The efficacy of proton beam therapy (PBT) for hepatocellular carcinoma (HCC) has been reported, but insertion of fiducial markers in the liver is usually required. We evaluated the efficacy and toxicity of respiratory-gated PBT without fiducial markers for HCC located within 2 cm of the gastrointestinal tract. From March 2011 to December 2015 at our institution, 40 patients were evaluated (median age, 72 years; range, 38–87 years). All patients underwent PBT at a dose of 60 to 80 cobalt gray equivalents (CGE) in 20 to 38 fractions. The median follow-up period was 19.9 months (range, 1.2–72.3 months). The median tumor size was 36.5 mm (range, 11–124 mm). Kaplan–Meier estimates of the 2-year overall survival, progression-free survival, and local tumor control rates were 76%, 60%, and 94%, respectively. One patient (2.5%) developed a grade 3 gastric ulcer and one (2.5%) developed grade 3 ascites retention; none of the remaining patients developed grade >3 toxicities (National Cancer Institute Common Terminology Criteria for Adverse Events ver. 4.0.). This study indicates that PBT without fiducial markers achieves good local control without severe treatment-related toxicity of the gastrointestinal tract for HCC located within 2 cm of the gastrointestinal tract.

To retrospectively evaluate the arterial blood supply to the posterior aspect of segment IV of the liver with computed tomography (CT) after transcatheter arterial chemoembolization (TACE) with iodized oil through the caudate arterial branch of the liver for treatment of hepatocellular carcinoma (HCC).Institutional review board approval and patient informed consent were not required for this retrospective study. Twenty-four patients (11 men and 13 women; mean age, 68 years) with HCC originating in the caudate lobe (n = 23) or posterior aspect of segment IV (n = 1) were selected. TACE of the caudate arterial branch was performed in all patients, including one patient with HCC in the posterior aspect of segment IV who underwent TACE of the caudate arterial branch after CT helped confirm that iodized oil was not distributed in the tumor after TACE of the medial segmental artery. The distribution of iodized oil in the posterior aspect of segment IV was analyzed with CT 1 week after TACE. The number and origin of all arteries supplying the caudate lobe and the number of arteries embolized were determined.Thirty-three caudate arterial branches were embolized. Twenty-nine branches were derived from the right hepatic artery and four were derived from the left hepatic artery. A single branch was seen in 17 patients, two branches were seen in five, and three branches were seen in two. Eight patients simultaneously underwent additional TACE of branches of the right hepatic artery (n = 6) or right inferior phrenic artery (n = 2). At CT, iodized oil was seen to be distributed entirely (n = 19) or partially (n = 5) in the caudate lobe. Distribution of iodized oil at the posterior aspect of segment IV was observed in 16 patients (67%), including 13 (54%) whose caudate arterial branches were derived entirely from the right hepatic artery.The results of this study suggest that the caudate arterial branch, which is mainly derived from the right hepatic artery, frequently supplies the posterior aspect of segment IV. This knowledge is important for managing HCC in the posterior aspect of segment IV by means of TACE.

Proton beam therapy (PBT) achieves good local control for hepatocellular carcinoma (HCC), and toxicity tends to be lower than for photon radiotherapy. Focal liver parenchymal damage in radiotherapy is described as the focal liver reaction (FLR); the threshold doses (TDs) for FLR in the background liver have been analyzed in stereotactic ablative body radiotherapy and brachytherapy. To develop a safer approach for PBT, both TD and liver volume changes are considered clinically important in predicting the extent of damage before treatment, and subsequently in reducing background liver damage. We investigated appearance time, TDs and volume changes regarding FLR after PBT for HCC. Patients who were treated using PBT and were followed up using gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid-enhanced magnetic resonance imaging (Gd-EOB-DTPA MRI) after PBT were enrolled. Sixty-eight lesions in 58 patients were eligible for analysis. MRI was acquired at the end of treatment, and at 1, 2, 3 and 6 months after PBT. We defined the FLR as a clearly depicted hypointense area on the hepatobiliary phase of Gd-EOB-DTPA MRI, and we monitored TDs and volume changes in the FLR area and the residual liver outside of the FLR area. FLR was depicted in all lesions at 3 months after PBT. In FLR expressed as the 2-Gy equivalent dose (α/β = 3 Gy), TDs did not differ significantly (27.0±6.4 CGE [10 fractions [Fr] vs. 30.5±7.3 CGE [20 Fr]). There were also no correlations between the TDs and clinical factors, and no significant differences between Child-Pugh A and B scores. The volume of the FLR area decreased and the residual liver volume increased, particularly during the initial 3 months. This study established the FLR dose for liver with HCC, which might be useful in the prediction of remnant liver volume for PBT.