Combining a symptom index, CA125 and HE4 (triple screen) to detect ovarian cancer in women with a pelvic mass.
By: Barbara A Goff, Kathy Agnew, Moni Blazej Neradilek, Heidi J Gray, John B Liao, Renata R Urban

Department of Obstetrics and Gynecology, University of Washington, Seattle, WA 98195, United States. Electronic address: bgoff@uw.edu.
2017-06-24; doi: 10.1016/j.ygyno.2017.08.020
Abstract

Objectives

To assess a simple algorithm of CA125, HE4 and Symptom Index to predict ovarian cancer in women with a pelvic mass.

Methods

This was a prospective study of women referred to a gynecologic oncology clinic for surgical evaluation of a pelvic mass. Preoperatively, women completed a SI and had serum markers drawn. Results were correlated with pathology. A triple screen was considered positive if at least 2 of the 3 markers were abnormal (positive SI, CA125≥35U/mL, HE4≥140pmol/L).

Results

218 patients enrolled in the study. 66 patients (30%) had ovarian or fallopian tube cancer (97% epithelial), 124 (57%) had benign masses, 17 (8%) had borderline tumors, and 11 (5%) had metastatic disease. The SI, CA125 and HE4 were positive in 87.9%, 74.2% and 60.6% of ovarian cancer patients, respectively. Of the 112 women with a positive SI 58 (52%) had ovarian cancer and 75 (67%) had non-benign masses. Excluding borderline and metastatic cancers the sensitivity of the triple screen was 79%; specificity 91%, PPV 83% and NPV 89%. CA125 alone had a sensitivity, specificity, PPV and NPV of 79%, 76%, 63% and 87% respectively. Requiring only one of the three tests to be abnormal resulted in a sensitivity of 97% but specificity dropped to 50%.

Conclusions

An algorithm using SI, CA125 and HE4 has good performance statistics for predicting cancer in women with pelvic masses. The triple screen has higher specificity and PPV than CA125 alone but similar sensitivity and NPV for predicting ovarian cancer.



Copyright © 2017. Published by Elsevier Inc.

PMID:28860006






Copyright 2026 InterMDnet | Privacy Policy | Disclaimer | System Requirements