Early effects of neoadjuvant chemotherapy in breast cancer using metabolic MRI

Abstract

Breast cancer patients undergoing NAC treatment experience side effects, due to the fact that chemotherapy drugs target all cells in the body that grow and divide rapidly. On top of that, in ~30% of the patients, NAC treatment does not result in tumor size reduction. At the moment there is no non-invasive method to evaluate the early effects of systemic treatment. The overall aim of this thesis was to evaluate early effects of NAC treatment in breast cancer patients using metabolic MRI. The first part of this thesis shows that we are able to measure metabolic changes after the first cycle of NAC treatment. In Chapter 2, metabolic ratios measured with 31P-MRSI were correlated to the pathological response. The PMEs and PDEs are known to be important in anabolism and catabolism of the cell membrane. In a small group of patients, it appeared that the PME/PDE ratio in partial and complete responders decreased after the first cycle of NAC treatment and the non-responders showed an increase in the PME/PDE ratio. This shows the potential of using PME/PDE ratio as a biomarker for early prediction of non-response to NAC treatment in breast cancer patients. In Chapter 3, a different contrast mechanism was used to acquire metabolic information about the tumor. Changes in amide signals originating from amide protons attached to the backbone of proteins and peptides, were measured with CEST-MRI. These are of interest as tumors show increased amide signals compared to healthy tissue due to increased cellular proliferation and subsequent accumulation of defective proteins in tumors. Partial and complete responders showed a decrease in amide signal while non-responders showed an increase in amide signal. In Chapter 4, we correlated the amide signal measured with CEST-MRI to the pH measured with 31P-MRSI. A correlation between amide signal and pH in the tumor was found in breast cancer patients. This correlation was opposite from the intrinsic relation between amide signal and pH, as this is base catalyzed. This demonstrates that the contribution of the concentration of mobile amide protons likely supersedes the influence of the exchange rate in the measured amide signal. This also shows that 31P-MRSI and CEST-MRI provide complementary information about the tumor emphasizing the importance of both techniques. In the last part of this thesis in Chapter 5, first steps were taken to merge metabolic imaging with ultra-high resolution MRI of the breasts. A new bilateral breast coil setup was developed. Improved image quality for breast imaging at 7 T was achieved using five meandering dipole antennas in a multi-transmit setup, combined with a high density receive array. The coil generates sufficient B1+ over a larger FOV in the breasts. This makes it possible to acquire T2-weighted images and to image the axillary lymph nodes, which has not been shown in previous studies at high field. The possibility of T2-weighted imaging is a crucial step towards translating routinely used breast imaging protocols from 3 T to 7 T.