Resection leads to cure in selected patients with hepatocellular

Overview: Where Does Radiation Therapy Fit in the Spectrum of Liver Cancer Local-Regional Therapies? Laura A. Dawson, MD*, Experience with radiation therapy for the treatment of hepatocellular carcinoma (HCC) and liver metastases has increased rapidly in the past decade. This is principally because of advances in imaging and radiation techniques that can conform high doses to focal cancers and to a better understanding of how to avoid radiation-induced liver toxicity. Guidelines on how to use radiation therapy safely are becoming more clearly established, and reports of tumor control at 2 to 5 years show the potential for cure after radiation therapy for early-stage HCC and liver metastases. For both HCC and liver metastases, the best outcomes after radiation therapy are found in patients with fewer than 3 lesions that are <6 cm in size, with intact liver function and no extrahepatic metastases. There is a strong rationale for using radiation therapy in patients unsuitable for or with expected poor outcomes after standard local-regional therapies. These patients tend to have advanced tumors (large, multifocal, or invading vessels) and/or impaired liver function, reducing the chance of cure and increasing the chance of toxicity. In these patients, the benefits of radiation therapy over systemic therapy or best supportive therapy should be established in randomized trials. Semin Radiat Oncol 21:241-246 2011 Elsevier Inc. All rights reserved. Resection leads to cure in selected patients with hepatocellular carcinoma (HCC) or liver metastases. Unfortunately, the minority of patients are suitable for resection, either because of medical contraindications, excessive burden of hepatic disease, or insufficient liver functional reserve. Nonsurgical local and hepatic directed therapies have a role to play in these patients, but these interventions have limitations, and there is a need for improved therapies. Radiation therapy has a well-defined role in many malignancies in which organ (functional) preservation is desirable (eg, head and neck cancer, esophageal cancer, anal canal cancer, and breast cancer). Given that hepatic function is required to sustain life, it is surprising that radiation therapy does not yet have an established role in hepatic malignancies. This is partially because of the incorrect perception of inevitable liver toxicity after radiation therapy, technical challenges in delivering focused radiation therapy to the liver, *Radiation Medicine Program, Princess Margaret Hospital, University Health Network, Toronto, Ontario, Canada. Department of Radiation Oncology, University of Toronto, Toronto, Ontario, Canada. Address reprint requests to Laura A. Dawson, MD, Radiation Medicine Program, Princess Margaret Hospital, University Health Network, 610 University Ave, Toronto, Ontario, Canada M5G 2M9. E-mail: Laura. dawson@rmp.uhn.on.ca and a general lack of training in hepatobiliary malignancies in most radiation oncology residency programs. The tolerance of the liver to radiation therapy is better understood now than 20 years ago, at least for classic radiation-induced liver disease as described in the article by Kavanagh et al in this issue (see pages 256-263). Advances in imaging, radiation planning, motion management, and image guidance at the time of radiation delivery make safe delivery of ablative doses of radiation therapy to focal liver tumors possible, as described in the article by Brock in this issue (see pages 247-255). However, despite these advances, the doses that can be delivered to liver tumors, especially those unsuitable for other local-regional therapies, are often limited by the tolerance of the liver or adjacent luminal gastrointestinal tissues. A variety of radiation strategies have been developed with the goal of escalating the dose to liver tumors while maintaining low doses to normal tissues, especially the hepatic parenchyma. All radiation therapy strategies have been shown to be most safe and effective in patients with intact liver function and early-stage liver cancers. The potential for radiation sensitizers to improve the therapeutic ratio is described in the article by Seong et al in this issue (see pages 303-311). In this overview, the potential of radiation therapy to improve outcomes in patients with HCC and liver metastases, relative to the already large spectrum of local-regional therapies available, is discussed. 1053-4296/11/$-see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.semradonc.2011.05.009 241

