Diagnostic value of serum HE4 in endometrial cancer: a meta-analysis

Abstract

Background: Endometrial cancer (EC) is a common female malignant cancer. The age of incidence has become younger than before. If the diagnosis is during stage I, then the survival rate is about 90%. To date, there are no specific tumor markers for endometrial cancer. We usually use serum CA125 to help in diagnosing it. However, a serum biomarker CA125 greater than 35 U/ml is not useful in diagnosing EC at an early stage. Now, human epididymis protein 4 (HE4) has been intensively studied, and has been described as a new marker for ovarian cancer. The goal of this study was to evaluate the clinical value of serum HE4 in the diagnosis of endometrial cancer by meta-analysis.

Methods: We used MEDLINE, EMBASE, Cochrane Library and CBM databases to search the literature. The meta-analysis was performed by using Meta-Disc 1.4 software.

Results: All data we obtained showed that the major advantage of HE4 lies in its specificity in endometrial cancer diagnosis. Its sensitivity in serum was not as high as expected. But this evidence is not enough.

Conclusions: Additional studies, particularly to evaluate HE4's capability in identifying EC at an early stage, will be needed.

Figure 1

Selection process for articles included in the meta-analysis.

Figure 2

Forest plots of sensitivity of HE4 for EC prediction. The red circles represent the sensitivity of one study; the black line shows its confidence interval. If a study reported accuracy data for more than one cut-off, its results are included more than once.

Figure 3

Forest plots of specificity of HE4 for EC prediction. The red circles represent the specificity of one study; the black line shows its confidence interval. If a study reported accuracy data for more than one cut-off, its results were included more than once.

Figure 4

Diagnostic odds ratio of HE4 for EC prediction. Heterogeneity existed among the study designs. Cochran-Q is 22.13 in diagnostic odds ratio, P = 0.0024. Heterogeneity existed among the studies’ designs. Cochran-Q is 22.13 in diagnostic odds ratio, P = 0.0024.

Figure 5

SROC of HE4 for EC prediction. Each circle represents each study in the meta-analyses. The size of each study is indicated by the size of the circle. The regression summary receiver operating characteristic curves summarize the overall diagnostic accuracy.