Purpose: The purpose of this study was to determine the diagnostic accuracy of tumor markers for malignancy in girls with ovarian neoplasms.
Methods: A retrospective review of girls 2-21 years who presented for surgical management of an ovarian neoplasm across 10 children's hospitals between 2010 and 2016 was performed. Patients who had at least one concerning feature on imaging and had tumor marker testing were included in the study. Sensitivity, specificity, and negative and positive predictive values (PPV) of tumor markers were calculated.
Results: Our cohort included 401 patients; 22.4% had a malignancy. Testing for tumor markers was inconsistent. AFP had high specificity (98%) and low sensitivity (42%) with a PPV of 86%. The sensitivity, specificity, and PPV of beta-hCG was 44%, 76%, and 32%, respectively. LDH had high sensitivity (95%) and Inhibin A and Inhibin B had high specificity (97% and 92%, respectively).
Conclusions: Tumor marker testing is helpful in preoperative risk stratification of ovarian neoplasms for malignancy. Given the variety of potential tumor types, no single marker provides enough reliability, and therefore a panel of tumor marker testing is recommended if there is concern for malignancy. Prospective studies may help further elucidate the predictive value of tumor markers in a pediatric ovarian neoplasm population.
Type of study: Retrospective Cohort Review.
Level of evidence: Level III.
Keywords: Oophorectomy; Ovarian neoplasms; Ovary-sparing surgery; Preoperative risk; Tumor markers.
Copyright © 2019 Elsevier Inc. All rights reserved.