Meeting Banner
Abstract #1926

On the Accuracy of AUC Ratio Method for Detecting Treatment Changes with Hyperpolarised 13C Dynamic Spectra

Deborah K. Hill1, Matthew R. Orton1, Martin O. Leach1, Yuen-Li Chung1, Thomas R. Eykyn1, 2

1CR-UK and EPSRC Cancer Imaging Centre, The Institute of Cancer Research and Royal Marsden NHS Trust, Sutton, Surrey, United Kingdom; 2Division of Imaging Sciences and Biomedical Engineering, Kings College London, St Thomas Hospital, London, United Kingdom

Signal enhancement from Dynamic Nuclear Polarisation allows estimation of apparent enzymatic rates in vivo in real-time. We have previously described a method for calculating metrics that are sensitive to treatment induced changes in the apparent forwards rate constant, based on AUC ratios of injected and generated metabolites (e.g. pyruvate and lactate). This AUC ratio approach is simpler and more robust than curve fitting using the modified Bloch equations, but suffers from bias. This theoretical abstract explores the assumptions required for the AUC ratio approach to be valid, and shows that the bias is less than 10% in a representative example.

Keywords

abstract according advantageous alternative apparent appropriate assume assuming assumption bias biomedical cancer complex computed conditions considered constant constants conversion correctly correlation correspond curves decay define depends depicted derived described detecting detection determine differential difficult dimensionless dots drug dynamic engineering england enhancements enzymatic equations equivalently error errors evaluate example exchange experimental finding fitting form formed forward fractional full function generate generated give grants greater health hill importantly in vivo included incurred independent independently individual induced injected input institute integrals investigate kinetic kingdom knowledge known lactate leach linear literature manipulated measured measuring metabolite metabolites metric model modified molecule noise novel nuclear particular particularly pathways peak plot post practice prediction previously priori proportional published quantify random react reaction readily relate relates relevant repetitions requires respectively reverse robustness royal sciences simple simpler simulated simulation since spectra structure suitable summed summing superscripts surrey system taken target term terms treatment treatments true trust useful usually variables various vitro worse