242 L.A. Dawson HCC HCC is the third most common cause of cancer death globally, with an estimated 564,000 new cases annually. 1 Although HCC is less common in North America, the incidence has increased from 1.4 to 2.4 per 100,000 over the past 2 decades, and it is expected to continue to rise in parallel with the increasing incidence of hepatitis C. 2 Liver and biliary cancers are the most rapidly increasing cause of cancer death in the United States, with a 30% increase in death rate from 1991 to 2006. 3 Cirrhosis from alcohol, viral hepatitis, autoimmune hepatitis, hemochromatosis, or nonalcoholic steatohepatitis (NASH) increases the risk of HCC developing. Patients with hepatitis C cirrhosis have a 5% to 20% 5-year cumulative incidence of HCC, and even in the absence of cirrhosis, hepatitis B infection is associated with a 15% risk of HCC. Many patients with cirrhosis have impaired liver function, and the degree of impairment impacts HCC prognosis and treatment options. The most commonly used measure of liver function is the Child-Pugh classification, which is based on the presence or absence of ascites and encephalopathy as well as bilirubin, albumin, and international normalized ratio (INR) levels. Survival is best in Child-Pugh class A and worst in Child-Pugh class C even in the absence of HCC. Given the importance of liver function as a prognostic factor and for treatment decision making, liver function is included in most staging and classification systems for HCC, such as the widely used Barcelona Clinic Liver Cancer (BCLC) (Fig. 1). 4,5 A weakness of the BCLC system is that BCLC advanced stage (C) includes patients with vascular invasion, lymph node involvement, or metastases. Prognosis varies considerably in these patients, and local therapies may be an option in focal HCC with vascular invasion but not in patients with diffuse extrahepatic metastases. The great majority of HCC patients die of tumor progression within the liver. Improved hepatic-directed therapies, including radiation therapy, have the potential to improve hepatic control, overall survival, and quality of life in HCC patients. Outcomes After Standard Therapies in HCC Cure from a variety of therapies is most likely in early-stage HCC, comprising 30% of the total HCC population. 4 Another 20% of HCC patients present in terminal stage in which palliative treatment is recommended. The remaining 50% of Where RT fits Unsuitable for resection, transplant or RF Definitive RT RT as bridge to transplant Unsuitable/refractory to TACE Definitive RT Symptomatic Low dose RT Portal invasion Definitive RT & sorafenib Randomized trials needed to demonstrate benefit Figure 1 BCLC classification for HCC including estimated incidence and accepted therapeutic options. Potential roles for RT are shown in the blue boxes in the bottom rows. An opportunity for randomized trials of radiation therapy is indicated by diamonds. For portal invasion, the optimal timing of radiation therapy and sorafenib is unknown, and radiation therapy and concurrent sorafenib is only recommended on a clinical trial. PST, performance status based on Eastern Cooperative Oncology Group score; N, nodal stage; M, metastases stage; CLD, cadaver liver transplantation; LDLT, living donor liver transplantation; RF, radiofrequency ablation; PEI, percutaneous ethanol injection; TACE, transarterial chemoembolization. (Modified with permission. 5 )

Radiation therapy and liver cancer 243 patients have incurable locally advanced disease or metastatic disease, possibly suitable for regional or systemic therapies, as shown in Figure 1. Liver transplantation is argued to be the ultimate therapy for HCC, especially in the setting of severe cirrhosis (Child-Pugh B or C) because both the cancer and the underlying liver dysfunction are simultaneously treated. The most widely accepted selection criteria for transplantation are the Milan criteria defined as a single tumor 5 cm or less or up to 3 tumors 3 cm or less with no extrahepatic spread or macrovascular involvement. When such criteria are followed, transplantation is associated with a 5-year overall survival of approximately 70%, and the recurrence rate is 15%. Unfortunately, there is a substantial wait time for transplantation because of a limited availability of donors, so many patients drop off the waitlist because of progression of HCC beyond the Milan criteria. Bridging therapies for patients on the waitlist may increase the percentage of patients who are transplanted. For patients with a solitary HCC without vascular invasion, with Child-Pugh A liver function, and no portal hypertension, partial liver resection is a treatment option. Operative mortality is low (1%-3%), and 5-year survival rates are approximately 50%. Mortality risks are higher (3%-13%) in the presence of cirrhosis. Unfortunately, recurrences and/or new foci of HCC occur in up to 70% of patients after resection. For patients with unresectable HCC, radiofrequency ablation (RFA) or other ablative approaches (ie, cryotherapy and microwave) are associated with a local control rate of 90% or higher in lesions 3 cm, away from large vessels. Larger lesions are more likely to recur after RFA. In a randomized trial of 180 patients with a solitary HCC 5 cm, survival was the same in patients treated with RFA versus resection. 