Chemical exchange saturation transfer (CEST) MRI is a novel molecular MRI technique that can detect subtle chemical changes by detecting interactions of protons between metabolite and bulk water. Amide proton transfer (APT) imaging is the representative endogenous CEST imaging that has been applied to some clinical cancer studies. Especially in brain, previous reports have demonstrated the usefulness of APT imaging in tumor detection and characterization(1-3). However, little is known about the clinical usefulness of APT imaging in breast cancer. The aim of this study were to determine the possible utility of amide proton transfer (APT) MR imaging in breast cancer.

Approval for this study was obtained from the Institutional Review Board of our institution, and informed consent was waived due to retrospective study design. Twenty-three female patients with pathologically confirmed invasive breast cancer underwent breast MRI. All examination was performed at a 3T scanner (MAGNETOM Trio, A Tim System, Siemens GmbH) with 16ch breast coil. Axial bilateral breast APT CEST imaging were obtained using a proto-type 3D gradient-echo pulse sequence with the following settings: TR/TE, 6.3/1.78ms; Flip angle, 12°; voxel size, 1.8×1.8×3.0mm³; 16slices. Pre-saturation pulses consisted of 5 consecutive RF pulses of 99-ms duration with 60-ms inter-pulse delays and 1.2 μT time-average amplitude. First, baseline images were acquired without pre-saturation pulse (S₀ image). And then, images with pre-saturation pulses at different offsets (0, ±0.6, ±1.2, ±1.8, ±2.4, ±3.0, ±3.6, ±4.2, and ±4.8 ppm) from the bulk water resonance were acquired (S_Sat). Selective excitation of water was used to avoid signal from lipid resonance. Total scan time was 4min 49s.

The asymmetry of the magnetization transfer rate (MTR_asym) at 3.5ppm (APT_asym) was calculated by integration of the attenuated signal (S_Sat) between +3.2 and +3.8 ppm and subtracting that from the integrated S_Sat between –3.2 and –3.8 ppm, normalized against S₀. S_Sat(Δω) was obtained after interpolation of the originally sampled points to a resolution of 0.05 ppm and subsequent correction for inhomogeneity of the static magnetic field by Z-spectrum shifting similar to a previously described method by Schmitt et al (4). Obtained APT_asym maps (APT images) were evaluated in comparison with B₀ (to rule out uncorrectable confounding B₀ inhomogeneities post shimming), dynamic contrast-enhanced (DCE) T1-weighted, T2-weighted and T1-weighted images. Tumor size was determined using DCE T1-weighted images. One lesion without DCE T1-weighted images was evaluated using diffusion-weighted images instead.

In total, 23 patients with 46 breasts were investigated and 24 cancer lesions were identified. In APT images, no lesions presented intratumoral hyperintensity compared with normal breast tissue and 6 lesions presented hyperintensity in peritumoral breast tissue (named peritumoral hyperintensity; PT-HI) compared with contralateral breast tissue. Two representative cases of PT-HI are shown in Fig 1 and 2a. Fig 2a&b shows the disappearance of PT-HI through neoadjuvant chemotherapy (NAC). Table 1 shows the characteristics of all lesions, lesions with PT-HI (PT-HI group) and lesions without PT-HI (non PT-HI group).

It is noteworthy that no breast cancer lesions presented intratumoral hyperintensity, which was different from previous reports in brain tumors demonstrating intratumoral hyperintensity compared with normal tissue. For now, we are not sure what contribute to the difference, but the difference in amide concentration and tissue physicochemical properties (pH and temperature) between breast cancer and brain tumors may exist. Also, PT-HI in breast cancer has not been reported before, as far as we know. Considering the disappearance of PT-HI through NAC treatment demonstrated in Fig 2a&b, we speculate PT-HI may reflect the change in microenvironment induced by cancer (e.g., interstitial protein accumulation, lymph stasis), though our data didn’t indicate the specific characteristics to PT-HI group. Further studies in larger number of lesions are needed.

We demonstrated a pilot study of APT imaging of breast cancer. Providing biochemical information non-invasively, APT CEST imaging is a potentially useful MRI technique for breast cancer.