Article Text

Download PDFPDF
Identifying Serum Biomarkers for Ovarian Cancer by Screening With Surface-Enhanced Laser Desorption/Ionization Mass Spectrometry and the Artificial Neural Network
  1. Jing Yang, MD*,
  2. Yanhui Zhu, PhD,
  3. Hongyan Guo, MD,
  4. Xiuyun Wang, MD,
  5. Ronglian Gao, MD,
  6. Lufang Zhang, MD,
  7. Yangyu Zhao, MD and
  8. Xiaowei Zhang, MD
  1. *Department of Obstetrics and Gynecology, Haidian Women’s and Children’s Hospital, Beijing 100080, China;
  2. Medical Informatics Center, Peking University, Beijing, China; and
  3. Department of Obstetrics and Gynecology, Third Hospital, Peking University, Beijing, China.
  1. Address correspondence and reprint requests to Xiaowei Zhang, MD, Third hospital, Peking University, 49 Garden N Rd, Beijing 100191, China. E-mail:


Objective The purpose of this study was to screen potential serum tumor biomarkers for the diagnosis of ovarian cancer.

Methods The study includes 3 sets. The first set of patients included 37 ovarian cancers and 31 healthy women (healthy controls). The second set included 42 ovarian cancers, 33 patients with benign ovarian tumor, and 29 healthy women (noncancer controls). The third set included 39 ovarian cancers and 35 patients with benign ovarian tumor (benign controls). Serum samples from ovarian cancers, healthy controls, noncancer controls, and benign controls were analyzed by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry.

Results A 3-peak model (peaks of mass-to-charge ratio values at 5766.379 d, 5912.586 d, and 11695.56 d) was established in the training set that discriminated cancer from noncancer with high sensitivity (10/11, 90.90%) and specificity (19/20, 95.00%). The peaks corresponding to 3 potential biomarkers increased significantly with the degree of malignancy.

Conclusions The proteins represented by these 3 peaks are biomarker candidates for ovarian cancer diagnosis and/or monitoring treatment response.

  • Ovarian cancer
  • Artificial neural network
  • Proteomic model

Statistics from

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.


  • This research was supported by National Natural Science Foundation of China (39970763, 30471807).

  • The authors declare no conflicts of interest.

  • Jing Yang and Yanhui Zhu contributed equally to this project and should be considered co–first authors.