Arun Sreekumar

Ph.D

Baylor College of medicine

Professor

PLCO (Learn more about this study)

PLCO-583

Feb 19, 2020

Developing a metabolite-based biomarker panel for prostate cancer detection and prognosis in African-American men

In some series, African American (AA) men have 60% higher incidence and two-times greater risk of dying of prostate cancer (PCa) than European American (EA) men. AA men have early onset of the disease with rapid progression to metastasis. With the limitation of Prostate Specific Antigen (PSA) alone as reliable marker for PCa detection, there is an urgent need to develop accurate bio-markers for early detection of this disease especially in AA men. Metabolites can be detected in non-invasive bio fluids. A metabolite-based bio-marker panel for cancer is currently lacking. Notably, there is a lack of predictive markers that can predict the likelihood of biochemical recurrence and/or castration resistant diseases in AA men with prostate cancer. From the patient perspective such a test needs to be non-invasive while being able to be integrated into the clinical setting. Thus, the long term goal of this proposal is to develop panel of metabolites for early detection and/or as predictive markers for prostate cancer progression.
This project will develop a non-invasive multiplex biomarker panel for the early detection and prognosis of prostate cancer in AA men. In the initial phase, the focus will be to discover the panel of metabolites in prostate tumors versus patient matched adjacent benign tissue in AA men. The absolute levels of the verified metabolites in sera will be validated in a newly developed metabolite quantification platform in our laboratory to associate the variour metabolite level with detection or prognosis of the disease for AA men. Prognosis for prostate cancer will be defined as two end points 1) time to biochemical recurrence (clinically measured by raise in PSA) and development of castrate resistant disease associated with local or distal metastasis. The analysis will use a training approach where in the clinical samples will be categorized a priori into subgroups (cancer versus benign or indolent versus aggressive) and association of metabolite or panels of metabolites with the clinical endpoint(s) will be determined.

In all cases, absolute levels of metabolites will be quantified using Multiple Reaction Monitoring. For each metabolite, the Limit of Detection, Limit of Quantification, matrix interference, biological and technical reproducibility, inter/intraday assay variability, accuracy and precision will be calculated, both for individual metabolites and multiplex panels. Calibration plots will be generated using spiked isotopically labeled internal standards. Pooled samples will be used as positive (pooled cancer samples) and negative controls (pooled healthy controls from individuals having no prior history of cancer). Following the development of the Training Model (that contains a list of key metabolites that are associated with the outcome), we will validate the nominated panel of metabolites in an independent set of clinical sera.

Hypothesis1: Metabolite levels are associated with prostate cancer detection and prognosis in AA men. We will discover a panels of metabolites in prostate tumors versus patient matched adjacent benign tissue for AA men using PLCO dataset. A list of metabolites will be nominated for the liklihood of early dectection (biopsy status), biochemical recurrence and castration disease in AA men.

Hypothesis 2: There is a sera-based metabolite marker for early detection of PCa in AA men. The absolute levels of the nominated metabolites will be quantified. A panel of metabolites will be correlate with the biopsy outtome.

Hypothesis 3: There is a sera-based metabolite marker asocited with prognosis for PCa in AA men. The absolute levels of the nominated metabolites will be quantified. A panel of metabolites will be correlate with the onset of biochemial recurrence and/or castration resistance/metabstasis.