Although the current standard treatment for hepatocellular carcinoma (HCC) with portal vein tumor thrombosis (PVTT) is sorafenib, many previous studies have established the need for a reliable local modality for PVTT control, which is a major cause of liver function deterioration and metastasis. Additionally, there is growing evidence for the prognostic significance of PVTT classification according to the location of tumor thrombosis. Favorable outcomes can be obtained by applying local modalities, including surgery or transarterial chemoembolization, especially in second-order or distal branch PVTT. Rapid control of PVTT could maintain or improve liver function and reduce intrahepatic as well as distant metastasis. Radiotherapy (RT) is one of the main locoregional treatment modalities in oncologic fields, but has rarely been used in HCC because of concerns regarding hepatic toxicity. However, with the development of advanced techniques, RT has been increasingly applied in HCC management. Randomized studies have yet to definitively prove the benefit of RT, but several comparative studies have justified the application of RT in HCC. The value of RT is especially noticeable in HCC with PVTT; several prospective and retrospective studies have reported favorable outcomes, including a 40% to 60% objective response rate and median overall survival of 15 mo to 20 mo in responders. In this review, we evaluate the role of RT as an alternative local modality in HCC with PVTT.

Hepatocellular carcinoma (HCC) is the fifth most common cancer and second leading cause of cancer mortality worldwide[1,2]. Despite remarkable improvements in the management of HCC, including antiviral agents[3-5], adaptation of magnetic resonance imaging[6] and major advances in local treatment modalities[7-9], this disease remains a concerning health issue. The major issue in HCC management is that approximately 70% of newly diagnosed patients are not candidates for curative local treatment modalities[10]; in particular, portal vein tumor thrombosis (PVTT), which is a common complication of HCC, is frequently an obstacle to the application of such modalities[11].

PVTT is known to be present in approximately 30% to 40% of cases of advanced HCC[10-12]. The association between PVTT and dismal outcomes has been confirmed by many studies[13,14]. Because of the role of PVTT as the most powerful prognosticator in HCC, curative local modalities are not recommended for HCC with PVTT in the Barcelona Clinic Liver Cancer (BCLC) staging and treatment system[15,16]. Nevertheless, local treatment modalities including surgical resection, trans-arterial chemoembolization (TACE), and radiotherapy (RT) are commonly applied in HCC with PVTT to obtain early and durable local control[17,18].

RT is a key locoregional modality in oncology[19]. It is minimally invasive and, unlike surgery, is not associated with pain or a long recovery period. In cancer patients with a relatively high risk of recurrence at other sites, as in those with oligo-metastatic disease, RT is generally preferred for obtaining local control over invasive surgical resection in order to avoid missing the optimal timing for systemic management[20]. However, RT has played a very limited role in past management of HCC, and is usually administered for palliation of symptoms related to extrahepatic metastasis, rather than for control of the primary liver tumor[21]. The source of this limitation is the radiosensitive nature of the cirrhotic non-tumorous liver and related concern regarding the maintenance of clinical liver function, which is a main aim in HCC management[22].

With the development of improved RT techniques and radiobiology, RT has recently been increasingly applied in HCC management. Although there is no definitive evidence for the role of RT, superior outcomes obtained by including RT in HCC management have been reported in several meta-analyses as well as many prospective and retrospective studies[23-35]. Based on these favorable clinical outcomes of innovative RT techniques, liver-directed RT has been recommended for HCC patients who are not candidates for curative loco-regional modalities in the guidelines from the Korean Liver Cancer Study Group[36] and the National Comprehensive Cancer Network[37]. However, even though the use of RT in unresectable HCC is gradually increasing worldwide as a result of these recommendations[38], RT is still not recommended by the BCLC system, which is the most widely used set of HCC management guidelines[16].

In this review, we introduce the current status of RT in HCC and recently reported outcomes and discuss why the addition of RT is beneficial in HCC patients with PVTT.

PVTT is present at diagnosis in approximately 30% to 40% of patients with advanced HCC[11,12], and is one of the most important poor prognostic factors of HCC, with a reported median overall survival (OS) of 2.7 mo to 4.0 mo[12,39].

There are several explanations for how PVTT results in dismal clinical outcomes. PVTT itself disturbs the portal blood supply in the normal liver and can therefore cause deterioration of liver function crucial to HCC management[40]. Furthermore, local tumor progression can accelerate the deterioration of liver function. In addition, PVTT might act as a distributor of intrahepatic and extrahepatic metastasis[41,42]. These possible unfavorable roles of PVTT were observed in our previous study evaluating the relationship between PVTT response and clinical outcomes in HCC patients with PVTT who were treated with TACE followed by RT[43]. Elevation of Child-Pugh scores representing liver function deterioration was significantly more frequent in the non-responders than the responders of treatment. Moreover, the rates of intrahepatic and extrahepatic metastasis were significantly lower in responders.

