Related Experiment Video
Updated: Jun 14, 2026

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
Predicting risk of malignancy in adnexal masses
John M McDonald1, Stacey Doran, Christopher P DeSimone
1From the Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, and Departments of Statistics and Biostatistics, University of Kentucky Chandler Medical Center-Markey Cancer Center, Lexington, Kentucky.
Objective:
To estimate the accuracy of preoperative ultrasonography, serum CA 125, and patient demographics as a means of predicting risk of malignancy in women with a ultrasonographically confirmed adnexal mass.
Methods:
Tumor morphology derived from ultrasonographic images, tumor size, tumor bilaterality, serum CA 125, and patient demographics were evaluated preoperatively in 395 patients undergoing surgery from 2001 to 2008. Tumor morphology was classified as complex, solid, or cystic. Preoperative findings were compared with tumor histologic findings at the time of surgery. Multivariable classification and regression tree analysis were used to identify a group of patients at high risk of ovarian malignancy.
Results:
One hundred eighteen patients had ovarian cancer, 13 patients had ovarian tumors of borderline malignancy, and 264 had benign ovarian tumors. Multivariable classification and regression tree analysis defined women at high risk of ovarian malignancy as those with an adnexal mass having complex or solid morphology and a serum CA 125 value greater than 35 units/mL. This definition had a positive predictive value of 84.7% and a negative predictive value of 92.4% and correctly identified 77.3% of patients with stage I and stage II ovarian cancer and 98.6% of patients with stage III and stage IV ovarian cancer.
Conclusion:
Patients with solid or complex ovarian tumors and an elevated serum CA 125 level (greater than 35 units/mL) are at high risk of ovarian malignancy.
Level Of Evidence:
II.
