Angiography of the celiac trunk and superior mesenteric artery via the femoral artery is used to evaluate the hepatic artery vascularization and portal vein patency. HAIC is contraindicated if there is any uncorrectable delivery of anticancer drugs to the gastrointestinal tract or reversed blood flow out of the liver. If necessary, nontarget vessels are embolized to minimize the flow of chemotherapeutic agents into both uninvolved liver parenchyma and extrahepatic tissues. After finding the artery feeding the HCC, the tip of the catheter is inserted into the proper hepatic or common hepatic artery. The proximal end of the catheter is connected to the injection port, which is implanted in a subcutaneous pocket in the right iliac fossa.

In clinical practice, TACE has long been used as palliative therapy for unresectable HCC. However, high-dose hepatic artery infusion chemotherapy has showed favorable outcomes in patients with intractable, advanced HCC. Kim et al. [16] compared TACE and HAIC in advanced HCC and found that the objective response rate was better with HAIC (16.7% vs. 0%; p=0.030). The OS was longer with HAIC (median survival 193 days vs. 119 days; p=0.026) and there were no serious adverse effects in the HAIC group.

Various chemotherapeutic agents have been tried individually or in combination for HAIC, with the most commonly used being the thymidylate synthase inhibitor 5-fluorouracil (5-FU) and cisplatin, a DNA cross-linking compound [17,18]. The mechanism of 5-FU might involve antitumor effects via cell-cycle arrest and the induction of apoptosis. Cisplatin amplified the effect of 5-FU and is regarded as a biochemical modulator of 5-FU [19,20]. Cisplatin not only inhibits DNA synthesis directly but it also interrupts methionine transport into tumor cells, resulting in a reduction in the intracellular methionine level. Simultaneously, cisplatin increases the levels of intracellular reduced folic acid (5,10-CH₂THF, THF), which is necessary for tight binding of FdUMP, a major active metabolite of 5-FU, to its target enzyme, thymidylate synthetase [21,22]. While various combinations of 5-FU and cisplatin have been investigated for use in HAIC [23,24], few studies have assessed these regimens in patients with advanced HCC.

In a recent prospective multicenter study in Korea, Woo et al. [25] compared high-dose (5-FU, 500 mg/m² on days 1 to 3 and cisplatin, 60 mg/m² on day 2) and low-dose (5-FU, 170 mg/m² and cisplatin, 7 mg/m² on days 1 to 5) HAIC. The objective response rate was significantly better in the high-dose group (16.7% vs. 0%; p=0.024). The median time to disease progression (145 days vs. 90 days; p=0.095) and OS (193 days vs. 153 days; p=0.108) were prolonged slightly in the high-dose group. HAIC showed more therapeutic benefit with the high-dose regimen.

In addition, Kim et al. [22] reported the long-term clinical outcome of high-dose HAIC. During a median of 9.5 months, the overall disease control rate was 62.3%. The median progression-free survival (PFS) and OS were 6.0 and 9.5 months, respectively. These results are similar to the reported efficacy of sorafenib.

However, the limitations of the HAIC studies are the small sample sizes and lack of large randomized trials. It will be necessary to compare sorafenib and HAIC in advanced HCC. Randomized trials comparing HAIC and sorafenib are ongoing and might validate the overall clinical advantages of HAIC.

Before a radioembolization procedure, all patients should undergo a ^99mTc-labeled macroaggregated albumin test to see whether the pulmonary shunt is less than 20%. If the pulmonary shunt exceeds 20%, systemic distribution of the radioactive isotope with resulting toxicity is anticipated. Any uncorrectable delivery of ⁹⁰Y to the gastrointestinal tract, reversed blood flow out of the liver, or complete PVT are contraindications of radioembolization, and the presence of these is evaluated together with the shunt study by angiography before treatment [26]. If necessary, nontarget vessels are embolized to minimize the flow of ⁹⁰Y microspheres into both uninvolved liver parenchyma and extrahepatic tissues. Estimation of the radiation dose actually delivered to the tumor tissue is complicated. The basic premise for ⁹⁰Y microsphere dosimetry is to ensure that the liver parenchyma exposure does not exceed 70 Gy, while at least 120 Gy must accumulate within the tumor to deliver a dose-dependent tumoricidal effect [27]. Accurate assessment of the target tumor and liver volumes is critical because they affect the absorbed radiation dose estimates directly. The desired ⁹⁰Y activity is calculated using a partition model or body surface area model [28]. Infusion of the microspheres into one or both lobes can generally be accomplished in a single session, unless the patient has impaired liver function, in which case two procedures separated by 30 to 45 days might be required [29]. Unlike TACE, the response to radioembolization needs to be assessed approximately 3 months postprocedure due to the delayed radiation effect on the tumor [30].

Kulik et al. [31] evaluated a phase II study on the safety and clinical benefit of radioembolization in a larger cohort of patients with unresectable HCC complicated by PVT. The partial response (PR) rate was 42.2%. Stable disease (SD) was seen in 34.7%. The median survival of patients with branch PVT after ⁹⁰Y was 261 days, while that of patients with main PVT was 148 days. In an Italian study, the objective response was 34.3%, whereas the disease control rate was 74.3% [32]. During a median follow-up of 36 months, the median TTP was 11 months with no significant difference between PVT versus no PVT (7 months vs. 13 months). The median OS was 15 months with a non-significant trend in favor of non-PVT versus PVT patients (18 months vs. 13 months). The tumor response was significantly correlated with the absorbed dose in target lesions and a threshold of 500 Gy predicted the response. ⁹⁰Y radioembolization was effective in advanced HCC, particularly in the case of PVT.

In a recent prospective multicenter study in Korea, Kim et al. [33] reported the efficacy and safety of radioembolization in 40 HCC patients. The median follow-up was 29 months. At 3 months, a complete response (CR), PR, and SD were seen in 4 patients (10.0%), 19 patients (47.5%), and 15 patients (37.5%), respectively. At 3 months, the response rate was 57.5% (23/40) and the disease control rate was 95% (38/40), while the respective values at 6 months were 63.9% (23/36) and 83.3% (30/36). The median TTP was 18 months. The most common adverse event was mild to moderate abdominal pain, and all of the complications were manageable.

Approximately 20% of the patients develop liver-related toxicity and 3% suffer treatment-related death [34]. No randomized controlled trial has compared the efficacy of ⁹⁰Y radioembolization with sorafenib in advanced HCC patients, although a randomized trial comparing ⁹⁰Y radioembolization and sorafenib is ongoing.