Because of its well-known role as the most powerful unfavorable prognosticator in HCC, the BCLC system does not recommend curative local modalities for patients with HCC with PVTT[16]. Based on its survival benefit in two large prospective randomized phase III trials[44,45], sorafenib, an oral multikinase inhibitor of platelet-derived growth factor receptor, vascular endothelial growth factor receptor and Raf, is the only standard treatment in these patients. However, physicians hesitate to use sorafenib in advanced but localized disease because of the disappointing objective response rate of 2% to 5% and extremely short median time to progression of 2.8 mo found in a study conducted in the Asia-Pacific region[46]. A more reliable local modality that can rapidly control vascular invasion to the portal vein and thus maintain liver function while eliminating the source of metastasis is needed to improve clinical outcomes.

Furthermore, there is a wide spectrum of vascular invasion clinical manifestations of PVTT, and prognoses are clearly separated according to the extent of PVTT[47-50]. PVTT can be classified into four categories as suggested by the Liver Cancer Study Group of Japan[51] (Figure 1): the presence of tumor thrombus distal to the second-order branches of the portal vein is defined as Vp1, in the second-order branches of the portal vein as Vp2, in the first-order branches of the portal vein as Vp3, and in the main trunk or both first-order branches of the portal vein as Vp4. Although definitive evidence is lacking, customized treatments for HCC patients with PVTT according to PVTT classification combined with other recognized factors, such as liver function, performance status, tumor size and number, should be explored.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Classification of portal vein tumor thrombosis. PVTT is classified into four categories according to the site of tumor thrombus as suggested by the Liver Cancer Study Group of Japan. PVTT: Portal vein tumor thrombosis.

The importance of local treatment modalities in the management of HCC cannot be overstated. Immediate methods of tumor elimination, such as surgical resection including liver transplantation (LT) and radiofrequency ablation (RFA), can prevent reductions in liver function by preventing tumor encroachment into the surrounding liver and removing the source of intra and/or extrahepatic metastasis[52-55]. Long-term survival is achieved in approximately 60% to 70% of HCC patients treated with curative local modalities, but recurrence eventually occurs in more than half of patients, with the exception of those who undergo LT. However, the majority of patients are not candidates for curative local modalities, mainly due to tumor burden, and PVTT is a major obstacle to the application of curative local modalities[10-12,39].

Notable survival rates of 20% to 40% 5 years after surgical resection were reported by the Liver Cancer Study Group of Japan, mainly in HCC patients with minor class Vp1 or Vp2PVTT[56,57]. Although selection bias is inherent in those studies, the results indicate that a proportion of patients might benefit from early elimination of the tumor and PVTT. Furthermore, some cases of long-term survival were reported after surgical resection of HCC with Vp3 or Vp4 PVTT[58,59].

As an alternative local modality with proven efficacy, TACE is widely used in cases of large multinodular HCC[60-62]. However, a complete response is rarely obtained and the duration of local control is unsatisfactory, especially for large tumors[59]. Furthermore, because of the rationale for TACE, which is selective tumor embolization through the arterial blood supply of HCC, in contrast to the dual supply of the surrounding normal liver, TACE had previously been contraindicated in HCC with PVTT. Recent studies, however, have supported that TACE can be safely administered in these patients because of collateral circulation[63,64]. In clinical practice, the super-selective method of TACE, with or without another modality, is one of the most frequently used treatments in HCC patients with PVTT[48,65].

Surgical outcomes are generally reported to be superior to TACE in HCC with PVTT. Liu et al[66] reported superior outcomes of surgical resection compared to TACE in 108 pairs of matched patients selected using propensity score analysis. It is very hard to directly compare clinical outcomes between surgical resection and TACE because of the clear differences in patient and tumor characteristics, including liver function, performance status, and the extent of disease. Nevertheless, local control could enhance clinical outcomes in HCC with PVTT.

Historically, RT was not used in HCC management, especially for primary lesion of the liver. The main reason for the avoidance of RT in HCC was the low tolerance of liver tissue to RT exposure and concerns regarding radiation-induced liver disease (RILD), which typically presents as anicteric hepatomegaly and ascites 2-3 mo after RT[22]. The pathology of radiation-induced liver damage shows changes similar to veno-occlusive disease, including endothelial swelling, terminal venule occlusion and sinusoidal congestion[67]. RILD is a major concern in liver RT, because there is no established treatment and some patients will die of liver failure, although a few patients may recover[68]. Recently, the incidence of classic RILD has decreased, but non-classic RILD (which appears with jaundice and marked elevation of serum transaminase) is still problematic in the application of RT in HCC[69].

