Figure 1. Interval between hepatic resection and tumor recurrence in patients with hepatocellular carcinoma with and without cirrhosis.
Figure 2. A, Comparison of disease-free survival curves of patients with and without cirrhosis and TNM stage I disease. B, Comparison of overall survival curves of patients with and without cirrhosis and TNM stage I disease. C, Comparison of disease-free survival curves of patients with and without cirrhosis and TNM stage II disease. D, Comparison of overall survival curves of patients with and without cirrhosis and TNM stage II disease. E, Comparison of disease-free survival curves of patients with and without cirrhosis and TNM stage III disease. F, Comparison of overall survival curves of patients with and without cirrhosis and TNM stage III disease.
Figure 3. Comparison of disease-free survival curves of patients with hepatocellular carcinoma with and without cirrhosis who have transplantable tumors.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Chang C, Chau G, Lui W, Tsay S, King K, Wu C. Long-term Results of Hepatic Resection for Hepatocellular Carcinoma Originating From the Noncirrhotic Liver. Arch Surg. 2004;139(3):320–325. doi:10.1001/archsurg.139.3.320
In patients with hepatocellular carcinoma who do not have cirrhosis, the clinicopathologic characteristics and long-term postresectional outcomes must be clarified and liver transplantation may also have a role in future treatment strategy.
Case series. The mean (SD) follow-up time was 52.4 (33.8) months.
A tertiary care medical center.
From a prospective database, 445 patients with hepatocellular carcinoma who underwent hepatectomy were classified into 2 groups—those without cirrhosis (n = 223) and those with cirrhosis (n = 222). Clinicopathologic factors and postresectional outcomes were compared between these groups based on the new American Joint Committee on Cancer/Union Internationale Contre le Cancer TNM (sixth edition) staging system and the patient selection criteria for undergoing transplantation.
Main Outcome Measures
Postresectional disease-free and overall survival rates.
Compared with patients with cirrhosis, patients without cirrhosis were younger, had a lower rate of viral hepatitis type C infection, and had more advanced TNM stage III disease. Also more of the patients who did not have cirrhosis had undergone major resection. The tumor recurrence rate was significantly lower in the noncirrhotic group than in the cirrhotic group (59.5% vs 69.5%, P = .03). The 5- and 10-year disease-free and overall survival rates of the noncirrhotic group were 36.8% and 25.7%, and 53.0% and 36.9%, respectively. The survival of the members of the noncirrhotic group was better than the survival of the members of the cirrhotic group for patients with early stage (TNM stage I or transplantable) diseases. The 5-year disease-free and overall survival rates in patients without cirrhosis with transplantable diseases were 54.8% and 70.0%, respectively.
In early stage diseases, patients without cirrhosis had significantly better survival rates than patients with cirrhosis. For a small hepatocellular carcinoma originating in a noncirrhotic liver, hepatic resection is a reasonable first-line treatment. Liver transplantation can be reserved as salvage treatment for patients with recurrent disease after hepatic resection.
Hepatocellular carcinoma (HCC) is a common malignancy worldwide and is frequently associated with cirrhosis.1 At present, hepatic resection and liver transplantation are the mainstays of treatment with curative intent. For patients with early HCC and decompensated cirrhosis, liver transplantation is the treatment of choice since the procedure potentially cures both the cirrhosis and the HCC, and the outcome after liver transplantation is universally accepted to be better than hepatic resection.2
In about 20% to 50% of the patients with HCC the underlying liver has no cirrhotic change.3-5 Articles concerning the clinicopathologic and prognostic differences in HCC with or without cirrhosis have been controversial.4,6,7 Although hepatic resection is generally considered the primary therapy in patients with HCC and a noncirrhotic liver, whether liver transplantation has a role in the treatment of this type of patient remains unknown. A systematic review of 77 patients undergoing transplantation for noncirrhotic, nonfibrolamellar–type HCC without strict patient selection criteria showed a 5-year survival rate of only 11%.8
The 2-fold purpose of the present study was to evaluate the clinicopathologic features and prognosis of HCC originating from a noncirrhotic liver, and based on our long-term outcomes after hepatectomy, to evaluate whether liver transplantation has a role in future treatment strategy. All data were prospectively collected and retrospectively analyzed.
