Mechanism and Natural Course of Tumor Involution in Hepatocellular Carcinoma Following Transarterial Ethanol Ablation.

PURPOSE: To evaluate the microvascular distribution of lipiodol-ethanol, the histological change of the tumor lesion, and the status of tumor involution over time in hepatocellular carcinoma (HCC) following transarterial ethanol ablation (TEA), in lesions that showed CT evidence of complete tumor response. MATERIALS AND METHODS: Patients with unresectable HCC were treated (183 patients, 242 lesions) with TEA using lipiodol-ethanol mixture (LEM) mixed in 2:1 ratio by volume and followed with CT at 3-month intervals for a median of 14.1 months. Liver tumors (n = 131) that showed CT evidence of complete tumor response, defined as the absence of any enhancing tumor throughout the follow-up period, were included. The surgical specimens of five patients who subsequently received partial hepatectomy were available for histological assessment. The microvascular distribution of LEM and the degree of tumor necrosis were analyzed. Tumor involution over time was assessed with CT in lesions that showed complete response. RESULTS: Lipid stain revealed lipiodol infiltration throughout arterioles, intratumoral sinusoidal spaces, tumor capsule, and peritumoral portal venules. Complete tumor necrosis (100 %) occurred in all 5 surgical specimens. The median (IQR) percentage tumor volume compared to baseline volumes at 12, 36, and 60 months was 32 % (23.5-52.5 %), 22 % (8-31 %), and 13.5 % (6-21.5 %), respectively. CONCLUSION: Intrahepatic HCC lesion that showed CT evidence of complete tumor response following TEA is associated with histological evidence of LEM infiltration throughout the intratumoral and peritumoral vasculature and complete tumor necrosis, as well as sustained reduction in tumor volume over time.

Unresectable hepatocellular carcinoma: randomized controlled trial of transarterial ethanol ablation versus transcatheter arterial chemoembolization.

PURPOSE: To compare effectiveness of transarterial ethanol ablation (TEA) and transcatheter arterial chemoembolization (TACE) for unresectable hepatocellular carcinoma and determine whether TEA leads to better overall survival and tumor response than TACE. MATERIALS AND METHODS: In this institutional review board-approved preregistered randomized controlled trial (n = 200), informed consent was obtained. Primary outcome was overall survival; secondary outcomes were time to progression (TTP), progression-free survival (PFS), tumor response at computed tomography, and treatment-related toxicity. Eligible patients were randomized at a 1:1 ratio. Treatment included transcatheter delivery of ethiodized oil-ethanol mixture (2:1 ratio by volume up to 60 mL) for TEA and cisplatin-ethiodized oil emulsion (0.5 mg cisplatin per milliliter up to 30 mg), followed by 1-mm gelatin-sponge pellets, for TACE. Study was terminated after interim analysis (n = 98); 90 patients were available for analysis. Overall survival, TTP, and PFS were analyzed with Kaplan-Meier method; differences were compared with log-rank test. RESULTS: Study was terminated prematurely after interim analysis, which showed no difference in overall survival; this was unlikely to change with further patient accrual. Median overall survival in TEA and TACE was 24.3 months (95% confidence interval [CI]: 12.8, 32.7) and 20.1 months (95% CI: 9.3, 31.2), respectively (P = .358). Median TTP and PFS for intralesional progression were longer with TEA than TACE (TTP, 34.6 months [95% CI: 28.2, 41] vs 26.05 months [95% CI: 18.7, 33.3]; PFS, 14.8 months [95% CI: 10.2, 19.5] vs 9.3 months [95% CI: 7.1, 11.5]) (P = .028 and 0.029, respectively). Complete response rate on a tumor basis was persistently and significantly higher with TEA at 3 months (62 of 88 [70%] vs 39 of 76 [51%], P = .012), 6 months (64 of 88 [73%] vs 41 of 76 [54%], P = .012), and 12 months (66 of 88 [75%] vs 45 of 76 [59%], P = .031). CONCLUSION: Although there was no significant difference in overall survival, TEA demonstrated better complete tumor response, longer time to intralesional progression, and longer PFS.