Because the treatment options for RILD are very limited, most investigations have focused on identifying predictive factors of RILD development[69-74]. Factors related to RT planning, such as RT dose and liver volume, have been of particular interest. Mean liver dose was demonstrated by Dawson et al[71] to be a risk factor of RILD, and V_{20 Gy} was suggested by Liang et al[74] to be a dosimetric predictor. Kim et al[72] reported that grade 2 or higher radiation-induced hepatic toxicity was significantly related to the percentage of total normal liver volume exceeding 30 Gy irradiation. Our group reported that V_{30 Gy} was a significant risk factor of Child-Pugh score elevation of two points or greater[69]. There is also a possibility of a higher incidence of liver function deterioration after RT in HCC combined PVTT than in the other, but the dosimetric factors of liver are more important, generally[69,72].

With better information regarding RILD development and the application of computed tomography (CT) to RT planning, higher RT doses for local tumor control can be administered with an acceptable RILD risk level[73,75-77]. Further innovative RT technologies, including three-dimensional conformal RT (3D-CRT)[78], intensity-modulated RT (IMRT)[79], stereotactic body ablative RT (SABR)[80] and particle beam therapies[81], have been introduced and actively applied in HCC management (Figure 2). Additionally, image-guided RT[82], which is used as a supportive/supplementary method with other precision RT techniques, is also standard RT practice. These biological and technical developments have enabled delivery of higher-dose RT with improved precision and better conformation and without an increased probability of normal tissue complications. With these rapid developments in techniques and radiobiology, RT has become an accepted tool in the management of HCC[77].

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Radiation dose distribution according to radiotherapy technique. More conformal dose delivery to the main mass and tumor thrombosis with a reduced liver dose is achievable with 3D-CRT, IMRT, and SABR, as well as to an even greater extent with proton beam RT. 3D-CRT: Three-dimensional conformal RT; IMRT: Intensity-modulated RT; SABR: Stereotactic body ablative RT. RT: Radiotherapy technique.

Although evidence from well-designed randomized trials on the role of RT is still needed, several comparative studies have justified the application of RT in HCC management. The results of three systematic meta-analyses of randomized and non-randomized trials are displayed in Table 1[24,25,30]. In all three reports, the addition of RT to TACE showed a significant survival benefit over TACE alone, with an OR of 1.91 to 2.75 for 3-year OS rates. The complete response (CR) rate was also significantly higher in the TACE plus RT group with an OR of 2.58 to 2.73, and a similar effect was observed in the PVTT subgroup (OR = 2.38)[30].

Several comparative reports presented the efficacy of RT compared to other treatment modalities in HCC, and a summary of these studies is displayed in Table 2[23,26-28,31,33-35,83]. In the retrospective cohort study of Eun et al[83], RT showed a clear survival benefit in advanced HCC over best supportive care (median OS 45.9 mo vs 4.8 mo, P < 0.001). Yoon et al[35] compared concurrent chemoradiotherapy (CCRT) and other treatment modalities, including surgery, RFA and TACE, in locally advanced HCC using a propensity score matching technique and showed that CCRT yielded significantly longer-term OS than other modalities (median OS 11.4 mo vs 6.6 mo, P = 0.02).

Two studies compared RT with or without TACE to sorafenib as the current standard treatment in patients with BCLC stage C disease including PVTT, and both showed a positive effect of RT on survival[27,28]. Moreover, TACE with RT also led to a significant survival advantage in HCC with PVTT in a comparative study using propensity score matching[33]. Cho et al[27] reported that OS was significantly higher in the RT group in a propensity score-matched cohort (median OS 8.9 mo vs 3.1 mo, P < 0.001), as well as in all cohorts (median OS 14.1 mo vs 3.3 mo, P < 0.001). Tang et al[26] also compared surgical resection to RT in 371 patients with resectable HCC with PVTT and reported a median 2.3-mo survival advantage with RT when compared to resection (P = 0.03), with comparable progression-free survival.

The popularity of SABR, also known as stereotactic body RT, has recently increased in RT for HCC[29,31,32,34,84]. Two studies compared SABR and RFA, which is one of standard local modalities, and both showed that local control with SABR was comparable or even higher for larger tumors (≥ 2 cm) with similar OS in early HCC[31,34]. Although differences in characteristics originating from the retrospective design were compensated for in these studies, the results should be interpreted with caution.