13 Lesions near the dome of the liver are difficult to target with RFA effectively, and lesions near large vessels are not adequately heated because of heat sink from the adjacent vasculature; alternative local therapies would be beneficial in these situations. For patients with large or multifocal tumors, regional and systemic therapies are available. Hepatic tumors derive 80% of their blood supply from the hepatic artery, whereas the adjacent liver parenchyma is supplied by the portal vein. Thus, transarterial chemoembolization (TACE) or radioembolization, as described by Memon et al in this issue (see pages 294-302), can target HCC with specificity via hepatic arterial delivery. TACE has been shown in randomized trials to improve survival duration compared with symptomatic therapy alone in patients without macrovascular involvement. 14,15 In tumors larger than 10 cm, morbidity of TACE is increased and efficacy is reduced. 16 Sorafenib, an oral multikinase inhibitor, improves survival in HCC patients unsuitable for any of the previously described therapies. In the SHARP (Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol) trial, median survival with sorafenib was improved from 7.9 to 10.7 months (P.001), 17 whereas in the Asia-Pacific trial, median survival was improved from 4.2 to 6.5 months (P.014). 18 Outcomes after standard therapies in HCC are summarized in Table 1. Overall, there is a strong need for more effective safe therapies. Table 1 Outcome Estimates for HCC (Child-Pugh A) Survival, 5 Years (%) Liver Metastases Hepatic Recurrence (%) Transplant 70-75 8 Resection 45-60 68 RFA 50 5-20 (local recurrence) TACE <2 (31-63 at, 2 y) 100 sorafenib 0 100 The liver is a frequent site of metastases for a variety of malignancies, including colorectal carcinoma, breast carcinoma, melanoma, and neuroendocrine tumors. A substantial proportion of these patients die of hepatic confined metastases. Although most patients with liver metastases have occult diffuse micrometastases making them incurable despite any liver-directed therapy, a proportion have oligometastases, meaning that the radiographically detected metastases are the only site of disease and that local control of the metastases may improve overall survival. 6 Although the concept of oligo metastases has not changed, the ability to better identify patients with true oligo metastases is increasing as staging investigations improve. Approximately 50% of patients with colorectal carcinoma develop liver metastases over the course of their lifetime. Five-year survival for metastatic colorectal carcinoma has increased over the past few decades, caused largely by advances in systemic therapy 3 and the wider use of resection or ablation of liver metastases. With chemotherapy alone, the 5-year survival for patients with unresectable colorectal liver metastases is 2%. Evidence for the use of local therapy for liver metastases is strongest in colorectal carcinoma, in which 5-year survival rates range from 33% to 58% after the resection of liver metastases. 7-9 For noncolorectal liver metastases, there is a less clear role of resection although long-term survivors have been reported after the resection of liver metastases from breast cancer, 10 sarcoma, 11 cutaneous or uveal melanoma, 12 and gynecologic malignancies. Patients with liver-confined metastases from colorectal carcinoma and other selected tumor sites, who have medical contraindications to resection or unresectable metastases, may be suitable for nonsurgical ablative therapies, such as radiation therapy, that may improve outcomes over systemic therapy alone. Outcomes After Standard Therapies in Liver Metastases Outcomes after resection for liver metastases from colorectal carcinoma have improved over time, from 30% to 40% to 50% to 58% for 5-year survival in more recent series. 7-9 In parallel, the patient selection for liver resection is expanding. Fong et al 7 reported the following adverse prognostic factors after liver resection for colorectal carcinoma metastases: the number of tumors removed ( 1), the size of the tumors removed ( 5 cm), the preoperative carcinoembryonic antigen level (greater or less than 200 ng/ml), the extent of

