Breast tumour volume and blood flow measured by MRI after one cycle of epirubicin and cyclophosphamide-based neoadjuvant chemotherapy as predictors of pathological response

Accelerated 3D high-resolution T2-weighted breast MRI with deep learning constrained compressed sensing, comparison with conventional T2-weighted sequence on 3.0 T

PURPOSE

To evaluate the feasibility of isotropic 3D high-resolution T2-weighted imaging (T2WI) MRI sequences and compare the images reconstructed by integrating artificial intelligence-compressed sensing (AI-CS), compressed sensing (CS), and conventional 2D T2WI sequences for quality.

MATERIALS AND METHODS

Fifty-two female patients (ages: 26-80 years) with suspected breast cancer were enrolled. They underwent breast MRI examinations using three sequences: conventional T2WI, CS 3D T2WI, and AI-CS 3D T2WI. Image quality, signal-to-noise ratio (SNR), contrast-to-noise ratio, tumor volume, and maximal tumor diameter were compared using the Friedman test. Image quality was scored on a 5-point scale, with 1 indicating nonassessable quality and 5 indicating excellent quality. Tumor volume and maximal tumor diameter were compared based on AI-CS 3D T2WI (slightly high signal), conventional T2WI, and dynamic contrast-enhanced (DCE) sequences.

RESULTS

All three T2WI were successfully performed in all patients. 3D CS and AI-CS were significantly better than conventional T2WI in terms of lesion conspicuity and morphology, structural details, overall image quality, diagnostic information for breast lesions, and breast tissue delineation (P < 0.001). The SNR of conventional T2WI was significantly higher for 3D T2WI sequences. The contrast-to-noise ratio was significantly higher for AI-CS 3D T2WI than for conventional T2WI sequence. There was no significant difference in tumor volume between DCE (8.08 ± 16.51) and AI-CS 3D T2WI (8.25 ± 16.29) sequences and no significant differences in tumor diameter among DCE, AI-CS 3D T2WI, and conventional T2WI sequences.

CONCLUSION

Isotropic-resolution 3D T2WI sequences can be acquired using AI-CS while maintaining image quality and diagnostic value, which may pave the way for isotropic 3D high-resolution T2WI for clinical application.

Evaluation of Monoexponential, Stretched-Exponential and Intravoxel Incoherent Motion MRI Diffusion Models in Early Response Monitoring to Neoadjuvant Chemotherapy in Patients With Breast Cancer-A Preliminary Study

BACKGROUND

There has been a growing interest in exploring the applications of stretched-exponential (SEM) and intravoxel incoherent motion (IVIM) models of diffusion-weighted imaging (DWI) in breast imaging, with the focus on differentiation of breast lesions. However, the use of SEM and IVIM models to predict early response to neoadjuvant chemotherapy (NACT) has received less attention.

PURPOSE

To investigate the value of monoexponential, SEM, and IVIM models to predict early response to NACT in patients with primary breast cancer.

STUDY TYPE

Prospective.

POPULATION

Thirty-seven patients with primary breast cancer (aged 46 ± 11 years) due to undergo NACT.

FIELD STRENGTH/SEQUENCES

A 1.5-T MR scanner, T1 -weighted three-dimensional spoiled gradient-echo, two-dimensional single-shot spin-echo echo-planar imaging sequence (DWI) at six b-values (0-800 s mm-2 ).

ASSESSMENT

Tumor volume, apparent diffusion coefficient, tissue diffusion (Dt ), pseudo-diffusion coefficient (Dp ), perfusion fraction (f), distributed diffusion coefficient, and alpha (α) were extracted, following volumetric sampling of the tumors, at three time-points: pretreatment, post one and three cycles of NACT.

STATISTICAL TESTS

Mann-Whitney test, receiver operating characteristic (ROC) curve. Statistical significance level was P < 0.05.

RESULTS

Following NACT, 17 patients were determined to be pathological responders and 20 nonresponders. Tumor volume was significantly larger in nonresponders at each MRI time-point and demonstrated reasonable performance in predicting response (area under the ROC curve [AUC] = 0.83-0.87). No significant differences between groups were found in the diffusion coefficients at each time-point (P = 0.09-1). The parameters α (SEM), f, and f × Dp (IVIM) were able to differentiate between response groups after one cycle of NACT (AUC = 0.73, 0.72, and 0.74, respectively).

CONCLUSION

Diffusion coefficients derived from the monoexponential, SEM, and IVIM models did not predict pathological response. However, the IVIM-derived parameters f and f × Dp and the SEM-derived parameter α were able to predict response to NACT in breast cancer patients following one cycle of NACT.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY STAGE: 2.