Apparent diffusion coefficient measurements as very early predictive markers of response to chemotherapy in hepatic metastasis: A preliminary investigation of reproducibility and diagnostic value

Abstract

PURPOSE: To evaluate the reproducibility and diagnostic value of apparent diffusion coefficient (ADC) as an early predictor of response to chemotherapy of liver metastasis in routine clinical practice. MATERIALS AND METHODS: A prospective study of 20 patients with histologically proven primary tumors with liver metastases was undertaken. Diffusion weighted MRI was performed twice before and 12-14 days after the start of treatment. Absolute and liver normalized ADC values were calculated. Bland Altman statistics were used to assess the reproducibility of ADC change for predicting lesion response as measured by RECIST. RESULTS: Nineteen of 31 metastases responded. Significant increases in absolute and normalized ADC values were found in responding (mean +208.7 × 10(-6) m(2)/s and +18% respectively, both P < 0.001) compared with nonresponding lesions (mean +98.6 × 10(-6) m(2)/s and 2%, respectively, P = 0.09 and 0.519). Reproducibility was better using normalized ADC compared with absolute ADC values (within patient coefficient of variability 8.0% and 10.1%, respectively). Using the repeatability threshold of ±22.3% for normalized ADC, only 8 of 19 responding and all but one nonresponding lesions could be prospectively detected. CONCLUSION: Increases in ADC values in responding liver metastases occurred within days after the start of chemotherapy but were of smaller magnitude than the variability of ADC measurement. These preliminary data suggest that the presently used technique is not reliable enough to predict final response at such an early time point in individual lesions.