Serum HE4 and diagnosis of ovarian cancer in postmenopausal women with adnexal masses

Aleksandra Gentry-Maharaj, Matthew Burnell, James Dilley, Andy Ryan, Chloe Karpinsky, Richard Gunu, Susan Mallett, Jon Deeks, Stuart Campbell, Ian Jacobs, Sudha Sundar, Usha Menon

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)
137 Downloads (Pure)


BACKGROUND: Transvaginal ultrasound and serum CA125 are routinely used for differential diagnosis of pelvic adnexal mass. Use of human epididymis 4 was approved in the United States in 2011. However, there is scarcity of studies evaluating the additional value of human epididymis 4.

OBJECTIVE: The objective of the study was to evaluate the performance characteristics of transvaginal ultrasound, CA125, and human epididymis 4 for differential diagnosis of ovarian cancer in postmenopausal women with adnexal masses.

STUDY DESIGN: This was a cohort study nested within the screen arms of the multicenter randomized controlled trial, United Kingdom Collaborative Trial of Ovarian Cancer Screening, based in England, Wales, and Northern Ireland. In United Kingdom Collaborative Trial of Ovarian Cancer Screening, 48,230 women randomized to transvaginal ultrasound screening and 50,078 to multimodal screening (serum CA125 interpreted by Risk of Ovarian Cancer Algorithm with second line transvaginal ultrasound) underwent the first (prevalence) screen. Women with adnexal lesions and/or persistently elevated risk were clinically assessed and underwent surgery or follow-up for a median of 10.9 years. Banked samples taken within 6 months of transvaginal ultrasound from all clinically assessed women were assayed for human epididymis 4 and CA125. Area under the curve and sensitivity for diagnosing ovarian cancer of multiple penalized logistic regression models incorporating logCA125, log human epididymis 4, age, and simple ultrasound features of the adnexal mass were compared.

RESULTS: Of 1590 (158 multimodal, 1432 ultrasound) women with adnexal masses, 78 were diagnosed with ovarian cancer (48 invasive epithelial ovarian, 14 type I, 34 type II; 24 borderline epithelial; 6 nonepithelial) within 1 year of scan. The area under the curve (0.893 vs 0.896; P = .453) and sensitivity (74.4% vs 75.6% ;P = .564) at fixed specificity of 90% of the model incorporating age, ultrasound, and CA125 were similar to that also including human epididymis 4. Both models had high sensitivity for invasive epithelial ovarian (89.6%) and type II (>91%) cancers.

CONCLUSION: Our population cohort study suggests that human epididymis 4 adds little value to concurrent use of CA125 and transvaginal ultrasound in the differential diagnosis of adnexal masses in postmenopausal women.

Original languageEnglish
Pages (from-to)56.e1-56.e17
Number of pages17
JournalAmerican journal of obstetrics and gynecology
Issue number1
Early online date24 Jul 2019
Publication statusPublished - Jan 2020


  • ovarian cancer
  • ovarian neoplasm
  • CA125
  • HE4
  • TVS
  • ultrasound
  • risk of malignancy
  • diagnosis
  • adnexal mass
  • human epididymis 4
  • United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS)
  • transvaginal ultrasound
  • Humans
  • Middle Aged
  • Carcinoma, Ovarian Epithelial/diagnosis
  • Sensitivity and Specificity
  • Ultrasonography
  • Postmenopause
  • Female
  • Diagnosis, Differential
  • Logistic Models
  • WAP Four-Disulfide Core Domain Protein 2/metabolism
  • Ovarian Neoplasms/diagnosis
  • Randomized Controlled Trials as Topic
  • Membrane Proteins/metabolism
  • Aged
  • CA-125 Antigen/metabolism
  • Cohort Studies

ASJC Scopus subject areas

  • Obstetrics and Gynaecology


Dive into the research topics of 'Serum HE4 and diagnosis of ovarian cancer in postmenopausal women with adnexal masses'. Together they form a unique fingerprint.

Cite this