244 L.A. Dawson Table 2 Outcome Estimates for Colorectal Liver Metastases 5-Year Survival (%) Local Recurrence (%) Hepatic Recurrence (%) Mortality (%) Resection 33-58 1 50 1-3 RFA 17-55 6-39 12-57 <1 Systemic chemotherapy 2 100 100 0.5-1 resection (less than or greater than a lobectomy), the resection margin (negative or positive), and the presence of extrahepatic disease. 7 Now, the decision to resect is not limited by the size or number of metastases but primarily by the ability to preserve a minimum functional liver remnant. Chemotherapy can be used in an effort to convert unresectable metastases to resectable, and outcomes in patients treated with surgery are better than historic controls treated with chemotherapy alone. Because liver regeneration occurs after resection, repeat or staged resections are possible in colorectal carcinoma liver metastases. Post-resection morbidity and mortality rates are lower in metastases compared with HCC, ranging from 20% to 45% and 1% to 3%, respectively. 7,8 Resection is sometimes used together with RFA in multilobar metastases unsuitable for extensive resections. RFA on its own is an option for small unresectable liver metastases distal to large vessels. Liver recurrence rates tend to be higher after RFA compared with resection, ranging from 12% to 57%. 19 Five-year survival after RFA ranges from 17% to 55% (Table 2). 8 With the exception of surgery for colorectal liver metastases, a challenge for the oncology community is to try to prove the benefits of local therapy for oligometastases over systemic therapy alone. Bias exists in single-institution and retrospective series showing the odd long-term survivors, so prospective trials are warranted. Types of Radiation Therapy A variety of techniques may be used to deliver radiation therapy to liver cancers, with treatment goals ranging from palliation to cure. Details of palliative, conformal, and stereotactic body radiation therapy for the treatment of liver metastases and HCC are described by Schefter et al (see pages 264-270) and Feng et al (see pages 271-277), respectively, in this issue. Also in this issue of Seminars, Ricke et al (see pages 287-293) describe interstitial brachytherapy, and Memon et al review radioembolization for HCC and liver metastases. The best single-center outcomes seen in HCC patients treated with radiation therapy are in patients treated with charged particle therapy, as described by Krishnan et al (in this issue [see pages 278-286]). The 5-year survival rates in 162 HCC patients with impaired liver function, treated with 72-Gy cobalt equivalent in 16 fractions, was 24%; for 50 patients with Child-Pugh A and isolated HCCs, 5-year survival was 54%, which is similar to surgical series. 20 In both liver metastases and HCC, local control of lesions 3 cm is very high (80%-100% at 1-2 years) after conformal or stereotactic radiation therapy, charged particle therapy, or percutaneous brachytherapy. Some of these patients have no evidence of disease more than 5 years after therapy, showing that cure may be possible. These are the minority of patients treated, and long-term survival is uncommon in most radiation therapy series, largely because of the inclusion of patients with more advanced disease and poor prognostic factors, increasing the probability of extrahepatic progression of disease. Comparisons of outcomes of radiation techniques to each other and to surgery, ablation, and regional therapies are hampered because of the large variability in selection criteria and prior and subsequent therapies. In general, reported local control and survival are better than those reported after systemic therapies but not as good as outcomes in (better prognosis) patients treated with surgery or RFA. All radiation techniques are dependent on state-of-the-art imaging to define the liver target volume, as described by in the article by Brock in this issue (see pages 247-255). Quality assurance is also crucial; most techniques require sophisticated patient immobilization, motion management, and specialized training. All require collaboration with other members of the hepatobiliary team. A dose response has been shown for both HCC and liver metastases; however, the optimal dose and fractionation for HCC and metastases are not clearly defined. Even though high doses are desirable, hepatic and/or nonhepatic normal tissues often limit the maximal dose that can be safely delivered. Table 3 summarizes which radiation therapy techniques may be used in different situations, with more s indicating a stronger role (increased chance of tumor control and lower risks), and 0 indicating no role. When deciding which technique may be used in which situation, in addition to tumor and patient criteria, center availability, expertise and cost need to be considered. Most Appropriate Patients for Radiation Therapy The most appropriate patients to be referred for radiation therapy are those in whom radiation therapy can be delivered safely, with an expected benefit to the patient. These criteria do not necessarily overlap (ie, the technical ability to deliver high doses of radiation to focal liver cancers is different from the clinical indication for liver-directed therapy). Table 4 outlines factors to be considered in determining the safety of radiation therapy, and Table 5 outlines factors that are related to chance of patient benefit. These tables are not based on level 1 evidence, and they are meant to be used for guidance only.

