1 Introduction
Hepatocellular carcinoma (HCC) is the fifth most common malignancy worldwide and is the third most common cause of cancer-related mortalities [
1,
2]. While HCC tumors tend to be slow-growing, they are often rapidly fatal since discovery usually occurs once the disease is advanced [
3]. The major risk factors for HCC are male sex, cirrhosis, hepatitis B virus (HBV) infection, and hepatitis C virus (HCV) infection [
4]. HCC rarely occurs before the age of 40 and reaches a peak at approximately 70 years of age. Cirrhosis is found in 67%–80% of autopsies on HCC patients [
4,
5]. Other risk factors involve alcoholic liver disease, non-alcoholic steatohepatitis (NASH), oral contraceptives, aflatoxins, and microcystin-contaminated water. There are multiple modalities available for management of HCC, depending on number and size of tumors, disease stage, and previous treatments attempted. This review serves to emphasize current and new technologies emerging in the treatment of HCC.
2 Surgical resection
While some studies suggest that the best treatment for HCC is orthotopic liver transplantation (OLT) in that it allows for complete resection of the tumor as well as elimination of cirrhosis [
6], surgical resection (SR) is still considered the gold standard and should be the first consideration in all cases. While hemihepatectomy, segmentectomy and nonanatomical wedge resection are all modalities for tumor resection, only 3% to 30% of patients are resectable at the time of diagnosis. Additionally, there are several factors affecting outcome after SR. Such factors include the size and number of nodules, microscopic or macroscopic vascular invasion, capsular invasion, lymph node involvement, and tumor grade. Although the operative mortality rate for hepatectomy in the 1980s was 10%–20% [
3], recent literature has shown a decrease to less than 5% in patients with early HCC meeting the Milan criteria [
7] along with demonstration of good liver function [
8]. Non-cirrhotic patients can tolerate extensive resections. Well-compensated patients with cirrhosis can tolerate resection of up to two segments; patients with cirrhosis who require resection of more than 2 functional segments should be considered for preoperative portal vein chemoembolization [
3]. The 1-, 3-, and 5-year survival rates of patients undergoing SR are depicted in Table 1 [
9–
15]. However, as seen in Table 2 [
10–
14], the 1-, 3-, and 5-year recurrence-free survival rates of patients undergoing SR remain quite high. Two major factors contribute to high recurrence rates: (1) the actual inciting factor (e.g., HBV or HCV) is not completely removed at the time of SR [
16]; (2) there are no established adjuvant chemotherapy protocols following SR.
3 Orthotopic liver transplantation
Orthotopic liver transplantation (OLT) remains the gold standard for patients with cirrhosis complicated by HCC. The Milan criteria [
7] have traditionally defined eligibility for OLT, in which the presence of a tumor is≤5 cm in diameter in patients with a single tumor or no more than 3 nodules each being≤3 cm in diameter in patients with multiple tumors. While studies have shown that survival rates for OLT tend to be higher in patients meeting the Milan criteria than in patients exceeding them, several studies have also suggested that survival rates in patients exceeding the Milan criteria are acceptable.
More recently, the “survival benefit” concept has become a guide used to evaluate OLT results. For patients with cirrhosis and a Model for End-state Liver Disease (MELD) score<15, the 1-year mortality risk was greater with OLT than without OLT [
17]. Therefore, the United States United Network for Organ Sharing (UNOS) instituted the “Share 15” rule, in which patients with MELD scores≥15 have priority for OLT [
18]. For HCC patients with MELD scores≥15, OLT is therefore a reasonable option given their hepatic decompensation. For patients with HCC and MELD<15, the situation is more complicated. Tandoi
et al. [
19] have demonstrated that the source of the underlying liver disease significantly influences the long-term survival in these patients, with hepatitis C recurrence being the most common cause of death.
