Ultraselective transcatheter arterial chemoembolization for small hepatocellular carcinoma guided by automated tumor-feeders detection software: technical success and short-term tumor response.

Purpose: To analyze the technical success and tumor response of ultraselective transcatheter arterial chemoembolization (TACE) for small hepatocellular carcinoma (HCC) using automated tumor-feeders detection (AFD) software. Methods: Prototype AFD software was prospectively applied to cone-beam computed tomography images acquired during TACE for 155 consecutive HCCs ≤50 mm in 81 patients. The detectability of tumor-feeding subsubsegmental arteries was analyzed. Technical success of TACE was classified into three grades according to 1-week CT; the tumor was embolized with a safety margin (5 mm wide for tumors <25 mm, and 10 mm wide for tumors ≥25 mm) (grade A), without a margin in parts (grade B), or the entire tumor was not embolized (grade C). Tumor response at 2-3 months after TACE was also evaluated in 71 patients using the modified Response Evaluation Criteria in Solid Tumors. Results: One-hundred and twenty-eight (82.6%) tumors were classed as grade A, 17 (11%) as grade B, and 10 (6.5%) as grade C. AFD software could identify 211 (85.4%) of 247 tumor-feeders but not 36 (14.6%). Eighteen (7.9%) were false positive. The tumor response of target lesions in each patient was complete response (CR) in 49 (69%) patients, partial response (PR) in 19 (26.8%), and stable disease (SD) in 3 (4.2%). The overall tumor response was CR in 39 (54.9%) patients, PR in 15 (21.2%), SD in 1 (1.4%), and progressive disease in 16 (22.5%). Conclusions: AFD software has sufficient performance to identify tumor-feeders and contributes to the high technical success in ultraselective TACE. [ABSTRACT FROM AUTHOR]