An ultrasound test may more reliably detect ovarian cancer in premenopausal women than the Risk of Malignancy Index

1. This cohort study found that the IOTA ADNEX ultrasound model at a 10% risk threshold detected more ovarian cancers in premenopausal women than the currently used Risk of Malignancy Index, making it a better test for triaging patients in secondary care. 

Evidence Rating Level: 1 (Excellent) 

Ovarian cancer is difficult to diagnose early, especially in premenopausal women, since symptoms are vague, and CA-125 can be elevated for benign reasons. Many women get diagnosed at an advanced stage, and in the United Kingdom, a large proportion present as emergencies, which is associated with worse survival. Accurate preoperative triage is crucial so that high-risk patients are referred to specialist cancer centres for appropriate first surgery, which improves outcomes. Current pathways rely on CA-125, ultrasound, and the Risk of Malignancy Index (RMI). However, these tests have important limitations and can miss early cancers or lead to unnecessary surgery. Evidence comparing newer models has been limited, particularly in premenopausal women, where fertility preservation and the lower cancer prevalence make the balance between detecting cancer and avoiding overtreatment more delicate. The ROCkeTS study was designed to address this gap by directly comparing several modern ultrasound- and biomarker-based risk models against RMI in women referred to secondary care with suspected ovarian cancer. The study’s main outcome measured how accurately tests identified primary invasive ovarian-type cancers versus benign or normal findings, while the secondary outcome used a broader definition of malignancy, which also included metastasis and borderline tumours. Out of 1,211 premenopausal women, 88 (7.3%) were diagnosed with primary ovarian cancer. Among these individuals, 5.7% (49 of 857) were diagnosed before June 2018, and 11.0% (39 of 354) after recruitment. Using the current standard RMI 1 (threshold 250), sensitivity was 42.6% and specificity 96.5%, meaning it missed more than half of cancers but produced few false positives. Other models detected more cancers; however, there were also more false positives. Sensitivity was 55.1% for CA-125, 79.2% for ROMA, 89.1% for IOTA ADNEX (10%), 83.0% for IOTA Simple Rules Risk (10%), and 75.0% for IOTA Simple Rules. Specificity dropped to 89.0% (CA-125), 73.1% (ROMA), 75.1% (ADNEX), 76.0% (Simple Rules Risk), while Simple Rules remained high at 95.2% but was inconclusive in 120 women. Overall, IOTA ADNEX at a 10% threshold is more effective at detecting ovarian cancers than RMI. 

Click here to read the study in BMJ

Image: PD

©2026 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.