Evaluation of Neoadjuvant Chemotherapy in Locally Advanced Breast Cancer Based on MR Methodology

Abstract

Breast cancer is the most common malignancy in women. Within clinical cancer treatment there is a trend to move away from population-based treatment to more personalized treatment regimens. This shift requires identification and validation of factors that can help to evaluate the treatment response and to assess prognosis. Magnetic resonance (MR) has evolved as one of the most important imaging modalities within oncology. Today, MR of the breast is commonly applied for breast cancer detection, diagnosis and follow-up.

Several MR sequences can be applied within an examination time of 45 minutes. Together these provide detailed anatomical images as well as functional information regarding tumor perfusion, water diffusion and metabolism. Dynamic contrastenhanced MR imaging (DCE-MRI) provides information about tumor size, vasculature, permeability and blood supply, and functional changes related to this can be assessed. Aggressive tumors are often highly vascularised, and the formation of new blood vessels is poorly regulated resulting in a chaotic organization of vessels with incomplete vessel walls. Diffusion weighted MRI (DW-MRI) explores the motion of water. In tissue with high cellularity the water diffusion is restricted, and DW-MRI can thus be used to assess changes in cell density. MR spectroscopy (MRS) mirrors the tumor metabolism, and can be performed both in vivo and ex vivo (in tissue biopsies for instance). In cancerous breast tissue elevated levels of choline containing compounds have been observed. Total choline (tCho) has been suggested as a biomarker for response to NAC treatment. However, tCho can not be detected in all tumors.

Breast cancer is a heterogeneous disease, and the outcome may vary according to this. Patients diagnosed with locally advanced breast cancer (LABC) have a poor prognosis. These patients often receive neoadjuvant chemotherapy (NAC) prior to operation to downstage the disease. The aim of this work was to use MR methodology to evaluate NAC in LABC patients. The patients were examined at clinical 1.5 T or 3 T scanners, with protocols including DCE-MRI, DW-MRI, in vivo and ex vivo MRS.

Our results show that pre-treatment DCE-MRI can be used to evaluate the prognosis in breast cancer patients receiving NAC. Assessment of response to NAC can be monitored before the second cycle of NAC on both 1.5 T and 3 T, and tumor volume parameters were the strongest predictors of response. 3 T show more significant findings in DCE-MRI derived parameters that were related to response compared to the results from 1.5 T. Several MRI derived as well as clinical parameters were found to be significantly different in tumors with detectable tCho compared to tumors with negative tCho detection.

In conclusion, the work presented in this thesis presents promising tools to assess prognosis and response to treatment in breast cancer patients.