A recent study found that ultrasonography-based risk models for differentiation between benign and malignant ovarian tumors performed well in a diagnostic study conducted in a US cohort.
Ultrasonography-based risk models for differentiatiation between benign and malignant ovarian tumors performed well in a diagnostic study conducted in a US cohort, according to a study published in JAMA Network Open.
Authors of the retrospective diagnostic study aimed to determine whether 3 ultrasonography-based risk-stratification models primarily used in Europe would perform well in a US population. The models assessed in the study—the Simple Rules model, Assessment of Different Neoplasias in the Adnexa (ADNEX), and the Ovarian-Adnexal Reporting and Data System (O-RADS)—have previously shown high performance in European populations but have limited uptake in the United States.
mi_viri - stock.adobe.com
“These models were developed primarily for nonexperts to ease sonographic assessments, standardize reports, and improve consistency,” the authors wrote. “In a busy clinical practice, these models enable the nonexpert clinician to distill the complex presentation of adnexal masses into smaller, objective, simple variables, thus reducing the number of indeterminate reports that often lead to surgeries for benign lesions.”
The main outcome of the study was each model’s diagnostic performance measured by area under the receiver operating characteristic (ROC) curve (AUC), sensitivity analyses, specificity, positive predictive value, negative predictive value, and positive and negative likelihood ratios. Researchers also explored the models’ accuracies in pre- vs postmenopausal women and in racially diverse populations in secondary analyses.
A total of 511 female patients were included in the study cohort, 15.9% of whom had malignant tumors (n = 81). In the overall cohort, 200 (39.1%) were postmenopausal. The mean (SD) age of patients with benign tumors was 44.1 (14.4) years, and the mean age of those with malignant adnexal lesions was 52.5 (15.2) years. The cohort included both patients treated surgically and conservatively.
The ROC analysis found an AUC for discriminative performance of 0.96 (95% CI, 0.93-0.98) for the ADNEX model and 0.92 (95% CI, 0.90-0.95) for the O-RADS model. When the ADNEX continuous individualized risk was converted into the discrete ordinal categories of O-RADS, the AUC of the ADNEX performance was very similar to that of the O-RADS, at 0.93 (95% CI, 0.90-0.96). When the Simple Rules model was combined with expert reevaluation, the sensitivity was 93.8% (95% CI, 86.2%-98.0%) and specificity was 91.9% (95% CI, 88.9%-94.3%). When Simple Rules was combined with malignant classification, the sensitivity was 93.8% (95% CI, 86.2%-98.0%) and specificity was 88.1% (95% CI, 84.7%-91.0%).
ADNEX demonstrated 91.4% sensitivity (95% CI, 83.0%-96.5%) and 86.3% specificity (95% CI, 82.7%-89.4%), at a 10% risk threshold, compared with the O-RADS model’s 98.8% sensitivity (95% CI, 93.3%-100%) and 74.4% specificity (95% CI, 70.0%-78.5%) with the same risk threshold.
In a subgroup analysis stratified by menopausal groups, the sensitivities of the models were similar between groups. However, the specificities were significantly lower among postmenopausal women. There were no significant differences between sensitivities in White vs Black women for any of the models, and the overall performances of the models were high regardless of patient race.
The study is the largest to the authors’ knowledge to assess ultrasonography-based risk models for the evaluation of adnexal lesions in a US cohort, and all 3 models tested showed high performance. The evaluation of clinically relevant subgroups marked a strength of the study, although the authors noted the study’s single-center, retrospective design as a limitation. Other limitations were the retrieval of additional sonograms of patients with adnexal masses, which may have introduced selection bias, and that some patients had uncertain findings at follow-up and were included based on subjective expert assessment. Results of a sensitivity analysis excluding this group showed similar findings to the primary analysis, however.
“In this diagnostic study of ultrasonography-based risk models to differentiate between benign and malignant adnexal lesions, the Simple Rules, ADNEX, and O-RADS models performed well in the same US cohort, although they are currently rarely used across the US,” the authors concluded. “…While all models showed high diagnostic accuracy, ADNEX has further clinical advantages, such as assigning individual numerical malignant tumor risk that would allow more tailored management and estimation of the likelihood of malignant subclasses, thereby enhancing personalized care.”
Reference
Yoeli-Bik R, Longman RE, Wroblewski K, et al. Diagnostic performance of ultrasonography-based risk models in differentiating between benign and malignant ovarian tumors in a US cohort. JAMA Netw Open. 2023;6(7):e2323289. doi:10.1001/jamanetworkopen.2023.23289
"The Barriers Are Real": Antoine Keller, MD, on Geography and Cardiovascular Health
April 18th 2025Health care disparities are often driven by where patients live, explained Antoine Keller, MD, as he discussed the complex, systematic hurdles that influence the health of rural communities.
Read More
Personalized Care Key as Tirzepatide Use Expands Rapidly
April 15th 2025Using commercial insurance claims data and the US launch of tirzepatide as their dividing point, John Ostrominski, MD, Harvard Medical School, and his team studied trends in the use of both glucose-lowering and weight-lowering medications, comparing outcomes between adults with and without type 2 diabetes.
Listen
Racial Differences in CA-125 Levels Tied to Ovarian Cancer Treatment Delays
April 17th 2025Black and American Indian women with ovarian cancer were less likely to have elevated cancer antigen 125 (CA-125) levels at diagnosis, resulting in delayed chemotherapy initiation and highlighting the need for more inclusive guidelines.
Read More
Empowering Teams Begins With Human Connection: Missy Hopson, PhD
April 16th 2025Missy Hopson, PhD, Ochsner Health, discussed in detail the challenges of strengthening the patient-centered workforce, the power of community reputation for encouraging health care careers, and the influence of empowered workforces on patient outcomes.
Read More