There have been an increasing number of studies examining OLT in patients exceeding Milan criteria but meeting UCSF criteria [
20,
21]. UCSF criteria [
21] are defined as the following: a solitary tumor≤6.5 cm, or≤3 nodules with the largest lesion≤4.5 cm, and a total tumor diameter of≤8 cm. Liu
et al. [
20] found that patients meeting Milan criteria undergoing OLT versus patients exceeding Milan criteria but meeting UCSF criteria experienced no difference in survival rates. Secondly, there have been a number of studies investigating the survival rates of OLT in patients who recurred after SR, otherwise known as salvage OLT (SLT), despite exceeding Milan criteria [
22]. Such studies have found no significant differences in survival rates between patients undergoing OLT versus SLT. The 1-, 3-, and 5-year survival rates of these patient populations are depicted in Table 3 [
20,
23–
28]. The 1-, 3-, and 5-year recurrence-free survival rates of these patient populations are depicted in Table 4 [
24,
26,
29,
30]. When compared to Table 2, OLT clearly shows lower recurrence rates when compared to SR.
4 Radiofrequency ablation
SR is often contraindicated in cases where there is limited hepatic function in patients with cirrhosis; hence, ablation has become increasingly common in treatment of HCC. Radiofrequency ablation (RFA) is the primary ablative therapy at most institutions. RF current causes molecular friction locally as a function of current density. Molecular friction adjacent to the electrode causes heat that is transmitted to the surrounding tissue. Temperatures are raised to 85–105°C to enable coagulation necrosis. Once running, RF current is monitored in real time with temperature recording and ultrasound.
RFA when used with intraoperative ultrasound yields better sensitivity for detecting and measuring tumors. While computed tomography (CT) is able to pick up most lesions, intraoperative ultrasound is extremely sensitive for picking up smaller lesions that may be missed on CT. Literature has shown that diagnostic laparoscopy, when used with intraoperative ultrasound, drastically altered the operative management in 32% of patients [
31]. Laparoscopic ultrasound also upstaged the tumor in 32% of cases [
32].
Survival rates after RFA when compared to SR are summarized in Table 1. Feng
et al. [
33] conducted a randomized controlled trial studying the efficacies of RFA versus SR on small HCC<4 cm in diameter with≤2 nodules. They found that the 1-, 2-, and 3-year survival rates for RFA and SR were 93.1%, 83.1%, 67.2% and 96.0%, 87.6%, 74.8%, respectively, with recurrence-free survival rates at 86.2%, 66.6%, 49.6% and 90.6%, 76.7%, 61.1% respectively. There were no differences between the groups in overall survival rate; however, the authors conclude that percutaneous RFA is more likely to be incomplete in treating small HCCs and that SR may still be a better option, as RFA may leave residual tumors more easily due to guidance system limitations. Despite use of ultrasonography with RFA for better results, RFA is limited by its high frequency of local recurrences. Local recurrence rates vary from 2% [
34,
35] to 60% [
36]. Table 2 summarizes recurrence-free survival rates, demonstrating greater recurrence-free survival rates in the SR versus RFA groups. Meta-analysis of SR versus RFA has also demonstrated that SR is superior to RFA for treatment of early-stage HCC [
37].
Additional drawbacks to RFA are the heat sink effect and the tissue charring. Heat adjacent to a great vessel dissipates into the great vessel, thereby preventing the target tissue from reaching an optimal temperature and altering ablation size and shape [
38–
40]. This phenomenon limits the area of ablation that RFA is able to achieve. Charring occurs when heat is rapidly deposited onto tissue, causing boiling and charring of the tissue, which increase tissue impedance and also alter ablation size and shape [
41].
5 Microwave ablation
Microwave ablation (MWA) is the most recent development in thermal destruction of HCCs. MWA employs electromagnetic radiation to excite and thereby heat the intracellular water molecules of the surrounding tissue. This heat causes coagulation necrosis and destruction of the tissue. MWA is extensively used in the Asia and in the UK, whereas its use in the US remains in its infancy [
42]. MWA can be delivered through open laparotomy, laparoscopically, or percutaneously.
