Hepatocellular Carcinoma: Targeted Therapy and Multidisciplinary P29

16 Radiofrequency Ablation for Hepatocellular Carcinoma 265 DynamicMRIormultiphasichelicalCTperformedduringthefirst3monthsfollow-ing RFA often demonstrates a hypervascular enhancing rim of inflammatory tissue. This inflammatory response can be difficult to distinguish from tumor recurrence but typically resolves and is usually not evident by scans obtained after the first 6 months. Identifying a local recurrence following RFA can be problematic. Early follow-up scans may not be able to differentiate between a recurrence and an inflammatory response. A local recurrence may be detected as progressive local ingrowth of vas-cularized tissue into the necrotic cavity or as vascularized outgrowth away from the RFA cavity. The arterial phase of a dynamic imaging study is best to identify recur-rences because the tumor tissue may appear otherwise isodense with the normal parenchyma. In small pilot studies, positron emission tomography (PET) scans have been used effectively to identify local recurrence following RFA. In one study of 24 patients who were treated by RFA for 33 lesions, PET scans were compared with CT scans for tumor surveillance [13]. The overall detection rate was 92% for PET scan com-paredwith75%forCTscans.EarlydetectionwasalsoimprovedwiththeuseofPET scans. Other small studies in patients with metastatic hepatic tumors have supported the role of PET scans in surveillance post-RFA [14–16]. Indications for RFA of Hepatocellular Carcinoma The US Food and Drug Administration approved RFA for general tissue abla-tion in 1996 and for ablation of unresectable liver tumors in 2001. Generally, the goal of RFA in HCC patients is treatment with curative intent. Therefore, patients with preoperative or intra-operative evidence of extra-hepatic disease should not be considered for RFA. Recently, long-term outcomes were reported from a prospective collaboration between the M.D. Anderson Cancer Center and the G. Pascale National Cancer Institute in Naples, Italy [17, 18]. The initial study of 110 prospectively treated patients was reported in 2000. In this investigation, 149 HCC tumors were treated with RFA. Laparoscopic RFA was performed in 3 cases, open RFA in 31 cases, and percutaneous RFA in 78 cases. Median diameter of tumors treated percuta-neously (2.8 cm) was smaller than tumors treated open (4.6 cm). With a median follow-up of 19 months, only four patients (3.6%) had evidence of local recurrence. All recurrences occurred in patients with tumors larger than 4.0 cm in diame-ter. Complications occurred in 12.7% of the patients with no treatment-related deaths. Complications included symptomatic pleural effusion, subcutaneous and subcapsular hematoma, and ventricular fibrillation. One patient with Child’s class B cirrhosis developed intra-tumoral bleeding requiring transfusion and hepatic arterial embolization. No patients developed hepatic or renal failure or sustained thermal injury to the surrounding organs. This initial experience with RFA for cirrhotic 266 E.R. Camp et al. patients with HCC demonstrated this to be a safe and feasible alternative treatment strategy for early-stage tumors. The follow-up experience in an extended cohort of patients continued to support the use of RFA in cirrhotic patients with early-stage, unresectable HCC [18]. The authors analyzed 194 patients who underwent RFA for 289 tumors with a median follow-up of 34.8 months. Disease recurred in 103 patients (53%) with a local recur-rence rate of only 4.6%. The overall survival rates at 1, 3, and 5 years were 84.5, 68.1, and 55.4%, respectively. In further analysis, 5-year survival rates were sim-ilar between the groups of patients treated by percutaneous RFA, open RFA, and in combination with hepatic resection. The patients treated with a combination of surgery and RFA had a lower short-term survival rate, likely related to the increased surgical morbidity observed in cirrhotic patients. Similar long-term survival data have been demonstrated in large clinical series with institutions in the United States, Europe, and Asia reporting 5-year survival rates with RFA comparable to hepatic resection [19–23] (Table 16.1). An Italian prospective, intention-to-treat analysis of RFA in 206 patients with HCC reported a 5-year overall survival rate of 48%. Based on multivariable analysis, they deter-mined Childs-Pugh class and the presence of multiple tumors as predictors of long-term survival [20]. These long-term analyses highlight the potential curative results when RFA is used in appropriately selected patients. Table 16.1 Long-term survival and local recurrence results with the use of radiofrequency ablation for HCC Authors Raut et al. [18] Lam et al. [19] Lencioni et al. [20] Montorsi et al. [21] Chen et al. [22] Choi et al. [23] No. of No. of patients tumors 194 289 273 357 187 240 58 58 71 71 570 674 Median tumor size (cm) 3.3 3.0 2.8 <5 cm – 2.2 Median follow-up period (months) 34.8 