Comparison of proteomic biomarker panels in urine and serum for ovarian cancer diagnosis

TY - JOUR

T1 - Comparison of proteomic biomarker panels in urine and serum for ovarian cancer diagnosis

AU - Petri, Anette Lykke

AU - Simonsen, Anja Hviid

AU - Høgdall, Estrid

AU - Christensen, Ib Jarle

AU - Kjaer, Susanne Krüger

AU - Yip, Christine

AU - Risum, Signe Normann

AU - Pedersen, Anette Tønnes

AU - Hartwell, Dorte

AU - Fung, Eric T

AU - Høgdall, Claus

N1 - Copyright © 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

PY - 2010/3/1

Y1 - 2010/3/1

N2 - Purpose: The purposes of this study were to confirm previously found candidate epithelial ovarian cancer biomarkers in urine and to compare a paired serum biomarker panel and a urine biomarker panel from the same study cohort with regard to the receiver operating characteristic curve (ROC) area under the ROC curve (AUC) values. Experimental design: Four significant urine biomarkers were confirmed among 130 pelvic mass patients in the present study. The four biomarkers form a potential urine biomarker panel. From the same study cohort, the potential urine biomarker panel was compared to a serum biomarker panel, consisting of seven proteins/peptides, OvaRI. Results: Multivariate analysis of the urine panel demonstrated a significant differentiation (po0.0001) between epithelial ovarian cancer patients and patients with benign ovarian pelvic masses.The ROC AUC of the urine panel was 0.84 and the ROC AUC of OvaRI was 0.83. Combining the urine panel with OvaRI demonstrated a significant contribution from both, for urine peaks, OR = 2.12 and for OvaRI, OR = 1.39; the ROC AUC of this model was 0.88. Conclusions and clinical relevance: We demonstrated that both urine and serum can be used individually or in combination to potentially aid in ovarian cancer diagnostics. Urine proteomic profiling could provide biomarkers for the non-invasive test required in clinical practice.

AB - Purpose: The purposes of this study were to confirm previously found candidate epithelial ovarian cancer biomarkers in urine and to compare a paired serum biomarker panel and a urine biomarker panel from the same study cohort with regard to the receiver operating characteristic curve (ROC) area under the ROC curve (AUC) values. Experimental design: Four significant urine biomarkers were confirmed among 130 pelvic mass patients in the present study. The four biomarkers form a potential urine biomarker panel. From the same study cohort, the potential urine biomarker panel was compared to a serum biomarker panel, consisting of seven proteins/peptides, OvaRI. Results: Multivariate analysis of the urine panel demonstrated a significant differentiation (po0.0001) between epithelial ovarian cancer patients and patients with benign ovarian pelvic masses.The ROC AUC of the urine panel was 0.84 and the ROC AUC of OvaRI was 0.83. Combining the urine panel with OvaRI demonstrated a significant contribution from both, for urine peaks, OR = 2.12 and for OvaRI, OR = 1.39; the ROC AUC of this model was 0.88. Conclusions and clinical relevance: We demonstrated that both urine and serum can be used individually or in combination to potentially aid in ovarian cancer diagnostics. Urine proteomic profiling could provide biomarkers for the non-invasive test required in clinical practice.

U2 - 10.1002/prca.200900042

DO - 10.1002/prca.200900042

M3 - Journal article

SN - 1862-8346

VL - 4

SP - 304

EP - 314

JO - Proteomics - Clinical Applications

JF - Proteomics - Clinical Applications

IS - 3

ER -