17 Microwave Ablation and Hepatocellular Carcinoma 285 25. Kuang M, Lu MD, Xie XY et al (2007) Liver cancer: increased microwave delivery to ablation zone with cooled-shaft antenna–experimental and clinical studies. Radiology 242:914–924 26. Izumi N, Asahina Y, Noguchi O et al (2001) Risk factors for distant recurrence of hepato- cellular carcinoma in the liver after complete coagulation by microwave or radiofrequency ablation. Cancer 91:949–956 27. Yamanaka N, Tanaka T, Oriyama T et al (1996) Microwave coagulonecrotic therapy for hepatocellular carcinoma. World J Surg 20, 1076–1081 28. Liang P, Wang Y (2007) Microwave ablation of hepatocellular carcinoma. Oncology 72(Suppl 1):124–131 29. Iannitti DA, Martin RC, Simon CJ et al (2007) Hepatic tumor ablation with clustered microwave antennae: the US Phase II Trial. HPB (Oxford) 9:120–124 30. Shibata T, Murakami T, Ogata N (2000) Percutaneous microwave coagulation therapy for patients with primary and metastatic hepatic tumors during interruption of hepatic blood flow. Cancer 88:302–311 31. Seki T, Tamai T, Nakagawa T et al (2000) Combination therapy with transcatheter arte- rial chemoembolization and percutaneous microwave coagulation therapy for hepatocellular carcinoma. Cancer 89:1245–1251 Chapter 18 Transarterial Chemoembolization Christos Georgiades and Jean-Francois Geschwind Keywords Chemoembolization · TACE · Hepatocellular carcinoma · Drug eluting beads This chapter discusses transarterial chemoembolization (TACE), which has become the mainstay of treatment for unresectable hepatocellular carcinoma (HCC). Its suc- cess is attributable to the ability to deliver high-dose chemotherapy into the tumor vascular bed. The addition of emulsifying agents (i.e., lipiodol) and/or particles to the chemotherapy slows down the blood flow through the tumor blood supply and increases the chemotherapy residence time. Recent technological advances such as drug eluting beads further increase the intra-tumoral drug concentration and residence time, while limiting the plasma concentration. This results in increased tumoricidal effect and less systemic toxicity related to TACE. The survival benefit from TACE has been repeatedly shown to be more than double that of supportive care or systemic chemotherapy alone, with less toxicity. The approval of targeted agents for the treatment of unresectable HCC, such as Sorafenib, can have syn- ergistic effect with TACE on survival. Combination treatments that include TACE, ablation, and systemic maintenance chemotherapy will soon become the standard of care for patients with unresectable HCC. These treatments will also likely result in downsizing of many previously unresectable or non-transplantable patients, a likely benefit but also a challenge to ensure such treatment course is appropriate. Whatever the new standard treatment protocol is for HCC is undoubtedly TACE will play the central role. Many risk factors have been implicated in the development of HCC, includ- ing chronic active infection by the hepatotropic viruses, Wilson’s disease, chronic alcoholism, hemochromatosis, and α-1 antitrypsin deficiency. None of these condi- tions, however, has been proven to be directly carcinogenic. Rather, the increased risk of cancer is thought to result from the inflammation-related increased turnover J F. Geschwind (B) Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA 287 K.M. McMasters, J N. Vauthey (eds.), Hepatocellular Carcinoma, DOI 10.1007/978-1-60327-522-4_18, C Springer Science+Business Media, LLC 2011 288 C. Georgiades and J F. Geschwind of hepatocytes and consequently the increased rate of mutagenesis. The hepa- totropic viruses include the DNA hepatitis B (Hep. B) and the RNA hepatitis C (Hep. C). The latter has also been implicated in certain lymphoproliferative dis- orders, especially non-Hodgkin’s lymphoma, adding circumstantial evidence to a direct carcinogenic effect which has nonetheless not been proven. Aflatoxin B1 is the only chemical that has been shown to be directly carcinogenic to hepatocytes, while smoking, diabetes, and obesity appear to be synergistic factors that enhance the risk of HCC development [1]. Whatever the cause of HCC, the growing, viable tumor exhibits certain ubiquitous characteristics, one of which is increased vascu- larity compared to the surrounding, non-neoplastic liver parenchyma. This is the result of the strong pro-angiogenic effect exerted not only by the neoplastic cells themselves but also by the surrounding microenvironment and cell/chemical sig- naling cascades seen in chronic inflammatory states (cirrhosis in this case) and/or initiated directly by the hepatotropic viruses (i.e., increased concentration of matrix metalloproteinases [MMPs]). The importance of tumor angiogenesis for the sur- vival