Hepatic artery chemoembolization for 110 gastrointestinal stromal tumors - Response, survival, and prognostic factors

B ACKGROUND. The efficacy of hepatic artery chemoembolization (HACE) was evaluated for gastrointestinal stromal tumors (GISTs) metastatic to the liver. METHODS. Records for patients with metastatic GIST who underwent HACE between January 1993 and March 2005 were reviewed and cross-sectional images were used to determine objective tumor response. Progression-free survival in the liver (PFS-liver) and overall survival (OS) were calculated with the Kaplan-Meier method. Patient, tumor, and treatment variables were analyzed to identify factors influencing survival. RESULTS. Of the 110 patients identified, the radiologic response to HACE could be evaluated in 85 patients, 12 of whom (14%) demonstrated partial responses, 63 of whom (74%) demonstrated stable disease, and 10 of whom (12%) demonstrated progressive disease. PFS-liver rates were 31.2%, 8.2%, and 5.4% at 1, 2, and 3 years, respectively; the median PFS time was 8.2 months. OS rates were 62% at 1 year, 32% at 2 years, and 20% at 3 years; the median OS time was 17.2 months. Patients who had > 5 liver metastases and received only 1 HACE treatment were found to have a shorter PFS compared with patients with fewer metastases or those who received >= 2 HACE sessions. Extensive liver involvement, the presence of extrahepatic metastases, and progression of liver disease after HACE were associated with poor OS. Use of imatimb prolonged OS time. CONCLUSIONS. HACE produced a durable tumor response or disease stabilization in the majority of patients with GISTs metastatic to liver. Extent of liver disease, presence of extrahepatic disease, number of embolization treatments, and use of imatinib were found to have prognostic influence on PFS, OS, or both.