Meta-analysis of the diagnostic accuracy of HE4 for endometrial carcinoma

To analyze and evaluate the value of serum human epididymis protein 4 (HE4) for the diagnosis of endometrial cancer (EC).

Studies involving HE4 and the diagnosis of EC were retrieved from the following medical literature databases: Medline, PubMed, Web of Science, China National Knowledge Infrastructure, China Biology Medicine Disc, Vip Journal Integration Platform, and Wanfang Data Knowledge Service Platform. Quality assessment was performed independently by two reviewers using Review Manager 5.3 (Cochrane Collaboration Group). A quality table of included studies was made using Review Manager 5.3, and the pooled sensitivity (SEN), specificity (SPE), positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic accuracy, and receiver operating characteristic curve (SROC) were analyzed using Review Manager 5.3 and Meta-Disc 1.4 software.

Of 887 studies, 17 passed quality assessment and were included in the final study. The pooled SEN was 0.65 (95 % confidence interval [CI]: 0.63-0.67), SPE was 0.913 (95 % CI: 0.92-0.95), PLR was 10.06 (95 % CI: 4.75-21.35), NLR was 0.41 (95 % CI: 0.33-0.50), diagnostic odds ratio (DOR) was 26.7 (95 % CI: 11.7-60.93), and the area under curve (AUC) of the receiver operating characteristic curve (SROC) curve was 0.75 (95 % CI: 0.81-0.87).

HE4 is a potential biomarker for the diagnosis of EC with a high specificity and relatively low sensitivity. Considering high heterogeneity and selection bias, the value of HE4 for diagnosing EC should be further evaluated in strictly-designed diagnostic studies as well as in different pathological types and stages of EC.