Abstract PO-060: Blood based metabolomic biomarker assay to predict radiation late effects in prostate cancer patients

Abstract Introduction and Objective: Radiation therapy (RT) is an effective modality as a primary treatment of cancers, or as an adjuvant to surgery or chemotherapy. However, about 10 % of treated patients experience radiation treatment related late effects that adversely affect quality of life. Availability of a blood test that can be used to predict or monitor adverse symptoms would help stratify sensitive sub-populations of patients receiving radiation therapy. We have developed a kit based assay for the measurement of a small molecule based biomarker panel that is predictive of radiation treatment related toxicities in prostate cancer patients. Methods: We performed multiple reaction monitoring based targeted metabolomic/lipidomic analyses using baseline (samples obtained before radiation therapy) from the plasma of a cohort of 100 prostate cancer patients. Pre-processed data along with clinical annotations were used for the analyses. An optimized classifier algorithm was developed, using baseline plasma samples collected prior to the initiation of radiation therapy. We compared groups of sub-sets of patients who developed adverse symptoms to those who did not (control group), over the course of a two-year follow-up interval. Feature selection for building classifiers was performed using LASSO Logistic regression to avoid over-fitting of the model through iterative selection, and cross validation of candidate markers, in compliance with FDA biomarker development guidelines. Results: We have used a combination of metabolomics and clinical data for developing a kit based assay for identification of prostate cancer patients Metabolite signatures predictive of adverse responses to radiation therapy were developed in the patient cohort treated with stereotactic body radiation therapy (SBRT) for prostate cancer. We were able to develop high accuracy, predictive algorithms for recurrence, urinary symptoms and rectal proctitis episodes in this cohort. A plasma metabolite index was developed for the evaluation of susceptibility to adverse outcomes of radiation therapy. Furthermore, a visualization software was developed to integrate clinical and biomarker data for easier clinical translation. Conclusions: We developed candidate predictive biomarker panels that can be used for identifying patients who are at increased risk for developing adverse symptoms following radiation therapy. Following analytical validation, this kit based assay is a candidate for a clinical validation study to determine clinical utility. Citation Format: Amrita K. Cheema, Scott Grindrod, Yaoxiang Li, Shivani Bansal, Simeng Suy, Sean Collins, Anatoly Dritschilo. Blood based metabolomic biomarker assay to predict radiation late effects in prostate cancer patients [abstract]. In: Proceedings of the AACR Virtual Special Conference on Radiation Science and Medicine; 2021 Mar 2-3. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(8_Suppl):Abstract nr PO-060.