and growth of HCC is further underlined by the recent positive outcomes in clinical trials where HCC patients were treated with Sorafenib (Bayer HealthCare, Leverkusen, Germany), a multi-kinase inhibitor with strong predilection for VEGF receptor kinases [1, 2]. The increased vascularity of HCC compared to the sur- rounding liver parenchyma provides an opportunity for intra-arterial locoregional treatment. Vascular Anatomy of HCC A number of different cell types are the target of tumor angiogenesis signals, but the final common denominator is the recruitment of vascular endothelial cells result- ing in arteriolar and venular angiogenesis and lymphangiogenesis. The increased vascularity is more pronounced on the inflow side of the HCC, which manifests as large, tortuous, and disorganized hepatic arterioles (Fig. 18.1). The increased tumor blood supply can on occasion be so pronounced that results in shunting of blood from the rest of the liver (“sump” effect) or even an angiographically visible shunt between the arterial and the hepatic venous side of the tumor. Interestingly, the pro-angiogenic effect associated with HCC has only a weak effect on the portal venous side leaving the hepatic artery as the main HCC supplier (Fig. 18.2). This phenomenon has therapeutic implications in the field of Interventional Radiology. Transarterial Chemoembolization (TACE) has become the mainstay of treatment for unresectable HCC. During TACE, a catheter is placed in the branch of the hepatic artery supplying the tumor, and high-dose chemotherapy emulsified with ethiodol (Savage Laboratories, Melville, NY, USA) is infused to near occlusion aided by variable size particle embolization. Since the tumor receives most of its blood sup- ply from the hepatic artery and the normal l iver parenchyma from the portal vein, TACE selectively delivers chemotherapy to the tumor while mostly sparing nor- mal liver. In addition, it has long been known based on empirical observations that 18 Transarterial Chemoembolization 289 Fig. 18.1 Vascular supply of HCC. Angiogram just prior to TACE (a) performed via a selective microcatheter placed in a branch of the right hepatic artery (black arrowhead) shows the disorga- nized nature of the hepatic arterioles supplying the tumor (white arrowheads). Post-TACE image (b) shows pooling of the lipiodol–chemotherapy mixture in the abnormal vascular bed of the tumor (arrowheads) Fig. 18.2 Schematic of the vascular supply of HCC. The majority of the hepatic parenchyma blood supply (∼70–80%) i s via the portal vein. Tumor-induced angiogenesis recruits mostly hepatic arterial branches thus HCC supply is almost exclusively by the hepatic artery. Therefore, TACE will preferentially treat the HCC and mostly spare normal liver parenchyma ethiodol selectively embolizes in the abnormal vascularity of tumors, further increasing the chemotherapy concentration and tumor residence time. The reasons for the tropism of ethiodol toward the tumor vascularity have not been clarified yet. Figure 18.3 shows the chemoembolization procedure. Figures 18.4 and 18.5 290 C. Georgiades and J F. Geschwind Fig. 18.3 Transarterial chemoembolization in a patient with unresectable HCC. Coronal T1- weighted, contrast-enhanced MRI image of the l iver (a) shows a mass in the lower part of the right lobe of the liver (arrows) and a smaller lesion (arrowhead) more cranially. Digitally subtracted, selective, right hepatic arteriogram (b) shows the two masses, indicated by arrows (larger)andarrowhead (smaller) to enhance. Post-TACE, coronal, non-enhanced CT image of the liver (c) shows the distribution of the lipiodol–chemotherapy mixture to correspond to the larger (arrows) and smaller (arrowhead) masses. Axial, T-1 weighted, contrast-enhanced MRI prior to TACE (d) and axial non-enhanced CT of the liver post-TACE show distribution of the lipiodol– chemotherapy mixture within the vascular portions of the tumor (arrows) while sparing the necrotic (devascularized) portions (arrowheads) showcase varied responses to treatment which include both Response Evaluation Criteria in Solid Tumors (RECIST) and European Association for the Study of Liver (EASL) responses. In reality, most HCCs will show both types of response to TACE; however, usually EASL criteria r esponse is more pronounced than RECIST response. Clinical Trials and Current Evidence At the time of diagnosis the majority of HCC patients (approximately 85%) are not candidates for transplantation or resection, the two choices offering the best chance for cure. The lack of effective chemotherapy and the poor response of HCC to radiation treatment left TACE as the only realistic treatment option. Because of the lack of competing therapies, researchers were not compelled to seriously study the efficacy