MWA has been shown in several clinical trials to be a safe and effective option for unresectable HCCs. Phase I clinical trials in the US demonstrated that MWA can achieve a maximal ablation zone of 50.8 cm
3 with gross and microscopic analysis demonstrating clear coagulation necrosis, even near large hepatic vessels [
43]. Phase II clinical trials in the US [
44] demonstrated that after MWA and a mean follow-up time of 19 months, 41% of patients were alive without evidence of recurrence. Martin
et al. [
45] reported a 100% ablation success rate for the 17 patients in their study with HCC, along with an ablation recurrence rate of 6%. They further report that their median disease-free survival was 18 months, with a median overall survival of 41 months. Another study showed that MWA yielded 1-, 3- and 5-year cumulative survival rates of 93%, 72%, and 51%, with tumor recurrence at 8% [
46]. These survival rates are comparative to those of RFA, and these two modalities should be further compared for efficacy and successful ablation and recurrence rates.
MWA presents with some theoretical advantage over RFA in that microwave instrumentation operates between 900 MHz to 2.45 GHz, whereas RFA ranges from 300 kHz to 1 MHz, thus having lesser thermal destruction capability and at slower rates of destruction. Liu
et al. [
47] studied the effectiveness of the 915 MHz versus the 2.45 GHz models for ablation of HCC greater than 4 cm in diameter. They found that 85.7% of tumors ablated with the 915 MHz models achieved complete ablation, whereas only 73.7% tumors were completely ablated with the 2.45 GHz models. Also, the 915 MHz group required fewer antenna insertions, which they attributed to its ability to penetrate deeper, thereby producing a larger ablation zone. Lastly, in the 915 MHz group, local tumor progression was seen in 14.3% cases, whereas the rate was 26.3% for the 2.45 GHz group. One recent study performed in the US utilized both 915 MHz and 2.45 GHz generators [
48]. In this study, the 915 MHz model is preferred for multiple small volume ablations, while the 2.45 GHz model can target a larger ablation zone over a shorter period of time.
Along with its capacity to create larger zones of ablation compared to RFA, MWA is a good option for perivascular tumor treatment. MWA does not seem to be as susceptible to the heat sink effect that RFA is limited by, as MWA yields a greater power density in the targeted ablation zone [
42]. Thus, MWA is able to achieve greater uniformity in the ablation zone. There has been some discussion about how large a target zone MWA can successfully ablate. RFA and MWA have been shown to be effective in treating HCC measuring less than or equal to 3 cm, but there is minimal literature comparing ablation success in small versus larger HCCs. Liu
et al. [
49] treated 52 patients with a main tumor measuring 3–5 cm and an additional 57 patients with a main tumor measuring 5–8 cm. Of the 52 patients with tumors measuring 3–5 cm, complete ablation was possible in 94.2% of patients. In patients with tumors measuring 5–8 cm, the complete ablation rate was 75%. There was a significant difference in complete tumor ablation rates between these two groups (
P = 0.033). In addition, the local recurrence rate in patients with tumors measuring 5–8 cm was significantly higher than those between 3 and 5 cm (
P = 0.026). The 1-, 2-, 3-, and 5-year overall survival rates for patients presenting with tumors 5–8 cm were 60.7%, 46.4%, 39.3%, and 13.1%. These data are consistent with other literature which recognize that tumor size is an important factor in determining local occurrence [
50] and long-term survival [
51].
6 Drug-eluting beads used in conjunction with transarterial chemoembolization
Transarterial chemoembolization (TACE) is a method in which chemotherapeutic drugs are combined with embolization particles and then injected into the hepatic artery that supplies the tumor. TACE has been used as treatment for HCC in cases where SR is not a viable option or a means of downstaging HCC to fit within Milan criteria for possibility of further management with OLT. TACE has traditionally been done with chemotherapeutics such as doxorubicin or cisplatin emulsed in lipiodol and gelatinin-sponge particles. Two randomized controlled trials have been done evaluating the efficacy of TACE. Table 5 [
52,
53] shows the survival rates for each of these studies. The trials performed by the Hong Kong [
52] and Barcelona [
53] groups showed a significant increase in survival rates when compared to the control groups (
P = 0.005 and
P = 0.025, respectively).
