| Summary: | Objective: Proteomic profiling of plasma or serum is a technique to identify new biomarkers in disease. The objective of this study was to identify new plasma biomarkers in ovarian cancer patients using mass spectrometry protein profiling and artificial intelligence.
Methods: A total of 65 plasma samples obtained from women with ovarian cancer (n = 35) and age-matched disease-free controls (n = 30) were applied to anion exchange protein chips for protein profiling by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS).
Results: SELDI-TOF MS was highly reproducible in detecting ovarian tumor-specific protein profiles. One protein peak (relative molecular mass, Mr, 11,537 Da) was identified in plasma from women with ovarian cancer but not in controls. Two peaks, Mr 5,147 and 8,780 Da, were present in the plasma of controls but not of women with ovarian cancer. After a training analysis, classification analysis generated by univariant or linear combination split was performed to reach a discriminant protein signature pattern. After cross validation, a sensitivity of 84% and specificity of 89% for all studied cases and controls was reached.
Conclusion: This study clearly demonstrates that the combined technology of SELDI-TOF MS and artificial intelligence is effective in distinguishing protein expression between normal and ovarian cancer plasma. The identified protein peaks may be candidate proteins for early detection of ovarian cancer or evaluation of therapeutic response.
|
|---|
