Intergrated Systems Biology Approach for Ovarian Cancer Biomarker Discovery

Abbreviated Name: Diamandis-Discovery of Ovarian Cancer Biomarkers

Protocol ID: 405

Lead Investigator: Diamandis, Eleftherios P. — Mount Sinai Hospital

Coordinating Investigator:

No coordinating investigator defined.

Involved Investigator Sites:

Diamandis, Eleftherios P. — Mount Sinai Hospital

Biomarker Name: AGRN, ANXA2, BCAM, DKK3, DSC2, DSG2, ECM1, FBLN1, FOLR1, FSTL1, GRN, KLK6, PROS1, PK1, VDBP.

Design: No design specified.

Field(s) of Research: Proteomics

Collaborative Group: Breast and Gynecologic Cancers Research Group

Phased Status:

Abstract:

The overall objective is to validate serum protein markers for early diagnosis of ovarian cancer with the ultimate goal being to develop a multiparametric panel consisting of 2-4 novel markers with 10 known markers for phase 3 analysis. In phase 1, we will screen for markers able to pass a threshold of 98% specificity and 30% sensitivity in a cohort of 300 women. Markers that pass phase 1 validation will be investigated in a phase 2 PRoBE cohort with a 98% specificity and 70% sensitivity cut-off. Finally, markers that pass phase 2 validation will be evaluated in EDRN CVC laboratory specimens with a cut-off of > 98% specificity and 90% sensitivity.

Aims: The major objective of this study is to examine if a panel of newly discovered ovarian cancer biomarkers, in combination with the classical ovarian cancer biomarkers, CA125 and other promising biomarkers such as HE4, constitutes a new, multiparametric serum panel for early ovarian cancer diagnosis with high sensitivity and specificity. We have developed a three-phase validation strategy in order to accomplish this objective. Thus far, 15 markers have been examined in the phase 1 cohort with five markers showing significant elevations in the serum of ovarian cancer patients compared to healthy controls. These 5 candidates were further evaluated in order of: (1) a retrospective cohort consisting of healthy controls (n=100), patients with benign gynecological conditions (n=100), and patients with ovarian cancer (n=100); (2) a retrospective cohort of patients who presented with a pelvic mass and later diagnosed with either a benign gynecological condition (n=50) or ovarian cancer (n=50); and finally (3) the phase 2 PRoBE cohort of patients who presented with a pelvic mass and were followed up to their diagnoses of either a benign gynecological condition (n=80) or ovarian cancer (n=130). From this extensive validation scheme, BCAM, FOLR1 and KLK6 were able to retain their discriminatory power after final evaluation in the phase 2 PRoBE cohort. CA125 and HE4 were also evaluated in order to assess how well our in-house markers perform against the FDA-approved markers as well as to assess if there exists any complementarity between the markers. Receiver operating characteristic curve analysis revealed that HE4 and FOLR1 performed better than CA125, BCAM and KLK6 in identifying ovarian cancer patients from patients with benign gynecological conditions. Further analysis revealed that the combination of HE4 and FOLR1 resulted in higher sensitivities than either marker alone and this was especially apparent at stringent specificities. To date, we have identified FOLR1 as a promising marker worthy of evaluation in the EDRN CVC laboratory specimens due to its improved performance compared to CA125 and possible complementarity with HE4. BCAM and KLK6 may also be worth of investigation as second-tier markers.

Analytic Method: ELISA

Comments:

Biomarkers:

Datasets:

No datasets are currently associated with this protocol.