of TACE until the early 2000s, a source of frequent criticism. In 2002, Llovetetal[3] published results from a randomized controlled trial, which was 18 Transarterial Chemoembolization 291 Fig. 18.4 50-year-old female with unresectable HCC. Three-month, sequential MRI (a, b, c, d, e) axial images of the liver after TACE. The initially large tumor replacing the entire right lobe of the liver shows gradual response based on RECIST criteria. At last follow-up, 5 years after initial TACE, the residual tumor is a 3 cm calcified nodule in the posterior right lobe of the liver (F) Fig. 18.5 34-year-old female with unresectable HCC. Pre- (a) and post-TACE (b), axial, contrast- enhanced, MRI images show the tumor (arrows) with mild RECIST response but significant EASL-based response, indicated by significant necrosis stopped early because TACE provided a s tatistically significant survival benefit in the treatment group (1- and 2-year survival of 82% and 63% for TACE vs 63% and 27% for the supportive care) (Fig. 18.6). A meta-analysis of five randomized controlled trials published in the same year also concluded that TACE reduced the 2-year mortality of patients with unresectable HCC (odds ratio 0.54, CI 95%, 0.33–0.89, p = 0.015) [4]. The benefits of TACE were cemented after a study by Lo et al [5] showed statistically significant survival benefit in patients with unre- sectable HCC treated with lipiodol–cisplatin chemoembolization. The 1-, 2-, and 292 C. Georgiades and J F. Geschwind Fig. 18.6 Survival of patients with unresectable HCC. The 1- and 2-year survival of patients treated with TACE (dark gray) is significantly better than those receiving supportive care alone (light gray) 3-year survival in TACE-treated patients was reported by Lo et al to be 57, 31, and 26%, compared to 32, 11, and 3%, respectively, in the control group. Finally, in a meta-analysis of randomized controlled trials, Llovet and Bruix [6] showed signif- icantly decreased 2-year mortality in patients treated with chemoembolization with an odds ratio of 0.53 (CI 95%, 0.32–0.89, p = 0.017). Recent evidence has emerged that compares resection vs. locoregional treatments for HCC with favorable con- clusions for TACE. Yamagiwa et al [7], for example, have shown that combination treatment using radiofrequency ablation (RFA) and TACE resulted in significantly longer overall 5-year survival compared to hepatic resection (72% vs 59%), albeit a shorter disease-free survival (14 vs 32 months). Combination locoregional treat- ment using RFA and TACE were also compared to hepatic resection for small HCCs (< 3 cm) by Yamakado et al [8]. The 1-, 3-, and 5-year survival was identical between the two groups at 98%, 94%, 75% and 97%, 93%, 81% for locoregional treatment and resection, respectively (Fig. 18.7). Also identical was the cancer recurrence rate at 36 and 37% for locoregional and surgical treatments, respectively. Fig. 18.7 Survival for patients with resectable HCC. Overall, 5-year survival is identical between patients treated with resection (light gray) and those treated with combination RFA+TACE (dark gray). The comparison was for patients with HCC ≤ 3cm 18 Transarterial Chemoembolization 293 One of the lingering criticisms of TACE is that the protocol has not been standardized yet. While most physicians will use a triple chemotherapy cocktail that includes cisplatin, mitomycin C, and doxorubicin, others will use double or monotherapy regimens while still a minority will only embolize the feeding ves- sel without chemotherapy. To date, there has not been a prospective, randomized trial comparing the efficacy of different chemotherapy regimens. Cisplatin causes DNA intrastrand crosslinks, mitomycin C is an alkylating agent that causes DNA interstrand crosslinking and inhibits DNA-dependent RNA polymerase, and doxoru- bicin intercalates DNA and inhibits proliferation-specific DNA polymerases. The choice of triple-regimen TACE rests on the theory that a multi-pronged attack on the neoplastic cell DNA is more effective and can overcome possible tumor drug resistance. TACE as a Bridge to Transplantation Recently there has been considerable interest regarding the role of TACE as a bridge to liver transplantation. It has been postulated that TACE shrinks and/or slows the progression of HCC thus possibly minimizing waiting list drop-off rates. This would theoretically be the case especially for those patients who are barely within Milan or San Francisco t ransplantation criteria. There is a lack of well-designed, prospective, randomized studies however, and the published ones have thus far been equivocal. One prospective (but not randomized) study showed a 1-, 2-, and 5-year survival after TACE and orthotopic liver transplantation (OLT) of 98, 98, and 93%, respec- tively [9], which is better than historical controls, suggesting that pretransplantation TACE does improve survival. The same study, however, concluded that downstag- ing of patients to within Milan criteria using TACE did not result in any survival benefit. Two other studies [10, 11] have correlated the degree of necrosis with out- come after OLT. One concluded that a high percent of lesion necrosis after TACE predicts lower tumor recurrence rates after OLT, whereas the other concluded that low necrosis rates after TACE “facilitate tumor recurrence.” The latter is unlikely, as there is no teleological effect for TACE. The results above, rather, suggest that good response to TACE indicates favorable disease biology. Overall, the current literature suggests (but is not definitive) that (1) pretransplantation TACE for patients within but close to falling out of criteria may be beneficial and (2) response to TACE may be predictive of disease biology and by extension, survival after liver transplanta- tion. Further studies are needed in order to define TACE’s role as a bridge to liver transplant. Quality of Life/Toxicity Profile HCC and cirrhosis are frequent comorbid conditions that have a significant impact on the patients’ quality of life. Pain from the expanding tumor, especially if it 294 C. Georgiades and J F. Geschwind is in a subcapsular location is a common presenting symptom for HCC. Ascites, edema, fatigue are symptoms related to the cirrhosis. Under these circumstances any treatment (except transplantation) has a potential of worsening the symp- toms, be it surgical resection, percutaneous ablation, systemic chemotherapy, or TACE. (Possible TACE-related complications are shown in Table 18.1.) Strong con- traindications to TACE include Child-Pugh C liver cirrhosis and poor performance status, i.e., Eastern Cooperative Oncology Group (ECOG) ≥ 3 or Karnofsky sta- tus <60. Performing TACE in patients with minimal if any liver functional reserve increases the risk of hepatic failure and in any case is unlikely to prolong sur- vival. When indicated, however, TACE has been shown to result in much lower plasma concentration and greater intra-tumoral chemotherapy concentration, com- pared to systemic treatment. Data from our group (pending publication) show a very favorable toxicity profile for TACE. The group recorded hematologic and non-hematologic toxicities r elated to TACE and categorized them based on the Common Terminology Criteria for Adverse Events (CTCAE). Grade 3–4 toxic- ities were found in about 20% of patients, a number much lower than reported for systemic chemotherapy [12]. This is an important consideration because TACE is usually reserved as salvage treatment in many patients with moderate cirrhosis. Furthermore, recent evidence suggests that the newer form of TACE – i.e., drug elut- ing beads – shows an even better pharmacokinetic profile than lipiodol-based TACE, including lower plasma chemotherapy levels and lower related systemic toxicity (Fig. 18.6)[13]. Table 18.1 Possible TACE-related complications. Third column shows the overall percent risk as reported in literature if the risk factor in column 2 is present. For example, patients who have Child-Pugh C liver cirrhosis (row 1, column 2) have a significant risk for liver failure (column 3) with TACE. If TACE can be performed in a superselective manner (column 4) thus sparing most of the liver then the chances of liver failure are minimized Complication Risk factor % Risk Risk mitigation action Liver failure, death, encephalopathy Child-Pugh C 5–10% for Child-Pugh C Superselective embolizationT. bilirubin > 4 mg/dl Albumin < 2 mg/dl Poor performance status Liver abscess Compromised Sphincter of Oddi 30–80% (else <5%) Broad spectrum Abx/ GI preparation Non-target embolization Aberrant anatomy especially left or right gastric artery <10% Place catheter distal to origin of gastric artery/watch for chemo reflux Pulmonary embolism Tumor shunting <1% Gelfoam embolization of shunt Upper GI bleeding Gastroesophageal varices Unknown (rare) Pre-TACE banding? Acute renal failure Renal insufficiency, diabetes 0.05–5% Hydration, renoprotection, minimize contrast . prospective, randomized studies however, and the published ones have thus far been equivocal. One prospective (but not randomized) study showed a 1-, 2-, and 5-year survival after TACE and orthotopic. chemotherapy into the tumor vascular bed. The addition of emulsifying agents (i.e., lipiodol) and/ or particles to the chemotherapy slows down the blood flow through the tumor blood supply and increases. survival benefit in the treatment group (1- and 2-year survival of 82% and 63% for TACE vs 63% and 27% for the supportive care) (Fig. 18.6). A meta-analysis of five randomized controlled trials published