24 24 25.7 27.9 26 Local recurrence (%) 4.6 12.8 5.3 35 – 12.1 Overall survival (%) 5 yr – 55.4 5 yr – 38 5 yr – 41 4 yr – 45 4 yr – 67.9 5 yr – 58 Although effective treatment of large (>4.0 cm) HCC is possible with RFA, the concern of increased local recurrence from incomplete tumor destruction in this patient population still remains a significant problem. A recent report from China evaluated the effectiveness of RFA for HCC tumors between 3.1 and 8.0 cm in largest diameter [24]. In a total of 35 patients with tumors larger than 3.0 cm with a medianfollow-upof11months,completeablationwasachievedingreaterthan80% of the cases. Using this treatment strategy, complications occurred in 17% of the patients with a mortality rate of 3%. During this short-term follow-up period, local 16 Radiofrequency Ablation for Hepatocellular Carcinoma 267 recurrence, distant intra-hepatic recurrence, and extra-hepatic recurrence occurred in only 3, 24, and 6% of the patients, respectively. The cumulative 6-, 12-, and 18-month survival rates were 85, 81, and 76%, respectively. Other series reporting RFA to treat large HCC tumors have demonstrated higher local recurrence rates with longer follow-up times highlighting the inherent tech-nical difficulties associated with producing complete thermal destruction of larger tumors [17, 25]. In 84 patients treated with hepatic tumor RFA at the John Wayne Cancer Center, local recurrences were identified in 15 patients [25]. Approximately one third of the patients with tumors larger than 3 cm in diameter recurred despite treating larger tumors with multiple overlapping ablations to include the tumor with a 1 cm margin. Comparing RFA to Surgical Resection for HCC Investigations comparing RFA to surgical resection continue to show equivalent long-term outcomes in patients with early-stage HCC. A prospective, randomized trial from China was recently reported comparing RFA to surgical resection in appropriate surgical candidates with a solitary HCC tumor less than 5 cm in diam-eter [22]. Complete tumor necrosis was achieved in 91.5% of the patients treated with RFA. Major procedure-related complications were significantly greater in the surgical group (55%) compared with RFA treatment (4%). In this intention-to-treat trial of 180 patients, the RFA and surgical resection groups had similar overall sur-vival rates (67.9% vs. 64%, respectively) as well as disease-free survival (46.4% vs. 51.6%, respectively). Retrospective series have demonstrated equivalent survival outcomes between surgery and RFA for HCC tumors. Montorsi et al. compared laparoscopic RFA with surgery in patients with solitary HCC tumors smaller than 5 cm [21]. The 4-year sur-vival rates were equivalent between surgical resection and RFA. In contrast, local recurrence was significantly higher in the RFA subset compared with surgery (53% vs. 30%). In a second prospective study from Italy, RFA at one specialized institu-tion was compared to resection performed at a second institution for the treatment of HCC [26]. The overall 3-year survival rates and disease-free survival results heav-ily favored resection for Child’s class A patients. The 3-year survival rates for RFA and resection were 33 and 65%, respectively. Similarly, the 3-year disease-free sur-vival rates for RFA and resection were 20 and 50%, respectively. For Child0s class B patients, the results were equivalent for the two treatment groups. Although this was a prospective study, the treatment groups were not comparable due to selection bias. The RFA group had a greater number of tumors per patient and worse hepatic function based on Child-Pugh class than the patients treated with surgical resection. These differences should be considered when interpreting the results. So while there were significant differences inoutcomes between thetwogroups – RFA and surgery, the results might reflect the underlying disease rather than treatment effect. 268 E.R. Camp et al. RFA in Combination with Surgical Resection RFA has been combined with surgical resection for the treatment of multiple HCC tumors. This combined approach is ideal for patients with bilobar tumors. Using a combined approach, surgical resection can be used for the largest lesion or for segments with the majority of the tumor burden reserving RFA for the preserved liver and smaller residual tumors. The experience from M.D. Anderson and the G. Pascale National Cancer Institute described this strategy for complex cases [18]. Of the 54 patients treated by open RFA, 22 (41%) patients underwent partial hepatic resection. Major hepatectomy was performed in only five patients. Procedure-related mortality (20%) and morbidity were increased when a combined approach was used. Open RFA with or without resection resulted in high local recurrence rates compared to the percutaneous approach, likely due to selection bias. Early survival rates were significantly worse in the combined resection and RFA group directly related to the high procedure-related