Diffusion-weighted MR imaging in liver metastases of colorectal cancer: reproducibility and biological validation.

Objectives: Before diffusion-weighted imaging (DWI) can be implemented in standard clinical practice for response monitoring, data on reproducibility are needed to assess which differences outside the range of normal variation can be detected in an individual patient. In this study we assessed the reproducibility of the apparent diffusion coefficient (ADC) values in colorectal liver metastases. To provide a biological basis for these values, their relation with histopathology was assessed.

Methods: DWI was performed twice in 1 week in patients scheduled for metastasectomy of colorectal liver metastases. Correlation between ADC values and apoptosis marker p53, anti-apoptotic protein BCL-2, proliferation marker Ki67 and serum vascular endothelial growth factor (VEGF) concentration were assessed.

Results: A good reproducibility coefficient of the mean ADC (coefficient of reproducibility 0.20 × 10(-3) mm(2)/s) was observed in colorectal liver metastases (n = 21). The ADC value was related to the proliferation index and BCL-2 expression of the metastases. Furthermore, in metastases recently treated with systemic therapy, the ADC was significantly higher (1.27 × 10(-3) mm(2)/s vs 1.05 × 10(-3) mm(2)/s, P = 0.02).

Conclusions: The good reproducibility, correlation with histopathology and implied sensitivity for systemic treatment-induced anti-tumour effects suggest that DWI might be an excellent tool to monitor response in metastatic colorectal cancer.