As discussed previously, a common chemotherapeutic agent used with TACE is doxorubicin. A more novel take on TACE is the use of drug-eluting beads (DEB) loaded with doxorubicin. Doxorubicin is loaded onto the beads
in vitro prior to the procedure, and beads varying in size from 100 to 700 μm are injected into the artery feeding the tumor prior to embolization. However, it should be emphasized that there are no major studies demonstrating that systemic doxorubicin is efficacious against HCC. Since the advent of DEB-TACE, there has been question as to whether DEB-TACE is safer or more effective than conventional TACE. Varela
et al. [
54] have demonstrated in their pharmacokinetics study that systemic doxorubicin levels were significantly lower in DEB-TACE patients than in patients treated with conventional TACE, despite having delivered greater doses of doxorubicin via DEB-TACE. Furthermore, Recchia
et al. [
55] have shown that DEB-TACE allows for the gradual and continuous release of doxorubicin over 14 days post-procedure, allowing for prolonged antitumoral effect. This study further showed that AST and LDH levels were significantly lower in the DEB-TACE group versus the conventional TACE group. They also found that hospital stay was significantly shorter and postoperative complications were significantly less in the DEB-TACE group. The PRECISION V study [
56] conducted a randomized trial comparing conventional TACE and DEB-TACE. In their study, the overall response rate, defined by complete response plus partial response, was 51.6% versus 43.5% for the DEB-TACE group versus the conventional TACE group. There were no significant differences between the disease control rates. However, the study did find that in the 67% of patients with Child Pugh B, Eastern Cooperative Oncology Group (ECOG) stage 1, bilobar, or recurrent disease, the incidence of objective response and disease control rates were significantly greater in the DEB-TACE group. The same group found significantly lower serious liver toxicity postchemoembolization in the DEB-TACE group.
There has been some discussion as to whether DEB-TACE can be used to downstage T3N0M0 HCC to meet Milan criteria for transplantation. Patients undergoing locoregional treatments have been shown in prospective studies [
57–
60] to demonstrate acceptable disease-free survival after transplantation and that downstaging HCC with locoregional treatments predicts a favorable OLT result [
61]. Green
et al. [
62] found that 77% of 22 patients undergoing DEB-TACE had their disease downstaged to meet Milan criteria per modified Response Evaluation Criteria In Solid Tumors (mRECIST) measurements, seven eventually undergoing OLT.
7 Transarterial radioembolization
Transarterial radioembolization (TARE) is an emerging treatment of HCC used for palliative measures or to bridge to OLT. Through TARE, micron-sized particles loaded with a radioisotype are injected transarterially. The most commonly studied and used radioisotype is Yttrium-90 infused microspheres [
63].
Yttrium-90 microspheres are 20–40 μm sized particles that emit β radiation [
63]. There are two modalities in which Yttrium-90 microspheres are introduced. TheraSphere® (MDS Nordion, Ottawa, Canada) consists of nonbiodegradable glass beads measuring between 20 μm and 30 μm. It was approved by the US FDA in 1999 through a Humanitarian Device Exemption (HDE) for patients with unresectable HCC and more recently approved for use in patients with HCC and portal vein thrombosis (PVT). SIR-Spheres® (Sirtex, Lane Cove, Australia) consists of biodegradable resin microspheres and was approved by the FDA for metastatic colon cancer to the liver. That being said, most centers in the US using TARE for HCC utilize TheraSphere®, while centers outside of the US have expanded to using SIR-Spheres® as well [
64].
Prior to TARE, pretreatment evaluations must be performed, including angiography and testing for pulmonary shunting. Prophylactic embolization of the gastroduodenal artery and right gastric artery is recommended to minimize risks of hepato-enteric flow, particularly when using resin microspheres [
63]. Shunting of Yttrium-90 microscrospheres to the lungs may cause radiation pneumonitis; hence,
99mTC-MAA particles, which closely mimic Yttrium-90 particles in size, are utilized to predict splanchnic and pulmonary shunting to calculate the lung shunt fraction to minimize the risk of radiation pneumonitis.
