Chemical Shift MRI Monitoring of Chemoembolization Delivery for Hepatocellular Carcinoma: Multicenter Feasibility of Initial Clinical Translation

Abstract

Purpose To demonstrate the feasibility of using chemical shift fat-water MRI methods to visualize and measure intrahepatic delivery of ethiodized oil to liver tumors following conventional transarterial chemoembolization (cTACE). Materials and Methods Twenty-eight participants (mean age, 66 years ± 8 [SD]; 22 men) with hepatocellular carcinoma (HCC) treated with cTACE were evaluated with follow-up chemical shift MRI in this Health Insurance Portability and Accountability Act-compliant prospective, institutional review board-approved study. Uptake of ethiodized oil was evaluated at 1-month follow-up chemical shift MRI. Measurements of tumor size (MRI and CT), attenuation and enhancement (CT), fat content percentage, and tumor:normal ratio (MRI) were compared by lesion for responders versus nonresponders, as assessed with modified Response Evaluation Criteria in Solid Tumors and European Association for the Study of the Liver (EASL) criteria. Adverse events and overall survival by the Kaplan-Meier method were secondary end points. Results Focal tumor ethiodized oil retention was 46% (12 of 26 tumors) at 24 hours and 47% (18 of 38 tumors) at 1 month after cTACE. Tumor volume at CT did not differ between EASL-defined responders and nonresponders (P = .06). Tumor ethiodized oil volume measured with chemical shift MRI was statistically significantly higher for EASL-defined nonresponders (P = .02). Doxorubicin dosing (P = .53), presence of focal fat (P = .83), and a combined end point of focal fat and low doxorubicin dosing (P = .97) did not stratify overall survival after cTACE. Conclusion Chemical shift MRI allowed for assessment of tumor delivery of ethiodized oil out to 1 month after cTACE in participants with HCC and demonstrated tumor ethiodized oil volume as a potential tool for stratification of tumor response by EASL criteria. Keywords: MRI, Chemical Shift Imaging, CT, Hepatic Chemoembolization, Ethiodized Oil Clinicaltrials.gov registration no.: NCT02173119 Supplemental material is available for this article. © RSNA, 2023.

Keywords: CT; Chemical Shift Imaging; Ethiodized Oil; Hepatic Chemoembolization; MRI.

Graphical abstract

Figure 1:

In-phase (I_in) and out-of-phase (I_out) images next to noncontrast CT image demonstrates high attenuation due to tumor ethiodized oil uptake, with a corresponding schematic of I_out signal changes (dashed black line) as a function of changes in microscopic fat fraction (solid black line). I_out demonstrates signal “drop out” corresponding to microscopic tumor ethiodized oil, as shown on the corresponding noncontrast CT image.

Figure 2:

(A) Postcontrast MRI scan demonstrates an enhancing hepatocellular carcinoma (arrow) in segment 8 adjacent to the diaphragm and right atrium. (B) Posttransarterial chemoembolization noncontrast CT scan demonstrates ethiodized oil uptake within the tumor (outlined). (C) Tumor fat content overlaid on the corresponding in-phase section. The red region of interest was transferred to the (D) fat percentage chemical shift MRI map, demonstrating increased fat percentage within the lesion (outlined).

Figure 3:

Images in a 60-year-old man with hepatocellular carcinoma. (A) CT scan after treatment with conventional transarterial chemoembolization (cTACE) confirmed ethiodized oil delivery with diffuse tumor uptake. Ethiodized oil retention is observed at (B) 1-month post-cTACE chemical shift (CS) MRI and corresponding (C) baseline and (D) 1-month post-TACE contrast-enhanced (CE) MRI.

Figure 4:

(A) Noncontrast CT scan obtained within 24 hours after transarterial chemoembolization (TACE) to segment 8. (B) Chemical shift (CS) MRI fat percentage map shows diffuse uptake within the target tumor (outlined), with faint nodular signal in segment 7 (arrow) suspicious for a satellite lesion. (C, D) One-month follow-up chemical shift MRI scan shows retained ethiodized oil within the target tumor (outlined), as well as within multiple satellite lesions (arrows).

Figure 5:

Images in a 73-year-old man with hepatocellular carcinoma with noncontrast CT within 24 hours after transarterial chemoembolization to segment 8 (top left). CT depicted ethiodized oil accumulation in the tumor and surrounding hepatic parenchyma 4 months after treatment. Noncontrast CT more than 1 year after treatment was performed following chemoembolization to segment 4a, with markedly less attenuation in the treated segment 8 tumor. Follow-up contrast-enhanced (CE) MRI scan demonstrates anterior focal nodular enhancement that correlated with signal void on the fat and fat percentage images. Arterial phase enhancement colocalized with absence of fat was thought to represent progressing disease, and the participant was treated after interval growth and local progression of this segment 8 tumor.

Figure 6:

Kaplan-Meier survival curves indicate no evidence of a difference in overall survival by log-rank testing for (A) participants with (orange line) or without (blue dashed line) focal fat signal at chemical shift MRI (P = .70), (B) participants who received high (>20 mg; orange line) versus low (<20 mg; blue dashed line) doses of doxorubicin (dox) (P = .53), or (C) participants with focal fat and less than 20-mg dose of doxorubicin (orange line) versus those without (blue dashed line) (P = .97).