Search tips
Search criteria 


Logo of bmccancBioMed Centralsearchsubmit a manuscriptregisterthis articleBMC Cancer
BMC Cancer. 2012; 12: 258.
Published online Jun 19, 2012. doi:  10.1186/1471-2407-12-258
PMCID: PMC3443004
Does risk for ovarian malignancy algorithm excel human epididymis protein 4 and ca125 in predicting epithelial ovarian cancer: A meta-analysis
Fake Li,#1 Ruxiu Tie,#1 Kai Chang,1 Feng Wang,1 Shaoli Deng,1 Weiping Lu,1 Lili Yu,corresponding author2 and Ming Chencorresponding author1
1Department of Clinical Laboratory, Institute of Surgery Research, Daping Hospital, Third Military Medical University, ChongQing, Peoples Republic of China
2Department of Obstetrics and Gynecology, Institute of Surgery Research, Daping Hospital, Third Military Medical University, ChongQing, Peoples Republic of China
corresponding authorCorresponding author.
#Contributed equally.
Fake Li: lfk2346/at/; Ruxiu Tie: tieruxiu112/at/; Kai Chang: changkai0203/at/; Feng Wang: wangfnew/at/; Shaoli Deng: dengsl072/at/; Weiping Lu: ronnylu/at/; Lili Yu: lililiyuyu/at/; Ming Chen: chenming1971/at/
Received January 5, 2012; Accepted May 23, 2012.
Risk for Ovarian Malignancy Algorithm (ROMA) and Human epididymis protein 4 (HE4) appear to be promising predictors for epithelial ovarian cancer (EOC), however, conflicting results exist in the diagnostic performance comparison among ROMA, HE4 and CA125.
Remote databases (MEDLINE/PUBMED, EMBASE, Web of Science, Google Scholar, the Cochrane Library and and full texts bibliography were searched for relevant abstracts. All studies included were closely assessed with the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2). EOC predictive value of ROMA was systematically evaluated, and comparison among the predictive performances of ROMA, HE4 and CA125 were conducted within the same population. Sensitivity, specificity, DOR (diagnostic odds ratio), LR ± (positive and negative likelihood ratio) and AUC (area under receiver operating characteristic-curve) were summarized with a bivariate model. Subgroup analysis and sensitivity analysis were used to explore the heterogeneity.
Data of 7792 tests were retrieved from 11 studies. The overall estimates of ROMA for EOC predicting were: sensitivity (0.89, 95% CI 0.84-0.93), specificity (0.83, 95% CI 0.77-0.88), and AUC (0.93, 95% CI 0.90-0.95). Comparison of EOC predictive value between HE4 and CA125 found, specificity: HE4 (0.93, 95% CI 0.87-0.96) > CA125 (0.84, 95% CI 0.76-0.90); AUC: CA125 (0.88, 95% CI 0.85-0.91) > HE4 (0.82, 95% CI 0.78-0.85). Comparison of OC predictive value between HE4 and CA125 found, AUC: CA125 (0.89, 95% CI 0.85-0.91) > HE4 (0.79, 95% CI 0.76-0.83). Comparison among the three tests for EOC prediction found, sensitivity: ROMA (0.86, 95%CI 0.81-0.91) > HE4 (0.80, 95% CI 0.73-0.85); specificity: HE4 (0.94, 95% CI 0.90-0.96) > ROMA (0.84, 95% CI 0.79-0.88) > CA125 (0.78, 95%CI 0.73-0.83).
ROMA is helpful for distinguishing epithelial ovarian cancer from benign pelvic mass. HE4 is not better than CA125 either for EOC or OC prediction. ROMA is promising predictors of epithelial ovarian cancer to replace CA125, but its utilization requires further exploration.
Articles from BMC Cancer are provided here courtesy of
BioMed Central