Objective A systematic review was performed to estimate the accuracy of human epididymis protein 4 (HE4) assay in the diagnosis of ovarian tumors.

Methods A comprehensive search of the MEDLINE (PubMed), EMBASE, Cochrane Central Register of Controlled Trials, IBECS, BIOSIS, Web of Science, SCOPUS, congress abstracts, and Grey literature (Google scholar; British Library) from January 1990 to April 2013 was conducted. Studies that evaluated HE4 levels for the diagnosis of ovarian tumors and compared them with paraffin-embedded sections as the diagnostic standard were included.

Results Forty-five studies were analyzed, which included 10,671 women and 3946 ovarian cancer cases. The pooled sensitivity for the diagnosis of borderline tumors or ovarian cancer was 78% (95% confidence interval, 77%–79%), and the specificity was 86% (95% confidence interval, 85%–87%). Summary receiver operating characteristic curves were constructed. For malignant and borderline ovarian tumors versus benign lesions, the area under the curve was 0.916. Besides the overall analysis, stratification was performed in premenopause and postmenopause, early and late stages, and for accuracy by enzyme-linked immunosorbent assay and chemiluminescence microparticle immuno assay.

Conclusions A HE4 level is a useful preoperative test for predicting the benign or malignant nature of pelvic masses.