TARE has been shown to downstage HCC to meet Milan criteria for OLT eligibility [
65,
66]. Furthermore, a retrospective analysis comparing TACE to TARE has shown that TARE may be a better tool for downstaging disease [
65]. Also, patients presenting with PVT are considered ineligible for TACE because of concerns of interruption of hepatic arterial blood flow resulting in a large segment of hepatic necrosis in patients whose blood supply is already compromised [
67,
68]. As far as time to progression (TTP) of disease, TARE has been shown to be superior to TACE in that TARE leads to longer TTP with less toxicity than TACE and with similar survival times [
69].
Complications with TARE include postradioembolization syndrome, which consists of fatigue, nausea, vomiting, anorexia, fever, abdominal discomfort, and cachexia. Radiation induced liver disease is a serious complication which occurs 4–8 weeks after TARE but usually occurs in only 0–4% of cases [
63]. Liver fibrosis causing portal hypertension may also occur; thus it is recommended that patients are observed routinely for radiologic and clinical evidence of portal hypertension [
70]. The incidence of radiation pneumonitis occurs in less than 1% if standard dosimetry protocols are followed [
71]. Gastrointestinal complications have been reported at less than 5%, so long as proper percutaneous methods are used [
72]. Microspheres may inadvertently spread to the gastrointestinal tract and cause ulceration [
72,
73]. Yttrium-90 induced ulcers usually originate from the serosal surface versus the mucosal surface [
63].
8 Systemic chemotherapy
8.1 Sorafenib
Sorafenib (Nexavar, Bayer/Onyx) is currently the only US FDA approved systemic chemotherapeutic agent to treat advanced HCC. Sorafenib is an oral multikinase inhibitor which blocks tumor cell proliferation by targeting Raf/MEK/ERK signaling at the level of Raf kinase [
74]. It also exerts an antiangiogenic effect by targeting vascular endothelial growth factor (VEGF) receptor-2/-3 and platelet-derived growth factor (PDGF) receptor-β tyrosine kinases [
75] (Fig. 1). The phase III study conducted by Llovet
et al. for the Sorafenib Hepatocellular Carcinoma Randomized Protocol (SHARP) Investigators [
76] demonstrated that the median overall survival in the sorafenib group was significantly greater at 10.7 months versus 7.9 months in the placebo group. They also found that while there was no significant difference symptomatic TTP, the median time to radiologic progression was significantly greater at 5.5 months in the sorafenib group versus only 2.8 months in the placebo group. Llovet
et al. noted similar toxicity profiles with sorafenib compared to the placebo as in the phase II study, with the added exception of hypophosphatemia. The phase III study of sorafenib among the Asia-Pacific population [
77] showed that the sorafenib group experienced a significantly increased medial overall survival at 6.5 months versus 4.2 months for the placebo group. Median TTP was significantly greater at 2.8 months in the sorafenib group versus 1.4 months in the placebo group. Toxicity profiles of sorafenib mirrored those reported in the phase II study. These data imply that despite the greater prevalence of HCV and alcohol cirrhosis contributing to HCC in the SHARP population and a greater prevalence of HBV contributing to HCC in the Asia-Pacific population, sorafenib is still relatively safe and efficacious as a systemic chemotherapeutic agent for both population groups. The discrepancies between the median overall survival rates and the median TTP between the SHARP and Asia-Pacific group have been postulated to be due to the more advanced disease in the Asia-Pacific group [
78]. Fig. 2 depicts changes in tumor size after 15 months of sorafenib.
There have been efforts to combine sorafenib with other modalities used to treat HCC, such as TACE and RFA [
79]. This study included patients in either stage B or C HCC according to Barcelona Clinic Liver Cancer (BCLC) criteria and ECOG≤1 and yielded a disease control rate of 33.3% by RECIST criteria with a median TTP of 15.3 months, median survival rate of 28.8 months. Previous studies have demonstrated that TACE-sorafenib yielded a median TTP of 8.5–10.6 months and a median overall survival of 12–18.5 months [
80–
82]. Li
et al. [
79] postulate that hypoxia induced by local treatment with modalities such as TACE (SPACE trial and ECOG1208) or Resction/RFA (STORM trial) is associated with production of VEGF and is thought to be a potential cause of recurrence. Therefore, the VEGF and PDGF receptor activity that sorafenib inhibits may contribute to the longer median survival rate and median TTP when compared to sorafenib alone. Final analysis of these two adjuvant studies is pending.
