Lisa Loi1, Ferdinand Zimmermann2, Steffen Goerke3, Andreas Korzowski4, Jan-Eric Meissner5, Katerina Deike-Hofmann6, Anne Stieber7, Peter Bachert8, Mark Edward Ladd9, Heinz-Peter Schlemmer10, Sebastian Bickelhaupt11, Sarah Schott12, Daniel Paech13. 1. Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany; Faculty of Medicine, University of Heidelberg, Im Neuenheimer Feld 672, 69120 Heidelberg, Germany. Electronic address: l.loi@dkfz-heidelberg.de. 2. Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany; Faculty of Physics and Astronomy, University of Heidelberg, Im Neuenheimer Feld 226, 69120 Heidelberg, Germany. Electronic address: f.zimmermann@dkfz-heidelberg.de. 3. Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: s.goerke@Dkfz-Heidelberg.de. 4. Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: a.korzowski@Dkfz-Heidelberg.dej. 5. Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: meissner@Dkfz-Heidelberg.de. 6. Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: k.deike@dkfz-heidelberg.de. 7. Department of Clinical and Interventional Radiology, University Hospital of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany. Electronic address: anne.stieber@med.uni-heidelberg.de. 8. Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany; Faculty of Physics and Astronomy, University of Heidelberg, Im Neuenheimer Feld 226, 69120 Heidelberg, Germany. Electronic address: p.bachert@dkfz-heidelberg.de. 9. Faculty of Medicine, University of Heidelberg, Im Neuenheimer Feld 672, 69120 Heidelberg, Germany; Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany; Faculty of Physics and Astronomy, University of Heidelberg, Im Neuenheimer Feld 226, 69120 Heidelberg, Germany. Electronic address: mark.ladd@dkfz-heidelberg.de. 10. Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: h.schlemmer@Dkfz-Heidelberg.de. 11. Junior Group Medical Imaging and Radiology - Cancer Prevention, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: s.bickelhaupt@Dkfz-Heidelberg.de. 12. Department of Gynecology and Obstetrics, University Hospital of Heidelberg, Im Neuenheimer Feld 440, 69120 Heidelberg, Germany. Electronic address: sarah.schott@med.uni-heidelberg.de. 13. Division of Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Electronic address: d.paech@dkfz.de.
Abstract
PURPOSE: To investigate whether fat-corrected and relaxation-compensated amide proton transfer (APT) and guanidyl CEST-MRI enables the detection of signal intensity differences between breast tumors and normal-appearing fibroglandular tissue in patients with newly-diagnosed breast cancer. METHOD: Ten patients with newly-diagnosed breast cancer and seven healthy volunteers were included in this prospective IRB-approved study. CEST-MRI was performed on a 7 T-whole-body scanner followed by a multi-Lorentzian fit analysis. APT and guanidyl CEST signal intensities were quantified in the tumor and in healthy fibroglandular tissue after correction of B0/B1-field inhomogeneities, fat signal contribution, T1- and T2-relaxation; signal intensity differences of APT and guanidyl resonances were compared using Mann-Whitney-U-tests. Pearson correlations between tumor CEST signal intensities and the proliferation index Ki-67 were performed. RESULTS: APT CEST signal in tumor tissue (6.70 ± 1.38%Hz) was increased compared to normal-appearing fibroglandular tissue of patients (3.56 ± 0.54%Hz, p = 0.001) and healthy volunteers (3.70 ± 0.68%Hz, p = 0.001). Further, a moderate positive correlation was found between the APT signal and the proliferation index Ki-67 (R2 = 0.367, r = 0.606, p = 0.11). Guanidyl CEST signal was also increased in tumor tissue (5.24 ± 1.85%Hz) compared to patients' (2.42 ± 0.45%Hz, p = 0.006) and volunteers' (2.36 ± 0.54%Hz, p < 0.001) normal-appearing fibroglandular tissue and a positive correlation with the Ki-67 level was observed (R2 = 0.365, r = 0.604, p = 0.11). APT and guanidyl CEST signal in normal-appearing fibroglandular tissue was not different between patients and healthy volunteers (p = 0.88; p = 0.93). CONCLUSION: Relaxation-compensated and fat-corrected CEST-MRI allowed a non-invasive differentiation of breast cancer and normal-appearing breast tissue. Thus, this approach represents a contrast agent-free method that may help to increase diagnostic accuracy in MR-mammography.
PURPOSE: To investigate whether fat-corrected and relaxation-compensated amide proton transfer (APT) and guanidyl CEST-MRI enables the detection of signal intensity differences between breast tumors and normal-appearing fibroglandular tissue in patients with newly-diagnosed breast cancer. METHOD: Ten patients with newly-diagnosed breast cancer and seven healthy volunteers were included in this prospective IRB-approved study. CEST-MRI was performed on a 7 T-whole-body scanner followed by a multi-Lorentzian fit analysis. APT and guanidyl CEST signal intensities were quantified in the tumor and in healthy fibroglandular tissue after correction of B0/B1-field inhomogeneities, fat signal contribution, T1- and T2-relaxation; signal intensity differences of APT and guanidyl resonances were compared using Mann-Whitney-U-tests. Pearson correlations between tumor CEST signal intensities and the proliferation index Ki-67 were performed. RESULTS: APT CEST signal in tumor tissue (6.70 ± 1.38%Hz) was increased compared to normal-appearing fibroglandular tissue of patients (3.56 ± 0.54%Hz, p = 0.001) and healthy volunteers (3.70 ± 0.68%Hz, p = 0.001). Further, a moderate positive correlation was found between the APT signal and the proliferation index Ki-67 (R2 = 0.367, r = 0.606, p = 0.11). Guanidyl CEST signal was also increased in tumor tissue (5.24 ± 1.85%Hz) compared to patients' (2.42 ± 0.45%Hz, p = 0.006) and volunteers' (2.36 ± 0.54%Hz, p < 0.001) normal-appearing fibroglandular tissue and a positive correlation with the Ki-67 level was observed (R2 = 0.365, r = 0.604, p = 0.11). APT and guanidyl CEST signal in normal-appearing fibroglandular tissue was not different between patients and healthy volunteers (p = 0.88; p = 0.93). CONCLUSION: Relaxation-compensated and fat-corrected CEST-MRI allowed a non-invasive differentiation of breast cancer and normal-appearing breast tissue. Thus, this approach represents a contrast agent-free method that may help to increase diagnostic accuracy in MR-mammography.