Article Text
Abstract
Objectives For patients with endometrial cancer (EC), the screening value of serum human epididymis protein 4 (HE4) remains controversial. We performed meta-analyses to compare the screening accuracy of serum HE4 and carbohydrate antigen 125 (CA125) for EC.
Materials and Methods A search of diagnostic test studies was performed in 5 English databases: Pubmed, Cochrane Library, Web of Science, Science Direct, and Elton Bryan Stephens Co or EBSCO; and 2 Chinese databases including China National Knowledge Infrastructure or CNKI and VIP (Weipu Database), from their inception dates to early July 2015. Two reviewers independently selected trials, conducted critical appraisal, and extracted data. Meta-analyses were performed to compare the screening accuracy between HE4 and CA125. Summary receiver operating characteristic curve and the area under the summary receiver operating characteristic curve were performed. Subgroup analysis, meta-regression, sensitivity analysis, and Egger plot and the Egger test were also conducted.
Results Twenty-one studies were identified, and the methodological quality was generally fair. Meta-analyses revealed that the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratios for HE4 in screening EC were 0.56, 0.89, 6.41, 0.49, and 14.82, respectively, whereas the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio for CA125 in screening EC were 0.32, 0.81, 2.15, 0.83, and 2.74, respectively. The areas under the summary receiver operating characteristic curves for HE4 and CA125 were 0.7778 and 0.5474, respectively.
Conclusions This study indicates that serum HE4 may be superior to CA125 in screening accuracy of EC. This conclusion has to be interpreted cautiously owing to high heterogeneity and some limitations.
- Serum human epididymis protein 4
- Carbohydrate antigen 125
- Endometrial cancer
- Diagnostic test
- Meta-analysis
Statistics from Altmetric.com
Footnotes
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net).
Ethical approval: not required.