Radiation therapy and liver cancer 245 Table 3 Eligibility Criteria for Different Radiation Techniques CRT SBRT Proton* Brachy Yttrium-90 <3 cm 3-6 cm 6-10 cm >10 cm Diffuse 0 0 0 0 High bleeding risk 0 0 Child-Pugh B Vascular invasion Caudate lobe Target <1 cm from GI tissues *Proton, protons or any other charged particle therapy. GI tissues, luminal gastrointestinal tissues (eg, stomach, duodenum). Abbreviations: CRT, conformal radiation therapy; SBRT, stereotactic radiation therapy; Brachy, brachytherapy; Yttrium-90, hepatic arterial yttrium-90. The proximity of the liver target volume to luminal gastrointestinal tissues, such as the stomach, impacts the ability to deliver high doses safely, but the cutoff of what distance is safe is dependent on technical factors, such as motion management, image guidance used, and the dose gradient at the target-normal tissue interface. For example, it is possible to deliver ablative doses using most radiation techniques when the target is more than 2 cm from the stomach, but if the target is within 1 cm from the stomach wall, the dose that can be delivered safely to the target most often needs to be decreased to respect the stomach radiation tolerance. Of note, insertion of tissue expanders may help to move the stomach from the tumor in this situation. Because there are little long-term data suggesting radiation therapy may compete with surgery, the gold standard local therapy, radiation therapy should be used outside of a clinical trial only for unresectable tumors or medically inoperable patients. An ideal candidate for liver radiation therapy may be a patient with unresectable disease and all the best criteria from Tables 4 and 5. An example of such an ideal liver metastases candidate is a patient with an isolated unresectable 3-cm liver metastasis from colorectal cancer, adjacent to a large vessel, more than 2 cm from all luminal gastrointestinal tissues, who is medically unsuitable for surgery or systemic therapy. An example of an ideal HCC candidate is a patient with a focal recurrence post-tace, Child-Pugh A, in whom tumoricidal doses of radiation therapy are possible to Table 4 Factors to Consider in Determining the Safety of Radiation Therapy Liver function Child-Pugh A versus B versus C Planned residual functional liver >700 ml nontumor liver (vs <700 ml) Distribution of tumor(s) (focal versus diffuse) Number of tumors (<3 vs 3-5 vs >5) Liver dose volume constraints can be met Target proximity to luminal gastrointestinal tissues >2 cm from tumor (versus 1-2 cm versus <1 cm) Experience with specific radiation therapy technique deliver safely. In these situations, radiation therapy should be considered before the next line of systemic therapy because long-term control is likely after radiation therapy. In HCC, this population of patients suitable for focal irradiation extends from very early to intermediate BCLC stage as shown in Figure 1. The highest chance of tumor control after radiation therapy is in very early and early-stage HCC ( 3 cm); however, in intermediate stage HCC 3 to 6 cm in size, tumoricidal doses are often possible to deliver, and the chance of sustained local control is only slightly reduced. Radiation therapy should be considered early in these patients natural history. In intermediate-stage HCC with a lower chance of local control from radiation therapy (multifocal tumors, 10 cm, vascular invasion, or adjacent to luminal gastrointestinal tissues), the rationale for upfront radiation therapy is reduced. In these patients, radiation therapy may be used after TACE, which may reduce the bulk of HCC. For advanced-stage N0M0 focal nonbulky HCC with portal vein invasion, radiation therapy can lead to tumor thrombosis recanalization and sustained tumor control; in these patients, radiation therapy should be considered either before or early after progression or intolerance to sorafenib. Concurrent sorafenib and radiation therapy should only be delivered on clinical trials. Radiation therapy is not likely to improve Table 5 Factors to Consider in Determining Efficacy of Radiation Therapy Natural history (indolent vs rapidly progressive) Presence of extrahepatic disease (none vs possible vs definite) Risk of extrahepatic and occult hepatic metastases Tumor type (colorectal, breast, sarcoma, melanoma vs pancreatic, lung, duodenum) Biologic selection Refractory to 1 line of systemic therapy vs 1 or more Vascular invasion (portal vein branch vs main portal vein vs inferior vena cava) Tumoricidal dose deliverable Number of tumors (< 3 vs 3-5 vs >5) Size of tumors (<6 cm vs 6-10 cm vs >10 cm)

