian centres (12 oncological and 9 non-oncological). The outcome was benign for 995 (69.5%) patients and malignant for 436 (30.5%). The overall AUROC was 0.93 for ADNEX with CA125, the two-step strategy with CA125, and SRRisk, 0.92 for ADNEX without CA125 and the two-step strategy without CA125, and 0.87 for RMI. Calibration was best for SRRisk. RMI had the lowest utility to select patients for referral to a gynecological oncology centre. The other methods had similar utility for risk thresholds up to 0.2, for higher thresholds SRRisk had slightly higher utility. Subjective assessment had an overall sensitivity of 0.93 and specificity of 0.87. For subgroups based on menopausal status, type of centre (oncology vs non-oncology), and examiner experience, RMI had AUROCs between 0.84 and 0.90 whereas all other models had AUROCs between 0.90 and 0.96. Two step strategy with or without CA125, ADNEX with or without CA125 and SRRisk had similar performance to distinguish benign from malignant adnexal tumours in patients examined by both expert and non-expert operators.

Ultrasound in Obstetrics and GynecologyVol. 64(S1)
Lund University (SE), Ospedale Filippo Del Ponte Varese (IT), Agostino Gemelli University Polyclinic (IT), Ospedale Santa Chiara (IT), Ospedale di Bolzano (IT), Ospedali Riuniti di Foggia (IT), Istituto Ramazzini (IT), Azienda Universitaria Ospedaliera Consorziale - Policlinico Bari (IT), Azienda Sanitaria Locale di Asti (IT), Presidio Ospedaliero (IT), Ospedale San Bassiano (IT), Skåne University Hospital (SE), Azienda Sanitaria Ospedaliera S.Croce e Carle Cuneo (IT), Policlinico Abano Terme (IT), Azienda Ospedaliera Universitaria Policlinico "G. Martino" (IT), Fondazione Poliambulanza Istituto Ospedaliero (IT), Candiolo Cancer Institute (IT), Policlinico San Matteo Fondazione (IT), Istituti di Ricovero e Cura a Carattere Scientifico (IT), Azienda Usl Toscana Centro (IT), Azienda Ospedaliera Universitaria Pisana (IT), Imperial College London (GB), KU Leuven (BE)
Gender equality
Openalex Percentile: Top 78%
Ovarian cancer diagnosis and treatment
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.