Endogenous Chemical Exchange Saturation Transfer: Quantitative Modelling and Application in Cancer

This work describes the origins of the contrast mechanism chemical exchange saturation transfer (CEST) and methods for proton magnetic resonance imaging (MRI), with application in cancer. It focuses on endogenous techniques, which do not require the injection of a contrast agent, and the experimental and analytic techniques which allow quantitative metrics to be obtained. The theory behind CEST is presented, and it is modelled using a three pool system of Bloch equations which are a function of magnetization relaxation parameters, CEST pool concentration, and hydrogen exchange rate constant. A platform is created for optimization and parameter fitting, which is used to obtain parameter values and for pulse sequence experimental design. The contribution of semi-solid macromolecules to the CEST spectrum is determined, and evidence given that the CEST parameters (in particular, the exchange rate constant) can be obtained independently of this additional pool. It is shown that there is a linear relationship between the log of the exchange rate constant and pH. CEST spectra are obtained in cancer xenografts in mice, exhibiting features from amide, amine and aliphatic protons as well as from magnetization transfer. Semi-quantitative CEST parameter maps are derived, showing the distribution of CEST features in tumours and in contrast with the surrounding normal-appearing muscle. Several exploratory experiments are performed in protein-containing phantoms and cell pellets in order to validate the origins of CEST spectra and their behaviour in conditions of changing pH and temperature. Cell pellet experiments support the hypothesis that CEST is sensitive primarily to intracellular conditions, and furthermore that the cell nucleus is a concentrated source of CEST-contributing proteins. These validation experiments are supportive of the assumptions made in the next section, in which a ratiometric method of determining intracellular pH from CEST spectra is proposed and proof-of-concept experiments are performed. Ultimately, the work contained herein supports the hypothesis that the endogenous CEST experiments provide new information not obtainable from other MRI experiments, leading to quantitative absolute intracellular pH mapping. This has potential for predicting the effectiveness of treatment regimens based upon the relationship between intracellular pH and drug uptake, establishing regions of tumour which are actively proliferating or which may be resistant to therapy.Ph.D