STOP-HCC, a phase III, open-label prospective, randomized clinical trial investigating efficacy of sorafenib with or without TheraSphere® for unresectable HCC, is currently underway. The YES-P trial is also currently in phase III and is studying the use of sorafenib versus TheraSphere® for the treatment of advanced HCC with PVT. Concurrently there is the SORAMIC trial in phase II, in which patients will either receive RFA followed by sorafenib or placebo or SIR-Spheres® with sorafenib or sorafenib alone. The Sirtex®-supported SorAfenib versus Radioembolization in Advanced Hepatocellular carcinoma (SARAH) trial is currently in phase III. In addition, the SIRveNIB trial is in phase III and is a multicenter open-label randomized controlled trial comparing sorafenib to SIR-Spheres® in advanced HCC.
8.2 Sunitinib
Sunitinib (Sutent, Pfizer, NY) is a multitargeted tyrosine kinase inhibitor of VEGF receptors and other receptor tyrosine kinases [
83–
89] (Fig. 1). It has demonstrated in three phase II studies antitumoral effects in patients with advanced HCC [
90–
92]. An open-label, phase III study comparing the efficacy of sunitinib versus sorafenib was terminated early due to significantly greater toxicity in the patients treated with sunitinib [
93]. Median overall survival for sunitinib was 7.9 versus 10.2 months for sorafenib. However, median progression-free survival and TTP were comparable.
8.3 Brivanib
Brivanib (BMS-582664, Bristol-Myers Squibb, NY) is a tyrosine kinase inhibitor of the VEGF and fibroblast growth factor (FGF) signaling pathways [
94] (Fig. 1). A phase II study in patients with advanced HCC who had failed previous antiangiogenic treatment demonstrated an overall response rate of 11%, disease control rate of 72% using mRECIST criteria, and a medial overall survival rate of 9.8 months [
95,
96]. Phase III conclusions from the BRISK-FL trial determined that brivanib was not inferior to sorafenib as first-line treatment for HCC, brivanib demonstrating a median overall survival of 9.5 months versus 9.9 months for sorafenib; TTP, objective response rate, disease control rate were similar between the two [
96]. Despite these similarities and while expressing an acceptable adverse rate profile, sorafenib was better tolerated than brivanib, with adverse effects including anorexia, fatigue, hyponatremia, hypertension, nausea, vomiting, headache, dysphonia, and dizziness [
97]. Additionally, phase III studies from the BRISK-PS trial showed that brivanib did not significantly improve the overall survival rate in patients previously treated with sorafenib [
98].
8.4 Ramucirumab
Ramucirumab [IMC-1121B (LY3009806), ImClone Systems Inc.] is a human IgG1 monoclonal antibody that binds with high affinity to VEGFR-2 and blocks the interaction of VEGFR-2 and its ligands [
99] (Fig. 1). In a phase II study of ramucirumab [
99], median progression-free survival was 4.0 months, with median TTP at 4.2 months. Median overall survival was 12.0 months. Patients with BCLC stage C disease and Child-Pugh A cirrhosis were found to have greater median overall survival rates at 18.0 months versus patients at BCLC stage C disease and Child-Pugh B cirrhosis at 4.4 months. Ramucirumab is currently undergoing a phase III trial as a second-line option, post-sorafenib. The study is now closed to enrollment.
8.5 Axitinib
Axitinib (Inlyta, Pfizer) is a tyrosine kinase inhibitor which has been shown to inhibit VEGFR-1, -2, and-3 in
in vitro and preclinical models [
100] (Fig. 1). Currently axitinib is indicated in treatment for advanced renal cell carcinoma after failure of one systemic therapy but is currently undergoing phase II clinical trials in patients with unresectable HCC after having one prior antiangiogenic therapy. This study is a non-randomized, open-label trial with a primary end point of determining safety and efficacy and is now closed to enrollment.
