Article Text

Download PDFPDF

#675 A new score based on human epididymis protein 4 discriminates benign from malignant adnexal mass much better than risk of ovarian malignancy algorithm
Free
  1. Mikhail Katsyuba1,2,
  2. Rustem Khasanov1,
  3. Irina Kutsenko3,
  4. Maria Egunova3,
  5. Gulsiyar Muratova4,
  6. Timur Madzhidov4,
  7. Guzel Usmanova2,
  8. Aisylu Akhmetzianova2 and
  9. Vera Terentieva2,1
  1. 1Russian Medical Academy of Continuous Professional Education - Kazan State Medical Academy Branch Campus, Kazan, Russia
  2. 2Republican Clinical ?ancer Center named after prof. M.Z.Sigal, Kazan, Russia
  3. 3Siberian State Medical University, Tomsk, Russia
  4. 4Kazan (Volga region) Federal University, Kazan, Russia

Abstract

Introduction/Background Recently, ESGO/ISUOG/IOTA/ESGE Consensus Statement on pre-operative diagnosis of ovarian tumors implied that neither Human epididymis protein 4 (HE4) nor Risk of Ovarian Malignancy Algorithm (ROMA) improve the discrimination between benign and malignant masses compared with CA 125 alone. This statement might be reassessed if a novel algorithm, more effective than ROMA, will be implemented. Thereby the aim of this study was to validate a new predictive algorithm, based on serum CA125&HE4.

Methodology A novel Risk of Ovarian Cancer Kazan Index (ROCK-I), based on serum HE4, CA125 and patient’s age as variables, has been developed using a training dataset (n=284). ROCK-I provides an estimation of the risk of malignancy of adnexal mass in premenopausal patients. The validating dataset consisted of 333 consecutively operated premenopausal patients with pelvic mass out of which there were 281 cases of benign diseases, 43 cancers and 9 borderline ovarian tumors (BOT). Results on the validating dataset are reported below.

Results When benign diseases vs all cancers and BOT were considered, ROC-AUC of ROCK-I, ROMA and CA 125 in the validating dataset were 0.917, 0.864 and 0.874 respectively. When benign diseases vs all cancers and stages Ic2-III of BOT were considered, ROC-AUC were 0.96, 0.911 and 0.896 respectively. The superiority of ROCK-I was statistically significant over both ROMA (p=0.003) and CA 125 (p=0.002). When standard cut-off levels were applied the specificities of ROCK-I and ROMA were 93.5 and 85.1%, and the sensitivities for all cancers were 93 and 86% respectively. The performance of ROCK-I and ROMA in different scenarios of discrimination is shown in table 1.

Abstract #675 Table 1

Comparison of the performance of ROCK-I and ROMA in the validating dataset.

Conclusion ROMA provides suboptimal discrimination at least among premenopausal patients. If a large independent validation shows similar or even slightly lower superiority of the novel ROCK-I over ROMA, it may provide a new basis of routine-use of HE4 in premenopausal patients with pelvic mass.

Disclosures Authors has nothing to disclosure

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.