Breast cancer is the most common form of cancer in women. In the year 2004, it was estimated that approximately 217,000 new cases of breast cancer were diagnosed in the United States, and that 40,000 deaths were attributed to the disease. The key to successful treatment of breast cancer is early diagnosis, and the use of widespread mammography screening has resulted in significant improvements in breast cancer survival rates. However, a major problem with mammography is a lack of specificity;in some studies, as many of 70-80% of suspicious lesions on mammography referred for biopsy ultimately have a benign final diagnosis. These 'unnecessary"""""""" biopsies represent a significant economic burden on health care systems, and are also invasive and unpleasant for the patient. Therefore, there is a need for the development of new non-invasive, cost-effective, and safe diagnostic imaging procedures with enhanced specificity and sensitivity. Proton MR spectroscopic imaging (MRSI) is a non-invasive metabolic imaging technique that shows promise for the non-invasive diagnosis of human breast cancer. Preliminary data from our group and others suggests that an elevated choline signal (detected in the proton MR spectrum) is a marker of malignant breast disease. However, the low signal-to-noise ratio of proton MRSI currently limits this methodology to quite large lesions (e.g. 1 cm or greater), and makes detection of small Cho signals difficult. Also, few previous studies have investigated the spatial distribution of Cho in breast cancer lesions. In this proposal, we therefore propose to develop methods for MRSI on high field MR systems (3 and 7 Tesla) that are expected to exhibit higher sensitivity and resolution than lower field scanners. Methods will be developed for full breast coverage at high magnetic fields in short scan times, using phased-array receiver coils, and optimal water and lipid suppression. Methods will also be developed for the quantitative determination of lesion choline concentrations. These methods will be developed in years 1 and 2, and, in years 3 through 5, the clinical value of MRSI will be investigated. Specifically, choline levels will be compared between histologically defined tissue types (malignant and benign), in patients who are scheduled for breast biopsy. The sensitivity and specificity of proton MRSI in this patient group will be determined, and compared between field strengths. In addition, the diagnostic value of MRSI will be compared to that of conventional MRI in the same population.
