Synopsis

The fat correction method enables robust APT-CEST quantification in the human breast and proved its suitability for examinations in vivo. We present to the extent of our knowledge the first APT-CEST contrast corrected for fat signal contribution, spillover, B₁ field inhomogeneities and T₁ relaxation in a breast cancer patient. The CEST contrast increased threefold compared to the measurement of a healthy volunteer. Repeated CEST imaging over the course of one menstrual cycle in one healthy woman did not reveal a hormonal correlation of APT contrast. A clinical study in healthy premenopausal volunteers will now investigate the dependency on menstrual cycle.

INTRODUCTION

METHODS

One 58-year-old patient diagnosed with an invasive mucinous mamma carcinoma (G2) was examined. Beginning at the 10^th day of the menstrual cycle, one healthy volunteer (31-year-old) was examined six times over the course of 37 days. Her subsequent menstrual period started on the fifth measurement (29 days past the first examination).

A modified WASABI-sequence⁷ (single rectangular pulse of t_p=2.5ms, B₁=7µT, followed by five 90° fat saturation pulses) yielded robust field-mapping. A T₁-map was acquired by a saturation recovery sequence. CEST-Saturation was achieved by a train of 297 Gaussian-shaped pulses ($$$t_p~=~15~ms$$$, t_sat = 5.6 s, DC = 80%). 75 offsets, $$$\vec{M_z(\Delta\omega)}$$$, were acquired for two different saturation powers of B₁ = 0.6 & 0.9 µT. Single-slice imaging TE_in-phase = 2.04 ms, resolution 1.5x1.5x5 mm³, BW = 1220 Hz/pix) was performed on a 7T MR scanner (Siemens Magnetom) employing a bilateral breast coil (Rapid Biomedical).

Assuming total water signal saturation after pre-saturation at $$$\Delta\omega~=~0~ppm$$$ without lipid saturation one can estimate the collective fat signal as the residual magnetization: $$$\vec{M_z}(0 ppm) = \vec{F_0}$$$. This allows performing the fat corrected CEST-normalization for amide resonances^1,2: $$Z_{corr.} =\frac{|\vec{M_z}(\Delta\omega)~-~\vec{M_z}(0 ppm)|}{|\vec{M_0}-\vec{M_z}(0 ppm)|}=\frac{|\vec{M_z}(\Delta\omega)~-~\vec{F_0}|}{|\vec{M_0}~-~\vec{F_0}|}\quad\quad\quad\quad\quad\quad[Eq.1]$$ $$$\vec{M_0}$$$ denotes the fully relaxed image. Evaluation of complex valued raw data allowed voxelwise determination of $$$\vec{F_0}$$$ and calculation of corrected Z-spectra by Eq.1. A five-pool Lorentzian function⁹ was fitted to the APT signal and $$$AREX_{APT}~=~\frac{1}{DC\cdot~T_1}\cdot(\frac{1}{Z_{lab}}-\frac{1}{Z_{ref}})$$$ was calculated⁸ ($$$\Delta\omega=+3.5ppm$$$) and corrected for B₁-inhomogeneities (reconstructed B₁ = 0.8 μT) ⁹.

RESULTS

Slight changes of APT signal are observable comparing the uncorrected and corrected Z-spectra in the healthy volunteer (Fig.1A&B). An overall increase of the fat corrected contrast can be seen (Fig.1C-E), mainly in the transition from glandular to fatty tissue. The fat corrected APT contrast appears more homogenous. However, some artifacts are observable in fat dominated voxels due to overcompensation.

The APT contrast in the breast cancer patient increases approximately threefold in comparison to the healthy volunteer (Fig.2). No correlations to B₁-inhomogeneities are observed (Fig.2B). Furthermore, signal intensities of both, amide and amine resonances increase (Fig.2C, cf. Fig.1B).

Changes of T₁ relaxation time during the menstrual cycle are observed (Fig.3A). Fat-corrected AREX_APT appears to change as well (Fig.3B). However, the contrasts deviations (up to 0.005 Hz during the cycle) are negligible in comparison to the signal difference between healthy and tumorous tissue.

DISCUSSION

The presented protocol was successfully applied in repeated examinations in a healthy volunteer, as well as in the patient. The slight increase of APT contrast after fat correction is in line with expectations due to the chosen in-phase echo time and the observed comparatively small fat fractions in the gland (<15%)^1,2. The approach works robustly despite field inhomogeneities and diverse fat contents within fibroglandular tissue. It is therefore our method of choice in future pilot studies, in which we expect a greater diversity of tissue composition with regards to lipid content.

We showed the first APT-CEST contrast corrected for fat signal contribution, spillover, B₁ field inhomogeneities and T₁ relaxation in a breast cancer patient. However, no healthy tissue adjacent to the tumor was present and prevented a direct comparison of APT contrast. The benefit of quantitative APT-CEST imaging has therefore be evaluated in further patients.

The observed fluctuations of CEST contrast within one menstrual cycle might indicate a hormonal dependence, are however not statistically significant (n=1). A possible correlation has to be verified in a larger cohort. We hypothesize that hormonally induced changes of APT are minimal and, with regard to tumor imaging, negligible.

CONCLUSION

Acknowledgements

References

1. Zimmermann F, Korzowski A, Schuenke P, et al. Fat corrected APT-CEST in the human breast. Proceedings of the 7th International Workshop on CEST Imaging, 2018, Beijing

2. Zimmermann F, Korzowski A, Schuenke P, et al. A novel normalization to correct APT-CEST in the presence of fat. Proceedings of the 26th ISMRM 2018, Paris, Poster #2222

3. Lu J, Zhou J, Cai C, et al. Observation of true and pseudo NOE signals using CEST-MRI and CEST-MRS sequences with and without lipid suppression. Magnetic Resonance in Medicine. 2015; 73(4):1615–1622

4. Zhang S, Keupp J, Wang X. et al. Z-spectrum appearance and interpretation in the presence of fat: Influence of acquisition parameters. Magnetic Resonance in Medicine, 2017; doi:10.1002/mrm.26900

5. Delille J-PP, Slanetz PJ, Yeh ED, Kopans DB, Garrido L. Physiologic Changes in Breast Magnetic Resonance Imaging during the Menstrual Cycle: Perfusion Imaging, Signal Enhancement, and Influence of the T1 Relaxation Time of Breast Tissue. Breast J. 2005; 11(4):236-41.

6. Kuhl CK, Bieling HB, Gieseke J, et al. Healthy premenopausal breast parenchyma in dynamic contrast-enhanced MR imaging of the breast: normal contrast medium enhancement and cyclical-phase dependency. Radiology, 1997;203(1), 137-144.

7. Schuenke P, Windschuh J, Roeloffs V, et al. Simultaneous mapping of water shift and B1(WASABI)-Application to field-Inhomogeneity correction of CEST MRI data. Magnetic Resonance in Medicine. 2016; 77(2):571-580.

8. Zaiss M, Xu J, Goerke S, et al. Inverse Z-spectrum analysis for spillover-, MT-, and T1-corrected steady-state pulsed CEST-MRI – application to pH-weighted MRI of acute stroke. NMR Biomed. 2014; 27: 240–252.

9. Windschuh J, Zaiss M, Meissner JE, et al. Correction of B1-inhomogeneities for relaxation-compensated CEST imaging at 7T. NMR Biomed. 2015;28:529–537.