8.6 Tivantinib as a selective MET inhibitor
The hepatocyte growth factor (HGF)/mesenchymal-epithelial transition factor (MET) receptor tyrosine kinase pathway is often dysregulated in HCC [
101] (Fig. 1). Patients may present with high MET versus low MET activity. Overexpression of HGF or MET is correlated with increased metastases, cancer aggressiveness, and poor prognosis [
102]. Tivantinib (ARQ 197) is an oral, selective MET inhibitor which disrupts MET-dependent downstream signaling by blocking MET phosphorylation that is constitutive and HGF-regulated [
103]. In a phase Ib study of tivantinib [
101], 360 mg of tivantinib was administered twice daily to patients with Child-Pugh A and B liver cirrhosis until there was evidence of disease progression or unacceptable toxicity. In their study, the best response was stable disease at 5.3 months with a median TTP of 3.3 months. Further phase II studies [
104] demonstrated that TTP was longer in the tivantinib group at 1.6 months versus 1.4 months in the placebo group. This difference was even more pronounced in MET-high tumors such that TTP in the tivantinib group was 2.7 months versus 1.4 months in the placebo group. In both studies [
101,
104], common adverse effects of tivantinib were neutropenia, anemia, asthenia, leukopenia, anorexia, diarrhea, and fatigue. In the phase II study [
104], four deaths related to severe neutropenia were observed in the tivantinib group. In this study, the 38 patients receiving tivantinib 360 mg twice daily underwent dose reduction to 240 mg twice daily due to a high incidence of grade 3 or worse neutropenia. A population pharmacokinetic analysis of tivantinib, which included this study, revealed that tivantinib exposure is about three times higher in patients with HCC than in patients with other solid tumors; such exposure is correlated with the incidence of grade 3 or worse neutropenia [
104]. The authors of the phase II study [
104] propose that tivantinib could serve as a second-line option for treatment of HCC, particularly patients with MET-high tumors. Currently, there is a phase III trial with tivantinib dosed at 240 mg twice daily versus placebo administered to subjects with MET-high HCC treated with one prior systemic chemotherapeutic agent.
8.7 Bevacizumab plus erlotinib as antiangiogenesis agents
Increased HGF, epithelial growth factor receptor (EGFR), VEGF, insulin-like, and transforming growth factors have been implicated in the progression from normal liver to cirrhosis and HCC [
104]. Bevacizumab (Avastin, Genentech BioOncology, South San Francisco) with erlotinib (Tarceva, OSI Pharmaceuticals, Boulder, CO) have been studied in Phase II clinical trials in the US and in Asia [
105,
106]. Bevacizumab is a monoclonal antibody that binds circulating VEGF-A ligand and is approved by the US FDA for treatment of metastatic colorectal, non-small cell lung, and breast cancers. Erlotinib is an oral tyrosine kinase inhibitor that blocks phosphorylation of the EGFR transmembrane receptor and is approved for advanced adenocarcinoma of the pancreas and non-small cell lung cancer [
105] (Fig. 1). In the US phase II trial [
105], patients with Child-Pugh A or B liver cirrhosis and ECOG status 0, 1, or 2 were enrolled. The primary end point of progression-free survival at 16 weeks was 62.5%. With 40 patients treated, 10 patients experienced partial response, the median progression-free survival was 39 weeks, and the median overall survival was 68 weeks. However, in the Asia Phase II trial [
106], the progression-free survival at 16 weeks was only 35.3%, the median progression-free survival was only 2.9 months, and the median overall survival was only 10.7 months. The variations between these two trials include the dosing of the bevacizumab and the population sampled: the Asia trial used a bevacizumab dose of 5 mg/kg whereas the US trial used a dose of 10 mg/kg, and the Asia trial enrolled more patients with BCLC stage C disease, more patients with HBV infection, and patients of Asian descent only. However, an additional two phase II studies [
107,
108] enrolling patients from the US reported less favorable efficacies than those of earlier phase II trial. Therefore, data are conflicting about the efficacies of bevacizumab with erlotinib. Currently, a randomized phase II study comparing bevacizumab with erlotinib and sorafenib is ongoing in the US [
106].
