Locally advanced breast cancer (LABC) is most frequently treated by neoadjuvant systemic therapy prior to definitive surgical resection. Dynamic contrast enhanced breast magnetic resonance imaging (DCE-MRI) and positron emission tomography (PET) have been used to evaluate response to traditional cytotoxic neo-adjuvant chemotherapy (1
). In addition, there is growing interest in targeted breast cancer therapy, such as anti-angiogenic agents, for which non-invasive imaging may be particularly advantageous for response evaluation (3
Prior studies have examined DCE-MRI and PET functional studies independently. DCE-MRI defines the extent of breast cancer in vivo more accurately than any other imaging modality (18
). In addition, DCE-MRI data can provide measurements of tumor volume and enhancement kinetics. Changes in tumor volume following neoadjuvant therapy, as measured using DCE-MRI, are predictive of recurrence free survival (8
). Measurements of tumor enhancement, including semi-quantitative parameters such as initial percent enhancement (PE) and delayed signal enhancement ratio (SER), have been shown to be helpful and accurate for tumor detection and response evaluation (19
). Tumor enhancement likely reflects a combination of blood flow and capillary permeability (15
). However, the precise biologic factors underlying enhancement and the relative contributions of blood flow and capillary permeability are not completely understood.
F-fluorodeoxyglucose (FDG) PET has also been widely studied in the setting of neoadjuvant breast cancer therapy (2
). Changes in glucose metabolism by FDG PET have been shown to predict response to treatment following a single cycle of chemotherapy (28
). PET measures of tumor perfusion have also been helpful for response assessment through dynamic imaging with 15
O-water, a test that can provide an independent and quantitative measure of blood flow in breast cancer (30
). Changes in blood flow following neoadjuvant therapy, as quantified by 15
O-water PET, predict pathologic response and disease free survival (5
). Recent studies comparing 15
O-water and FDG PET have shown that the delivery of FDG, measured by the kinetic parameter FDG K1
, and blood flow measured by 15
O-water PET are highly correlated, suggesting that blood flow is an important factor in FDG delivery (11
). These results and the increased use of imaging to monitor therapy motivate an investigation of the relationship between DCE-MRI parameters, PET measures of blood flow and PET FDG kinetics to better understand their relationships and the nature of DCE-MRI kinetic parameters.
Prior studies have examined the relationship between DCE-MRI and FDG PET in predicting response for breast cancer (2
). Pilot studies have been performed to validate the correlation of 15
O-water PET blood flow with MRI in the brain, prostate and coronary arteries (34
). PET blood flow and DCE-MRI kinetic parameters have not been directly compared or correlated in the breast. In addition to a better understanding of the determinants of MRI contrast enhancement, comparing DCE-MRI and 15
O-water PET may improve our ability to predict therapeutic response and provide insights into the mechanisms of anti-angiogenic therapies through quantitative regional differences in blood flow and capillary permeability.
There were two purposes of this study: To (1) test the correlations between DCE-MRI initial percent enhancement (PE) and delayed signal enhancement ratio (SER) with 15
O-water PET blood flow and 18
F-FDG PET metabolic rate (MR) and tissue transport constant (K1
) and (2) improve our understanding of DCE-MRI tumor enhancement through independent measures of tumor metabolism and blood flow. Based upon our studies correlating blood flow and metabolism by PET, we hypothesized that, in LABC, DCE-MRI kinetics (peak PE and peak SER) would directly correlate with measures of tumor perfusion, including blood flow, as measured by 15
O-water PET and FDG K1
). We performed an exploratory retrospective study to test this hypothesis.