Because MRI can provide increased brain and liver lesion detection as compared with CT, the investigators hypothesize that FDG-PET/MRI will provide concordant or improved lesion detection as compared with FDG-PET/CT in breast cancer patients at a decreased radiation dose.

This study is an initial assessment of the ability of 18FDG-PET/MR imaging to detect systemic disease in breast cancer patients as compared with conventional 18FDG-PET/CT.

Rationale: Accurate assessment of the extent systemic disease in breast cancer patients both initially and throughout treatment and surveillance comprises the basis for clinical care including surgical planning, radiation and drug selection. Shortcomings in FDG-PET/CT, a common current method of disease detection and surveillance, include decreased sensitivity in brain, liver and bone and a relatively high radiation dose, which could cause up to 5 new cancers per 1,000 patients (young American women) per scan. Initial studies assessing whole body MRI, a technique which involves no radiation, in the staging of breast cancer patients show promise, but can suffer from decreased ability to detect nodal disease and from detection of false positive lesions. Manual fusion of PET and MRI data in bowel and breast organ systems has been shown to decrease false positive lesions as compared with MRI alone. In the past year, molecular MRI imaging machines which combine PET and MRI imaging have become available for research purposes.

Hypothesis 1: We therefore hypothesize that FDG-PET/MRI will provide concordant or improved lesion detection as compared with FDG-PET/CT in breast cancer patients at a decreased radiation dose.

Hypothesis 2: In addition, the simultaneously acquired PET and MRI data allows for analysis of local tumor characteristics including perfusion, cellularity and glucose metabolism. In the subgroup of patients with an index breast cancer, we hypothesize that these local tumor metrics correlate with number and sites of systemic lesions.

Methods: All newly diagnosed breast cancer patients who are undergoing imaging evaluation for initial extent of disease or disease surveillance at NYU Cancer Center and Bellevue hospital are candidates for study in this protocol. Patients will be invited for follow-up FDG-PET/MRI imaging at any time they are undergoing FDG-PET/CT as part of their clinical care.

Subjects are undergoing FDG-PET/CT at the Cancer Center on 34th St and Lexington Avenue as part of routine care and will be transferred directly to the 38th St. facility by car in a timely manner so that an additional dose of 18-FDG labeled glucose will be unnecessary for the experimental FDG-PET/MRI.

FDG-PET/MRI will be performed and is estimated to 45-60 minutes of scanning time. The patient will then be free to leave the department. Data will be transferred to a workstation and interpreted by two independent radiologists. Size and location of each lesion will be recorded in a confidentiality protected spreadsheet for each component of the scan and the lesion load will be defined by the union of all detected lesions from all three components (DCE-MRI, diffusion-weighted imaging (DWI), FDG-PET). Data will be compared on a per lesion, per organ system and per patient basis with the patient's routine FDG-PET/CT for presence of metastatic disease and size and number of metastatic lesions.

In the case where PET/MRI demonstrates metastatic disease not seen on routine modalities, the referring physician will be alerted. Additional dedicated radiologic studies and biopsy may be recommended and performed at the discretion of the referring physician.

Results: Results of FDG-PET/MRI and FDG-PET/CT findings will be reported in a descriptive format together with related biopsy results and histopathologic correlation. An exact 95% confidence interval will be derived for the true percentage of times the assessment of disease state lesion, per organ system and per patient from FDG-PET/MRI will be concordant with the assessments derived for the same patient using the FDG-PET/CT. In patients with an index breast cancer, Pearson and Spearman rank correlations will be used to characterize the association between local tumor metrics and metastatic burden and location.

Potential benefits: 1. Equal or improved ability to detect metastatic disease in breast cancer patients at a decreased radiation dose. 2. An improved understanding of index tumor metrics help stratify risk for metastatic disease based on the index tumor and enable individually tailored surveillance.

Inclusion Criteria:

Any patient with a history or breast cancer undergoing PET/CT either for initial staging or for disease surveillance

-

Exclusion Criteria:

• Pregnant patients and patients with known contraindications for whole body MR imaging (e.g., pacemakers, recent surgery, brain vascular clips, etc.) will be excluded from the study. Patients will be screened with a questionnaire to be sure they have no medical devices that could make the procedure unsafe. Patients with glomerular filtration rate (GFR) < 15 ml/min/1.73m2 or who are on dialysis will be excluded from the study.