Chemoembolization follow-up of hepatocellular carcinoma with MR imaging: Usefulness of evaluating enhancement features on one-month posttherapy MR imaging for predicting residual disease

Purpose To determine the sensitivity, specificity, and accuracy of contrast-enhanced magnetic resonance (MR) imaging performed 1 month after localized chemotherapy as a measure of tumor response, before detectable changes in size. Materials and Methods This trial was approved by the authors' institutional review board and was compliant with the Health Insurance Portability and Accountability Act (HIPAA). Inclusion criteria selected patients receiving chemoembolization for hepatocellular carcinoma (HCC) with MR imaging within 2 months before treatment, in addition to MR imaging after treatment at 1 month and 6 months. Pathology was used as a surrogate for 6-month follow-up if the patient underwent interval transplantation. The final population consisted of 23 tumors (occurring within 21 patients). MR imaging studies were evaluated separately by two radiologists. Tumors were scored as showing complete loss of enhancement or as showing some residual tissue enhancement. Changes in T1 and T2 signal and perilesional enhancement were tabulated and recorded. Lesion size was also measured on all MR imaging studies by using a one-dimensional measure of the longest dimension. Increase in tumor size from 16 months of 20% or greater was used as confirmation of residual disease. In 5 of 23 tumors, review of pathology served as the surrogate standard. Sensitivity, specificity and accuracy were computed for each rater. Results The sensitivity, specificity, and accuracy of 1-month follow-up MR imaging were 71.485.7%, 100%, and 91.395.7%. There was a high degree of agreement between the two readers for both the 1-month (κ = 0.88) and 6-month (κ = 1.0) MR imaging studies. Conclusions This investigation shows high accuracy for using tumor enhancement features on 1-month posttherapy MR imaging to predict residual disease after chemoembolization of HCC.