246 L.A. Dawson Table 6 Proposed Randomized Trials in HCC 1. Intermediate BCLC HCC: TACE vs TACE and radiation therapy 2. Advanced BCLC stage, N0M0 HCC: sorafenib vs sorafenib and radiation therapy 3. End-stage symptomatic BCLC HCC: best supportive care vs low-dose radiation therapy survival in patients with diffuse HCC or extrahepatic metastases. In end-stage HCC, low-dose radiation therapy may improve symptoms and provide palliative benefit. Ideally, randomized trials will be conducted to test the benefits of radiation therapy in these situations as outlined in Table 6. Randomized trials will be more difficult to conduct in liver metastases because of the multiple primary sites and variable natural history of patients with liver metastases as well as the wide range of changing systemic therapies for these patients. Conclusions A variety of radiation therapy strategies can be used safely across a spectrum of patients with liver metastases or HCC. Toxicity is increased in patients with advanced HCC and impaired liver function (Child-Pugh B or C). For both HCC and liver metastases, long-term local control and cure are most likely in patients with 3 or fewer tumors, 3 cm each, in the absence of extrahepatic disease. Radiation therapy may also help patients with 3 or more larger liver metastases, but the benefit is less as the tumor burden increases because of reduced tumor control probability and the increased risk of occult diffuse metastases. In intermediate- and advancedstage HCC, radiation therapy can delay the time to tumor progression and recanalize vascular thrombosis. In both HCC and liver metastases, there is a strong rationale for randomized trials of radiation therapy. References 1. Bosch FX, Ribes J, Diaz M, et al: Primary liver cancer: Worldwide incidence and trends. Gastroenterology 127:S5-S16, 2004 2. Parkin DM, Bray F, Ferlay J, et al: Global cancer statistics. CA Cancer J Clin 2005:74-108, 2002 3. Jemal A, Siegel R, Xu J, et al: Cancer statistics. CA Cancer J Clin 60: 277-300, 2010 4. Bruix J, Sherman M: Practice Guidelines Committee AAftSoLD. Management of hepatocellular carcinoma. Hepatology 42:1208-1236, 2005 5. Llovet JM, Di Bisceglie AM, Bruix J, et al: Design and endpoints of clinical trials in hepatocellular carcinoma. J Natl Cancer Inst 100:698-711, 2008 6. Hellman S, Oligometastases WRR. J Clin Oncol 13:8 10, 1995 7. Fong Y, Fortner J, Sun RL, et al: Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: Analysis of 1001 consecutive cases. Ann Surg 230:309-318, 1999 8. Lo SS, Moffat-Bruce SD, Dawson LA, et al: The role of local therapy in the management of lung and liver oligometastases. Nat Rev Clin Oncol 2011 [Epub ahead of print] 9. Wei AC, Greig PD, Grant D, et al: Survival after hepatic resection for colorectal metastases: A 10-year experience. Ann Surg Oncol 13:668-676, 2006 10. Hoffmann K, Franz C, Hinz U, et al: Liver resection for multimodal treatment of breast cancer metastases: Identification of prognostic factors. Ann Surg Oncol 17:1546-1554, 2010 11. Rehders A, Peiper M, Stoecklein NH, et al: Hepatic metastasectomy for soft-tissue sarcomas: Is it justified? World J Surg 33:111-117, 2009 12. Mariani P, Piperno-Neumann S, Servois V, et al: Surgical management of liver metastases from uveal melanoma: 16 years experience at the Institut Curie. Eur J Surg Oncol 35:1192-1197, 2009 13. Hong SN, Sy L, Choi MS, et al: Comparing the outcomes of radiofrequency ablation and surgery in patients with a single small hepatocellular carcinoma and well-preserved hepatic function. J Clin Gastroenterol 39:247-252, 2005 14. Lo CM, Ngan H, Tso WK, et al: Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 35:1164-1171, 2002 15. Llovet JM, Real MI, Montana X, et al: Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: A randomised controlled trial. Lancet 359:1734-1739, 2002 16. Thornton RH, Covey A, Petre EN, et al: A comparison of outcomes from treating hepatocellular carcinoma by hepatic artery embolization in patients younger or older than 70 years. Cancer 115:5000-5006, 2009 17. Llovet JM, Ricci S, Mazzaferro V, et al: Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 359:378-390, 2008 18. Cheng AL, Kang YK, Chen Z, et al: Efficacy and safety of sorafenib in patients in the Asia-Pacific region with advanced hepatocellular carcinoma: A phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol 10:25-34, 2009 19. Bleicher RJ, Allegra DP, Nora DT, et al: Radiofrequency ablation in 447 complex unresectable liver tumors: Lessons learned. Ann Surg Oncol 10:52-58, 2003 20. Chiba T, Tokuuye K, Matsuzaki Y, et al: Proton beam therapy for hepatocellular carcinoma: A retrospective review of 162 patients. Clin Cancer Res 11:3799-3805, 2005