Clinical Qualification of Imaging and Fluid-Based Tumor Monitoring Biomarkers for Metastatic Castration Resistant Prostate Cancer

Metastatic prostate cancer is the second largest cause of cancer-related death in American men. A major obstacle to their treatment and to developing new drugs for this disease is the lack of reliable quantitative treatment monitoring methods. Fluid-based tumor monitoring using circulating tumor cells (CTC) has recently proven valuable for prognosis and predicting response to treatment. Current imaging methods for treatment monitoring in prostate cancer involve lesion-counting of new bone metastases or selected index lesions of soft tissue disease; both focus on progression only and cannot assess the full disease burden. This proposal deploys two validated methods, automated bone scan index (aBSI) and lymph node segmentation, to quantify standard imaging results so that a measure of total disease burden on imaging can be integrated with CTC data in a predictive model. The hypothesis that this model will outperform current methods for monitoring response to therapy will be tested via these specific aims: 1) Correlate quantitative post-treatment changes in tumor burden as assessed by imaging and CTC’s; 2) Determine whether a combination of fluid-based tumor monitoring (liquid biopsy) and an imaging assay is more powerful than either assay alone in prognosticating for survival; 3) Identify the optimal combination of liquid biopsy and imaging for different therapies. Models will be built to predict overall survival using imaging, CTC, and prognostic factor data from phase III clinical trials of first-line treatments, docetaxel, docetaxel + radium-223, and abiraterone + prednisone, for an advanced stage of disease, metastatic castration- resistant prostate cancer (mCRPC). Clinical trials including this data are available for analysis due to the recommendations for serial imaging (CT and bone scintigraphy) and CTC assessment as endpoints in the Prostate Cancer Working Group (PCWG) 3 guidelines, co-authored by the study team. The results of this study will create the first reliable, multiparametric, and fully quantitative biomarker that captures clinically meaningful indices of both response and progression in mCRPC; generate evidence to support its inclusion in future PCWG guidelines; and develop methods applicable to new imaging modalities.