mortality, increased tumor burden, and the morbidity associated with hepatic resection in cirrhotic patients. Long-term sur-vival rates were equivalent among all patients highlighting the potential utility of aggressive liver-directed therapy even for patients with extensive bilobar disease. Five-year median survival rates for percutaneous RFA, open RFA without hepatic resection,andopenRFAwithhepaticresectionwere57.5,45.9,44.8%,respectively. Choi et al. reported a series of 53 patients with a total of 148 HCC tumors treated with a combined RFA and surgical resection approach [27]. Child-Pugh class A or B patients were considered candidates for a combined approach if (1) the indocyanine green dye retention rate at 15 min was less than 10%, (2) only one to three tumors <4 cm were left for RFA following resection, and (3) no extra-hepatic malignancy was present. There were no procedure-related deaths and an 8% morbidity rate was observed. Of the 66 tumors ablated, complete tumor ablation was achieved in 98% basedonthe1-monthfollow-upCTscan.Witha22monthmedianfollow-upperiod, local tumor progression occurred in two patients (3%) with, previously, completely ablated tumors. The 5-year survival rate for the entire group of patients was 55%. The only independent predictor of survival on multivariable analysis was resected tumor size. The estimated 1, 3, and 5-year cancer-free survival rates were 41, 28, and 0%, respectively. In addition to tumor size, the presence of microvascular inva-sion also increased the likelihood for recurrence based on multivariable analysis. These studies highlight an opportunity to impact survival in patients with multifocal HCC with a combination of surgical resection and RFA. Clearly, the patients must be carefully evaluated preoperatively for assessment of appropriate residual liver function before electing to pursue this strategy. Comparing RFA to Other Ablative Techniques In addition to RFA, other ablative techniques such as percutaneous ethanol injec-tion (PEI), acetic acid injection, and cryoablation therapy have been used to treat patients with HCC. PEI is commonly delivered with ultrasound guidance for 16 Radiofrequency Ablation for Hepatocellular Carcinoma 269 monitoring similar to RFA. As opposed to RFA which usually requires one pro-cedure, PEI usually requires multiple injections performed over weeks to achieve adequate tumor necrosis. The number of treatments and quantity of alcohol injected depends on the size of the tumor. A similar treatment plan is used for delivery of acetic acid. Cryoablation creates a region of subzero temperatures to destroy tumors. Using the cryoprobe, an iceball with rapid tissue freezing is created. Like RFA, cryoablation may be delivered by percutaneous, laparoscopic and open approaches. Typically, two freeze-thaw cycles are used with the goal of achieving 5–10 mm ablation margin of normal tissue around the tumor. Cryoablation is associated with a high morbidity rate (50%) as well as a treatment-related mortality rate of 1.6% [28]. PEI and RFA are the most widely used local tumor destruction techniques for HCC and have been compared in several recent investigations [29–31] (Table 16.2). Two randomized studies have recently compared RFA and PEI in the treatment of early-stage HCC and both trials favored RFA as the superior technique. Lencioni et al. randomized HCC patients with a solitary tumor less than 5 cm in diameter or multiple tumors all less than 3 cm to either RFA or PEI. The schedule of PEI delivery was determined based on the tumor burden [29]. Complete tumor ablation was achieved in 91% of the patients with an average of 1.1 treatments using RFA compared to 82% complete ablation with an average of 5.4 treatments using PEI. A trend toward improved survival was observed in the RFA group, although the short observation period limited this analysis. However, local recurrence was signif-icantly less in the RFA group compared with PEI. A Taiwan trial recently compared RFA, standard PEI, and high-dose PEI for HCC smaller than 4 cm in a total of 157 patients [30]. Similar to the previous study, the investigators observed improved tumor ablation requiring fewer sessions with RFA. Survival and local recurrence rates were significantly improved with RFA treatment compared with either PEI schedule. Multivariate analysis revealed tumor size and treatment strategy were sig-nificant factors determining outcomes. In summary, these trials suggest that RFA requireslesstreatmentsessionsandisassociatedwithimprovedoutcomescompared with PEI. Table 16.2 Comparison of percutaneous ethanol injection with radiofrequency ablation for HCC Authors Lencioni et al. [29] Lin et al. [30] Shiina et al. [31] No. of patients PEI/RFA 50/52 52/52 114/118 Median follow-up period (months) 23 24 36 PEI Local recurrence (%) 38 34.6 11.4 Overall survival (%) 2 yr–88 3 yr–50 4 yr–57 RFA Local recurrence (%) 4 14 1.7 Overall survival (%) 2 yr – 98 3 yr – 74 4 yr – 74 ... - tailieumienphi.vn