Abstract 2818: Urinary MMP-2 and MMP-9 predict the presence of ovarian cancer in women with normal CA125 levels

Abstract

Abstract Currently, the only widely used biomarker for monitoring ovarian cancer status and response to therapy is the cancer antigen 125 (CA125). Interestingly, the majority of newly developed ovarian cancer multi-marker assays also rely heavily on serum CA125 levels as one of the biomarkers in their panels. However, 20% of women with advanced ovarian cancer have blood CA125 levels within the normal range of 35 U/mL and below. These women represent a cohort of patients for whom no clinically useful biomarkers are available for monitoring disease status, stage or response to therapy. To address this need, we asked whether a panel of urinary biomarkers that we have previously validated for use in other cancers, could predict the presence of disease in women with ovarian cancer possessing normal CA125 levels. Urine samples were collected from 81 healthy controls and 26 ovarian cancer patients with blood CA125 levels below 35 U/mL. Using monospecific ELISAs, we determined that urinary levels of matrix metalloproteinases 2 and 9 (MMP-2 and MMP-9) and the neutrophil gelatinase lipocalin 2 (NGAL) are detected at significantly higher levels in the urine of ovarian cancer patients than in the urine of healthy controls. Additionally, receiver-operating characteristic (ROC) area under the curve (AUC) analysis of these groups revealed that, individually, MMP-2, MMP-9 and NGAL each could significantly distinguish ovarian cancer patients with CA125 < 35 U/mL from controls. The best combination of biomarkers for predicting the presence of disease was MMP-2 plus MMP-9 with an ROC AUC of 0.7332 (P<0.01). When these biomarkers were multiplexed with age, a know risk factor of ovarian cancer, the diagnostic accuracy of these biomarkers increased to an AUC of 0.820 (P<0.001). This combination of urinary MMP-2, MMP-9 and age demonstrated excellent diagnostic capability in differentiating ovarian cancer patients from healthy controls among the cohort of patients for whom CA125 values were uninformative. Acknowledgments: The authors gratefully acknowledge the support of Ovations for the Cure of Ovarian Cancer (MAM and UAM) and the Stuart Weitzman Foundation (MAM). Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2818. doi:10.1158/1538-7445.AM2011-2818