A new approach for identifying patients with ovarian epithelial neoplasms based on high-resolution mass spectrometry.

Abstract

We investigated the serum profiles of patients with ovarian neoplasm using high-performance liquid chromatography (HPLC) and high-resolution mass spectrometry (HRMS) to obtain serum "fingerprints" for use in identifying patients with these neoplasms. We used HPLC-HRMS to analyze serum samples from patients with ovarian neoplasms and control subjects. Serum samples from 145 patients were analyzed, including 85 with ovarian epithelial neoplasms. We also compared the results of this serum-fingerprinting approach with the results of the CA-125 test and imaging. Fingerprinting successfully permitted the separation of control patients and patients with ovarian neoplasms. The sensitivity and specificity of the test were between 96% and 100%. When the results of this test were concordant with the results of the CA-125 test, 99% of serum samples were correctly classified as being from a patient with an ovarian neoplasm or with no ovarian neoplasm. We found that a metabolite of molecular weight 472 is the main metabolite in the separation of patients with ovarian neoplasms from control subjects. HPLC-HRMS serum profiling could become a screening test for ovarian neoplasms.