Between July 1, 1991, and December 31, 1998, 445 patients with HCC underwent hepatectomy at Taipei Veterans General Hospital, Taipei, Taiwan. These patients had the tumor tissue completely removed macroscopically. Of 445 patients, 223 patients (50.1%) had a tumor originating in a noncirrhotic liver and 222 patients (49.9%) had a tumor originating in a cirrhotic liver. No patient had a fibrolamellar-type carcinoma. Three patients without cirrhosis (1.3%) and 9 patients with cirrhosis (4.2%) died in the hospital. The remaining 433 patients were followed up for a period of at least 4 years. The mean (SD) follow-up time was 52.4 (33.8) months.
The operative procedures have been described elsewhere.9 Major hepatic resection indicates resection of 3 or more Couinaud segments. After the surgical procedure, the macroscopic features of the tumor, including size, number, capsular formation, and vascular invasion were recorded. The absence of cirrhosis was confirmed histologically according to the criteria proposed by Anthony et al.10 The patients were classified according to the new American Joint Committee on Cancer/Union Internationale Contre le Cancer TNM staging system.11 Eligibility for liver transplantation for HCC was based on the following Milanese criteria12: a solitary tumor 5 cm or smaller or 2 or 3 tumor nodules (oligonodular) each 3 cm or smaller; the absence of extrahepatic metastasis; and the absence of macroscopic vascular invasion. DNA ploidy, analyzed with fresh tissue samples, was measured using a flow cytometer (Epics Profiles; Coulter Electronics, Hialeah, Fla) as previously described.13
After the operation, patients were followed up regularly every 2 to 3 months by means of measuring the serum α1-fetoprotein level, ultrasonographic and/or computed tomographic scans, or magnetic resonance imaging. When recurrence was indicated on any of these examinations, hepatic angiography was performed to confirm its existence.
Statistical calculations were made using commercially available computer software (SPSS/PC+; SPSS Inc, Chicago, Ill). Results are expressed as mean (SD). The t test was used to compare continuous variables; the χ2 test was used to compare discrete variables. Survival rates were calculated using the Kaplan-Meier method,14 and survival curves were compared using the log rank test. P<.05 was considered statistically significant.
Compared with patients with cirrhosis, patients without cirrhosis were significantly younger, had less history of acute hepatitis, had a lower frequency of viral hepatitis type C infection, had a higher platelet count, and had better preoperative liver function compared with patients with cirrhosis (Table 1). A significantly higher proportion of patients without cirrhosis underwent major resection. On clinicopathologic examination, patients without cirrhosis had significantly larger-sized tumors and more of these patients were in advanced TNM stage III disease (Table 2).
The overall incidence of tumor recurrence was significantly lower in the noncirrhotic group than in the cirrhotic group (59.5% vs 69.5%, P = .03). For patients with recurrent disease, 51 (38.9%) of 131 patients without cirrhosis had extrahepatic recurrences, whereas 47 (31.8%) of 148 patients with cirrhosis had extrahepatic recurrences (P>.05). For both groups of patients, most of the recurrences (77.1% in the noncirrhotic group and 64.2% in the cirrhotic group) occurred within 2 years after undergoing surgery (Figure 1).
In the noncirrhotic group, the mean disease-free survival was 38.2 (36.5) months. Including the operative mortality, the 5- and 10-year disease-free survival rates in the noncirrhotic group were 36.8% and 25.7%, respectively, while the 5- and 10-year disease-free survival rates in the cirrhotic group were 26.9%, and 12.6%, respectively. The 5- and 10-year overall survival rates in the noncirrhotic group were 53.0% and 36.9%, respectively, while the 5- and 10-year overall survival rates in the cirrhotic group were 47.9% and 29.3%, respectively. The disease-free survival rates were significantly better in the noncirrhotic group (P = .03).
