Background Due to lack of effective early diagnostic methods, ovarian cancer (OC) remains a disease with a very unfavorable prognosis and correct differentiation between benign and malignant ovarian tumors is often difficult. Metabolomic profiling has recently been widely used in the search for non-invasive cancer diagnostic methods. In this research the serum free amino acid profiles were investigated to identify potential novel biomarkers of OC and assess their performance in ovarian tumor differential diagnosis.
Results Serum samples from patients diagnosed with ovarian tumors were divided based on the histopathological result: epithelial OC (n=38), borderline ovarian tumors (n=6) and benign ovarian tumors (BOTs) (n=62). Serum free amino acid profiles were evaluated using aTRAQ methodology based on high-performance liquid chromatography electrospray ionization tandem mass spectrometry. In the performed analyses, the levels of eleven amino acids significantly differed between OC + borderline and BOTs. The highest AUC (0.787) was obtained for histidine. Cystine and histidine were identified as best single markers for early stage OC/BOT and type I OC. For advanced stage OC, seven amino acids differed significantly between the groups and citrulline obtained the best AUC of 0.807. Between type II OC and BOTs eight amino acids differed significantly and the highest AUC of 0.798 was achieved by histidine and citrulline (AUC of 0.778).
Conclusions Serum free amino acid profiles are significantly altered in OC patients as compared to those with BOTs. Histidine emerges as a potential new biomarker in differential diagnosis of ovarian tumors. Adding histidine to a multimarker panel together with CA125 and HE4 improved the differential diagnosis between OC and BOTs.