Hepatocellular Carcinoma

Resection versus Transplantation

Truman M. Earl; William C. Chapman


Semin Liver Dis. 2013;33(3):282-292. 

In This Article

Comparative Outcomes of Resection and Transplantation

Although the roles of resection and transplantation in well-compensated cirrhotics with early-stage HCC are often debated and choice of therapy is influenced by nonclinical factors such as surgeon specialty and evaluation at a transplant center, patients who are eligible for either therapy represent a small portion of patients presenting with newly diagnosed hepatoma. Certainly, in most Western centers where HCV infection is the predominant etiology of chronic liver failure, patients presenting with early-stage HCC and well-compensated liver disease without portal hypertension are the exception rather than the rule. Despite this, debate continues regarding optimal surgical therapy for this small group of patients. Transplantation offers lower risk of recurrence, and excellent long-term disease-free survival due to complete removal of the cirrhotic liver. However, preneoplastic liver has inherent risks including disease progression while awaiting an organ and the need for chronic immunosuppression. Moreover, transplantation of potentially resectable HCC reduces the number of available allografts for patients with end-stage liver disease. Resection, in comparison, is immediately available, carries lower morbidity and early mortality in well-selected patients, and has no impact on the deceased donor organ pool. Recurrence following resection is common and candidate selection can be challenging. Despite the high risk of recurrence following resection, several studies have shown that the overall 5-year patient-survival rate is comparable to that of transplantation (Fig. 1).[14,69,70,71,72]

Figure 1.

Actuarial patient survival after liver resection and liver transplantation. The number of patients remaining in the analysis at each time point is indicated. (From Margarit C, Escartín A, Castells L, Vargas V, Allende E, Bilbao I. Resection for hepatocellular carcinoma is a good option in Child-Turcotte-Pugh class A patients with cirrhosis who are eligible for liver transplantation. Liver Transplantation 2005;11(10):1242-1251. Copyright © 2005 American Association for the Study of Liver Diseases with permission).

In addition to clinical variables including liver and tumor factors, nonclinical factors have been shown to play a role in the selection of therapy. In a survey of 336 liver surgeons, Nathan et al surveyed surgeons as to the best treatment modality in patients with HCC in various scenarios.[73] Fifty-four percent of respondents performed liver transplantation, 94% performed liver resection, and 49% performed both. Liver-transplant surgeons were more likely to choose liver transplantation for HCC than nonliver transplant surgeons (63% vs. 50%; p < 0.001). This result was maintained in a multivariable analysis; however, when clinical data was allowed to vary with surgeon specialty, the effect of surgeon specialty became insignificant, indicating that transplant and nontransplant surgeons weigh various clinical factors such as etiology of cirrhosis, estimated waitlist time, and type of resection differently. Interestingly, 60% of liver transplant surgeons indicated that they decide for or against liver transplantation first, then consider other therapeutic modalities, whereas 42% of nonliver transplant surgeons first considered transplantation. In a second analysis of the data, the authors investigated institutional and surgeon volume factors on therapeutic decision making.[74] In this analysis, prior receipt of liver-transplant training and formal training in surgical oncology did not affect preferences among surgeons who did not perform liver transplantation. Annual volume impacted choice of therapy with higher-volume surgeons favoring liver transplantation. Surgeons at institutions where liver transplant was performed were more likely to choose liver transplantation even if they did not personally perform transplantation.

Randomized controlled trials of transplantation and resection for early-stage HCC in patients with preserved liver function are infeasible due to the large number of patients required and differences in practice patterns. This has led to numerous retrospective analyses, however, retrospective comparison of transplantation and resection is complicated by numerous factors including waitlist dropout while awaiting transplantation, nonstandard selection criteria for hepatic resection, differences in tumor size/stage, severity of liver disease, and patient demographics. Perhaps the greatest barrier to analysis of these patients has been the capture of patients listed for transplantation that did not undergo the procedure due to disease progression, death, or other contraindication to transplantation; or patients who were initially resected and recurrence was treated with salvage transplantation. This has led to several analyses based on the intention-to-treat principal. In 1999, Llovet et al presented an intention-to-treat analysis of 164 patients evaluated for surgery at the Barcelona Liver Clinic.[14] Seventy-seven patients were resected and 87 transplanted. Of those resected, 74 were CTP A. The 1-, 3-, and 5-year intention-to-treat survival rates were 85%, 62%, and 51% for resection, and 84%, 69%, and 69% for transplantation. When patients with clinically relevant portal hypertension were excluded from analysis (n = 35), 5-year survival postresection rate increased to 74%, indicating that in well-selected patients, resection may carry an overall survival rate similar to that of transplantation in this group of patients. In another intention-to-treat analysis, Facciuto et al analyzed 179 patients with cirrhosis and HCC who underwent either resection or transplantation.[70] Patients with CTP C cirrhosis, incidental HCC, and macrovascular invasion were excluded, leaving 157 patients analyzed. Of these, 51 underwent resection and 106 were listed for transplantation of which 84 were transplanted. The mean waiting time for transplant was 7 months; 21 patients were removed from the waitlist due to tumor progression. Overall survival from time of listing or liver resection did not differ between the groups with 1- and 4-year overall survival rates of 88% and 61% for resection compared with 92% and 62% for transplantation (p = 0.54). As expected, survival rates were significantly lower in both treatment groups for patients with tumors beyond Milan criteria (Fig. 2).

Figure 2.

Kaplan-Meier estimates for survival after liver resection or when listed for orthotopic liver transplantation (OLT) in (A) patients within the Milan criteria and (B) patients without the Milan criteria. Blue line = liver resection; red line = OLT. (From Facciuto ME, Rochon C, Pandey M, Rodriguez-Davalos M, Samaniego S, Wolf DC, Kim-Schluger L, Rozenblit G, Sheiner PA. Surgical dilemma: liver resection or liver transplantation for hepatocellular carcinoma and cirrhosis. Intention-to-treat analysis in patients within and outwith Milan criteria. HPB 2009;11:398-404. Copyright © International Hepato-Pancreato-Biliary Association with permission.)

Although overall 5-year survival for liver resection may not be different than that of transplantation, disease recurrence following resection is common. In an analysis of 217 patients treated with either resection or transplantation, Sapisochin et al found that 1-, 5-, and 10-year risk of recurrence rates were 19%, 67%, and 83% for resection compared with 4%, 18%, and 20% for transplantation.[75] This did not translate into a significant overall 1- and 4-year survival-rate difference (85% and 60% for resection and 82% and 62% for transplantation), likely indicating the effect of adjuvant locoregional therapies at extending survival following resection and near uniform fatality of recurrence following transplantation. There was, however, a significant difference in the 10-year survival rate with 33% of resected patients alive at 10 years compared with 49% of transplanted patients, indicating the effect of tumor recurrence and progression of liver disease on long-term survival. The primary difficulty in interpreting these studies is the inherent bias of retrospective studies regardless of intention to treat. In a meta-analysis of 10 retrospective studies comparing resection and transplantation, transplantation was found to have a survival advantage over resection (odds ratio [OR] = 0.581; 95% confidence interval [CI] 0.359-0.939;p = 0.027) (Table 1).[14,72,76,77,78,79,80,81,82,83,84] The studies included were heterogeneous and included patients with advanced liver disease and some were not based on intention to treat. Meta-analysis of three studies comparing transplantation and resection in patients with early HCC and well-compensated cirrhosis using an intention-to-treat analysis included 412 patients. A significant 5-year overall survival-rate advantage was demonstrated for patients undergoing transplantation (OR = 0.521; 95% CI 0.298-0.911;p = 0.022).[76]