To determine the performance of magnetic resonance (MR) imaging–based tumor metrics for evaluation of response to transarterial chemoembolization (TACE) in patients with unresectable intrahepatic cholangiocarcinoma (ICCA).

Ninety-four patients with unresectable ICCA underwent baseline and follow-up MR imaging after TACE and were followed up until death or end of study duration. Lesions were analyzed for anatomic (Response Evaluation Criteria in Solid Tumors [RECIST] and tumor volume) and functional (viable tumor volume, viable tumor burden, and apparent diffusion coefficient [ADC]) volumetric MR parameters by using semiautomatic software. Response was assessed by using changes in viable tumor volume by using modified RECIST (mRECIST)–derived thresholds (three-dimensional mRECIST), viable tumor burden, and ADC. Overall survival was the primary endpoint. Cox-regression and Kaplan-Meier survival analysis were used.

Tumor volume did not change after TACE (P = .07) whereas RECIST diameter showed a small change (−2.6%; P = .02). There was an increase in ADC (20.7%) and a decrease in viable tumor volume (−29.3%) and viable tumor burden (−29.1%; P < .001 for all). Higher overall survival was noted among responders by using thresholds of 25% increase in ADC, 66% decrease in viable tumor volume, and 50% decrease in viable tumor burden (log-rank test, P < .05). Hazard ratio for nonresponders by using ADC, three-dimensional mRECIST, and viable tumor burden at multivariable analysis was 2.9 (P = .004), 4.1 (P = .009), and 4.0 (P = .002), respectively. Survival differences were noted for patients who showed response by using all three parameters (ADC, three-dimensional mRECIST, and viable tumor burden) versus those who showed response by using either one or two of these parameters versus those who showed no response (P < .001).

Changes in volumetric ADC, viable tumor volume, and viable tumor burden at MR imaging provide prognostic information among patients with unresectable ICCA who undergo TACE.

^© RSNA, 2018