Francescamaria Donati1, Piero Boraschi2, Federica Pacciardi3, Rosa Cervelli3, Maura Castagna4, Lucio Urbani5, Fabio Falaschi2, Davide Caramella3. 1. Department of Diagnostic Imaging, Pisa University Hospital, Via Paradisa 2, 56124 Pisa, Italy. Electronic address: francescamaria69@gmail.com. 2. Department of Diagnostic Imaging, Pisa University Hospital, Via Paradisa 2, 56124 Pisa, Italy. 3. Unit of Diagnostic and Interventional Radiology, Department of Translational Research and New Technologies in Medicine and Surgery, University of Pisa, Via Paradisa 2, 56124 Pisa, Italy. 4. Unit of Pathology, Department of Translational Research and New Technologies in Medicine and Surgery, University of Pisa, Via Roma 67, 56126 Pisa, Italy. 5. Department of General Surgery, Pisa University Hospital, Via Paradisa 2, 56124 Pisa, Italy.
Abstract
PURPOSE: The purpose of the study was to correlate the apparent diffusion coefficient (ADC) values of diffusion-weighted MR imaging (DW-MRI) by 3T device with the histological tumour regression grading (TRG) analysis of colorectal liver metastases after preoperative chemotherapy. MATERIALS AND METHODS: Our study included thirty-five patients with colorectal liver metastases who had undergone MRI by 3T device (GE DISCOVERY MR750; GE Healthcare) after preoperative chemotherapy. DW-MRI was performed using a single-shot spin-echo echo-planar sequence with multiple b-values (0, 150, 500, 1000, 1500s/mm2), thus obtaining an ADC map. For each liver lesion (more than 1cm in diameter) the fitted ADC values were calculated by two radiologists in conference and three ROIs were drawn: around the entire tumour (ADCe), at the tumour periphery (ADCp) and at the tumour center (ADCc). All ADC values were correlated with histopathological findings after surgery. Hepatic metastases were pathologically classified into five groups on the basis of TRG. Statistical analysis was performed on a per-lesion basis utilizing the one-way analysis of variance (ANOVA). This retrospective study was approved by our institutional review board; written informed consent was obtained from all patients. RESULTS: A total of 106 colorectal liver metastases were included for image analysis. TRG1, TRG2, TRG3, TRG4 and TRG5 were observed in 4, 14, 36, 35 and 17 lesions, respectively. ADCe and ADCp values were significantly higher in lesions classified as TRG1 (2.40±0.12×10-9m2/s and 2.28±0.26×10-9m2/s, respectively) and as TRG2 (1.40±0.31×10-9m2/s and 1.44±0.35×10-9m2/s), compared to TRG3 (1.16±0.13×10-9m2/s and 1.01±0.18×10-9m2/s), TRG4 (1.10±0.26×10-9m2/s and 0.97±0.24×10-9m2/s), and TRG5 (0.93±0.17×10-9m2/s and 0.82±0.28×10-9m2/s). ADCe, ADCp and ADCc values were significantly different in TRG classes (p<0.0001). Statistical correlations were found between the ADCe, ADCp, ADCc values and the TRG classes (Spearman correlation coefficient were -0.568, -0.542 and -0.554, respectively). CONCLUSION: Our study showed a significant correlation between ADC values of 3T DW-MRI and histological TRG of colorectal liver metastases after preoperative chemotherapy.
PURPOSE: The purpose of the study was to correlate the apparent diffusion coefficient (ADC) values of diffusion-weighted MR imaging (DW-MRI) by 3T device with the histological tumour regression grading (TRG) analysis of colorectal liver metastases after preoperative chemotherapy. MATERIALS AND METHODS: Our study included thirty-five patients with colorectal liver metastases who had undergone MRI by 3T device (GE DISCOVERY MR750; GE Healthcare) after preoperative chemotherapy. DW-MRI was performed using a single-shot spin-echo echo-planar sequence with multiple b-values (0, 150, 500, 1000, 1500s/mm2), thus obtaining an ADC map. For each liver lesion (more than 1cm in diameter) the fitted ADC values were calculated by two radiologists in conference and three ROIs were drawn: around the entire tumour (ADCe), at the tumour periphery (ADCp) and at the tumour center (ADCc). All ADC values were correlated with histopathological findings after surgery. Hepatic metastases were pathologically classified into five groups on the basis of TRG. Statistical analysis was performed on a per-lesion basis utilizing the one-way analysis of variance (ANOVA). This retrospective study was approved by our institutional review board; written informed consent was obtained from all patients. RESULTS: A total of 106 colorectal liver metastases were included for image analysis. TRG1, TRG2, TRG3, TRG4 and TRG5 were observed in 4, 14, 36, 35 and 17 lesions, respectively. ADCe and ADCp values were significantly higher in lesions classified as TRG1 (2.40±0.12×10-9m2/s and 2.28±0.26×10-9m2/s, respectively) and as TRG2 (1.40±0.31×10-9m2/s and 1.44±0.35×10-9m2/s), compared to TRG3 (1.16±0.13×10-9m2/s and 1.01±0.18×10-9m2/s), TRG4 (1.10±0.26×10-9m2/s and 0.97±0.24×10-9m2/s), and TRG5 (0.93±0.17×10-9m2/s and 0.82±0.28×10-9m2/s). ADCe, ADCp and ADCc values were significantly different in TRG classes (p<0.0001). Statistical correlations were found between the ADCe, ADCp, ADCc values and the TRG classes (Spearman correlation coefficient were -0.568, -0.542 and -0.554, respectively). CONCLUSION: Our study showed a significant correlation between ADC values of 3T DW-MRI and histological TRG of colorectal liver metastases after preoperative chemotherapy.