9 Irreversible electroporation
Irreversible electroporation (IRE) is a novel ablative technique used to treat HCCs in cases where patients either fail or are not candidates for SR or ablative therapies such as RFA or MWA. During IRE, micro- to millisecond electrical pulses are delivered to targeted tissue, causing cell membrane permeabilization and eventual necrosis of the tissue. Davalos
et al. [
109] demonstrated application of IRE can reach temperatures up to 50°C. However, they suggest that its primary ablative technique is through direct damage of the cell membrane rather than thermal destruction. Further animal studies conducted by Lee
et al. [
110] show that IRE tissue, when examined histologically, tested positive for Bcl-2 oncoprotein staining and TUNEL assay, suggesting apoptotic cell death among the treated tissue. Guo
et al. [
111] demonstrated that 15 days post-IRE, 5 out of 6 treated lesions showed no remnant viable tumor, but did show giant cell reaction, hemosiderin-laden histiocyte reaction, and scarring fibrosis. This is consistent with the thought that the irreversibly permeabilized cells left
in situ would be cleared by the immune system [
109]. A preliminary study on the safety of IRE on humans demonstrated complete target tumor ablation of HCC in 82% tumors [
112]. Overall recurrence free survival at 3, 6, and 12 months have been reported in a prospective study at 97.4%, 94.6%, and 59.5%, respectively; however, these values include colorectal metastasis to the liver (45% of tumors), hepatic metastases from non-small cell lung cancer, breast cancer, three carcinoid/neuroendocrine tumors, melanoma, renal cell carcinoma, soft tissue tumor, as well as HCC (35% of tumors) [
113]. Both the preliminary study and the prospective study utilized the NanoKnife® system (Angiodynamics, Latham, NY), which was approved by the US FDA for ablation in soft tissue, including cardiac and smooth muscle. Fig.3 depicts pre- and post-NanoKnife® ablation treatment of a liver tumor. Note the significant early edema zone seen at 24 h which ultimately resolves over time.
IRE also has advantages over RFA in that IRE does not seem to be affected by or damage blood vessels or bile ducts in the targeted area [
111,
113]. Lee
et al. [
110] further comment that the vessels in the ablated zone demonstrated intact endothelium with minimal endothelial damage in small vessels and preserved vessel integrity in large vessels. Furthermore, IRE does not seem to be subject to heat sink effects that RFA is limited by when near vasculature [
109,
113].
Potential adverse effects of IRE include cardiac arrhythmia. Delivery of the pulses must be synchronized to the patient’s electrocardiogram such that the pulses are timed to be delivered during the absolute myocardial refractory period 50 ms after the R-wave; however, the NanoKnife® pulse generator has incorporated this feature into its system [
113]. That being said, cardiac arrythmias can still occur such that in the prospective study discussed earlier, two procedures were aborted before completion of the planned number of ablations due to arrhythmia, ventricular tachycardia in four patients, and transient supraventricular tachycardia in one patient.
10 Summary
Surgical resection and OLT are still the gold standards in managing hepatocellular carcinoma. However, most patients are not eligible for these treatement modalities. While there are several thermal ablative techniques and irreversible electroporation that may serve as a palliative means of care or as a means of downstaging tumors to fit criteria for OLT, such techniques themselves have relatively high survival rates on their own. Sorafenib has shown promise as a systemic chemotherapeutic agent, and studies are investigating its efficacy with other chemotherapeutics.
11 Compliance with ethics guidelines
Marielle Reataza and David K. Imagawa declare that they have no conflict of interest. This manuscript is a review article and does not involve a research protocol requiring approval by the relevant institutional review board or ethics committee.
Higher Education Press and Springer-Verlag Berlin Heidelberg