Scientific Session 04 — Genitourinary - Reproductive/EndocrineMonday, May 1, 2017
3035. Can Ultrasound Predict Malignancy in an Adnexal Mass?
Morani A*, Sun J, Wei W, Faria S, Bhosale P. MD Anderson Cancer Center, Houston, TX
Address correspondence to P. Bhosale (Priya.Bhosale@mdanderson.org)
Objective: The purpose of this study was to determine the ability of ultrasound to independently differentiate benign versus malignant ovarian and adnexal masses and survival based on ultrasound, tumor marker CA125, and pathology results.
Materials and Methods: Retrospective institutional review board approval was obtained, and 182 randomly selected women from our radiology database with a history of cancer other than ovarian cancer who presented with an adnexal mass were included in the study. Women who did not undergo surgical excision of the pelvic mass were excluded. The mass or masses in the adnexa were documented as benign or malignant on the basis of ultrasound findings. The indeterminate adnexal masses were recorded as malignant. Masses containing intramural nodules with flow on Doppler evaluation, septations thicker than 3 mm, and vascularity within the septations were considered malignant. Patient demographics, age, pathology results, and CA125 values were recorded. CA125 serum blood level less than 35 U/mL was considered normal. All patients underwent surgical resection. Wilcoxon rank sum test was used to compare the difference in CA125 for benign and malignant masses. Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of ultrasound were estimated using pathology results as a reference standard. Overall survival was estimated using the Kaplan-Meier method and was compared using a log-rank test.
Results: Mean patient age was 50.6 years (range, 16–83 years). CA125 levels were significantly higher in patients who had malignant tumors according to pathology (p = 0.0001) and ultrasound findings (p = 0.0003). Accuracy, sensitivity, specificity, PPV, and NPV of ultrasound were 82%, 39%, 93%, 60%, and 85%, respectively. Patients with malignant tumors had significantly worse overall survival according to pathology (p = 0.009) and ultrasound findings (p = 0.001). Patients with CA125 levels of 35 U/mL or higher had worse overall survival than those whose levels were below 35 U/mL, but the difference was not statistically significant (p = 0.093). Median survival was 3.9, 7.6, and 7.6 years for malignant tumors based on ultrasound, pathology, and CA125 level, respectively. The 5-year survival rates were 47% (95% CI, 23–93%), 65% (95% CI, 47–90%), and 69% (95% CI, 50–97%), respectively.
Conclusion: Ultrasound had a high specificity of 93% and accuracy of 82%. Patients who were predicted to have ovarian malignancy on the basis of ultrasound findings had worse overall survival probability than those predicted to have malignancy on the basis of CA125 level or pathology findings. This study shows that ultrasound has the ability to predict survival on the basis of the imaging characteristics of the adnexal mass.