Functional Magnetic Resonance Imaging Response of Targeted Tumor Burden and Its Impact on Survival in Patients With Hepatocellular Carcinoma

Objective: The aim of this study was to evaluate response of the targeted tumor burden by functional magnetic resonance imaging (MRI) including volumetric diffusion-weighted imaging and volumetric contrast-enhanced MRI (CE-MRI) and its impact on survival in patients with hepatocellular carcinoma treated with intra-arterial therapy (IAT). Materials and Methods: This institutional review board-approved, Health Insurance Portability and Accountability Act-compliant retrospective study included 157 hepatocellular carcinoma lesions in 97 patients (78 men and 19 women; mean age, 64 years) treated with IAT. All patients had pretreatment and 3-to 4-week follow-up MRI with diffusion-weighted imaging and CE-MRI. All lesions 2 cm or larger that were targeted during the first session of IAT were segmented using research software (MR-Oncotreat) to determine targeted tumor burden relative to liver volume (%). Targeted tumor burden was stratified into low (<= 10%) or high (> 10%). Response using volumetric functional apparent diffusion coefficient (ADC; increase by >= 25%) and CE-MRI (decrease by >= 50% and >= 65% in arterial and venous enhancement [VE], respectively) was assessed in all targeted tumors (range, 1-11) using paired t tests. Kaplan-Meier survival analysis was performed and log-rank test was used to compare pairs of survival curves. Multivariate Cox regression analysis was performed to determine the simultaneous effect of treatment response and tumor burden on survival after adjusting for age, sex, and Child Pugh status. Results: There was a significant increase in volumetric ADC (median, 15%; P < 0.001) and a decrease in volumetric arterial enhancement (AE) and VE (median AE, -43% and portal venous phase (PVP), -29%, respectively; P < 0.001) 3 to 4 weeks after treatment in the targeted tumor burden. Multivariable Cox regression demonstrated that both ADC response and low tumor burden were independently associated with greater survival (hazard ratios, 0.53 and 0.55; P values, 0.025 and 0.016, respectively) after adjustment for age, sex, and Child Pugh status. Multivariable Cox regression models demonstrated no statistically significant relationship between AE response and survival after adjusting for tumor burden. However, multivariable Cox regression demonstrated that VE response was associated with greater survival only in those with low tumor burden (hazard ratio, 0.10; P = 0.001), indicating a strong interaction between VE response and tumor burden. Conclusion: Quantifying targeted tumor burden is important in predicting patient survival when using functional MRI metrics in assessing treatment response.