As mentioned above, the wide variety of clinical outcomes of PVTT reflects not only the extent of PVTT, but also other factors[13,14,39,47-50]. Previously, our group reported a prognostic model in HCC with PVTT treated with RT that clearly stratified patient survival outcomes based on other validated prognostic models of HCC[93]. In addition, our prognostic model was validated by another independent data set from a multicenter cohort[115].

A comparative study evaluating surgical resection vs conformal RT combined with TACE for resectable HCC with PVTT including more than 50% Vp3 or Vp4 cases showed a significant survival advantage in the TACE and RT group vs the surgical group (median 12.3 mo vs 10.0 mo, P = 0.03)[26]. In our unpublished analysis comparing surgical resection vs TACE followed by RT in HCC with Vp1/2 PVTT, the OS of patients that underwent surgical resection was slightly higher in cases with small tumor size (< 3 cm), solitary tumor and low α-fetoprotein level (< 200 ng/mL), although statistical significance was not reached.

The detrimental role of PVTT as an obstacle to liver function maintenance and a source of metastasis indicates that valid local modalities based on PVTT extent, clinical liver function and tumor extent are urgently needed to improve clinical outcomes for HCC. Suggested local modalities for control of HCC with PVTT according to PVTT classification and/or disease extent are displayed in Figure 3, although prospective studies are still needed on the role of these modalities in the treatment of HCC. Large, prospective, randomized, controlled studies evaluating local treatment modalities and survival outcomes compared to the current standard treatment of sorafenib should be performed.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Suggested local treatment modalities for hepatocellular carcinoma with portal vein tumor thrombosis according to portal vein tumor thrombosis classification. To rapidly eliminate tumor thrombosis, RT with or without TACE could be considered, except in the case of technically resectable Vp1/2 disease with a favorable prognosis. HCC: Hepatocellular carcinoma; PVTT: Portal vein tumor thrombosis; RT: Radiotherapy; TACE: Trans-arterial chemoembolization.

Although RT has shown promising results including favorable treatment response and survival rates in HCC with PVTT, treatment decisions regarding RT should be made cautiously because of several remaining limitations[116].

Liver function is one of the most important factors determining the method and purpose of HCC management[16], and is a key concern when considering RT. In patients with a Child-Pugh score ≥ 8, liver-directed RT is not generally recommended. Thus, there have been few studies in which RT was applied in those patients[36,37]. Culleton et al[117] reported a prospective study of SABR outcomes in patients with Child-Pugh class B or C HCC, and significantly lower survival was detected in patients with a Child-Pugh score ≥ 8 (9.9 mo in those with scores of 7 vs 2.8 mo with scores of 8 or higher, P = 0.01). In another prospective phase I study of SABR for HCC, 3 of 11 patients with Child-Pugh class B disease developed grade III hepatic toxicity, while none of the 17 patients with Child-Pugh class A disease did[98]. A significantly higher incidence of grade II or higher liver toxicity in patients with Child-Pugh class B disease was reaffirmed by a large retrospective SABR study (36.0% vs 11.9% of Child-Pugh class A patients)[118].

Another important obstacle to RT application in HCC is the radiation susceptibility of the bowel, including the stomach and duodenum. The positive correlation between the incidence of symptomatic bowel toxicity and RT dose/bowel volume has been confirmed in several studies[119-121]. Our group also reported that liver function deterioration is related to a higher incidence of symptomatic bowel toxicity[122].

Before applying RT in HCC, these unresolved obstacles need to be considered. Although promising outcomes were achieved in HCC patients with poor liver function (Child-Pugh class C) or adjacent to the bowel in small studies using particle beam RT[123-125], in these high-risk patients the use of RT should be carefully restricted or limited to prospective clinical trials.

While acceptable local control using RT in HCC has been reported, especially with SABR, frequent intrahepatic recurrence remains an unresolved issue[88,126-128]. Several studies have suggested that RT may induce intrahepatic metastasis via viral reactivation[129,130] and/or expression of vascular endothelial growth factor[131]. With the development and optimal application of antiviral agents and targeted agents like sorafenib, intrahepatic recurrence might be minimized after RT[3-5,131,132]. Further studies on combination treatment with RT are needed.

Considering the role of PVTT in liver function deterioration and metastasis, a reliable local modality for early control of PVTT is urgently needed. Because there is considerable variation in patient prognosis, customized application of local modalities including surgical resection and TACE should be based on the extent of the main mass and/or PVTT. With recent technological advances, RT could be an effective local modality for HCC control, especially for patients with HCC combined with PVTT.