When analyzed by TNM stages, the survival rates showed no statistically significant differences between the noncirrhotic and cirrhotic groups in patients with stage II or III disease. However, patients without cirrhosis with early stage I disease had significantly better disease-free (P<.001) and overall survival rates (P = .005) than patients with cirrhosis (Figure 2). For subgroups of patients with transplantable HCC, the disease-free survival rate (P<.001) and overall survival rate (P = .004) remained significantly better in the noncirrhotic group. In patients without cirrhosis who had livers that were transplantable, the 1-, 3-, 5-, and 10-year disease-free survival rates were 85.3%, 65.2%, 54.8%, and 35.5%, respectively (Figure 3); the 1-, 3-, 5-, and 10-year overall survival rates were 97.3%, 86.6%, 70.0%, and 50.1%, respectively.
In subgroups of patients with transplantable HCC, comparison of the tumor-related factors and outcome between the noncirrhotic and cirrhotic groups is given in Table 3. The noncirrhotic group had significantly larger-sized tumor, had a significantly higher incidence of solitary tumors, underwent major resection with a wider surgical margin (>1 cm) in the resected specimen, and had preserved liver function.
There has been controversy concerning whether differences existed in the long-term prognosis for HCC in patients with or without coexisting cirrhosis.4-6 In our study, patients without cirrhosis were significantly younger and their liver function was better than that of patients with cirrhosis. Tumors were larger with more patients being in advanced TNM stage III disease in the noncirrhotic group. Our data failed to support previous reports that vascular invasion and high-grade tumors were more prevalent in HCCs arising in cirrhotic livers.4 There was also no difference in the percentages of patients with tumor DNA aneuploidy in both groups.
In 2002 a simplified American Joint Committee on Cancer/Union Internationale Contre le Cancer TNM staging for HCC was proposed.11 With this new TNM staging system, we have found that there is a clear prognostic demarcation of our patients between each stage, both in the noncirrhotic and cirrhotic groups (Figure 2). One special finding of our study is that only in patients with early stage (TNM stage I or transplantable HCC) disease did the presence of cirrhosis have an unfavorable effect on long-term survival after surgery. The lower recurrence rate and the better prognosis in the patients with early stage disease without cirrhosis may be due to multiple factors and can be attributed to the differences in the tumor-related and surgical factors between the 2 groups. In our study, and as reported by Sasaki et al,6 the major cause of death in patients with and without cirrhosis was cancer recurrence, and in both groups most of the recurrences occurred within 2 years after undergoing surgery (Figure 1). It has been reported that in the cirrhotic liver, the presence of fibrosis, regenerative nodules, and parenchymal necrosis produces a heterogenous parenchyma that makes detection of HCC nodules more difficult than in patients without cirrhosis.15 The HCC nodules in a liver with cirrhotic background can be easily underdetected by radiological imaging.16,17 Thus, the presumably higher frequency of occult nodules in patients with cirrhosis may be the likely reason for the poorer prognosis in these patients. Because patients with and without cirrhosis have a high probability of developing intrahepatic metastasis at the more advanced stages of the disease, the prognosis of both groups is expected to be the same and equally dismal. At the early stages of the disease (TNM stage I and transplantable HCC), the probability of intrahepatic metastasis in patients without cirrhosis is low and effective control of local cancer is useful to avoid tumor recurrence.18 An extensive resection with a wide tumor clearance margin, which is feasible only in patients without cirrhosis, can result in a more favorable prognosis in this group of patients.19,20 In addition to these factors, the residual cirrhotic liver after hepatic resection continues to have malignant potential and may be associated with a higher postresectional recurrence rate.21
In general, the results of liver transplantation are better than those of resection for the treatment of HCC in patients with cirrhosis, particularly in patients who meet the Milanese criteria.12 Based on the success of transplantation in the treatment of cirrhotic HCC, the role of liver transplantation for HCC is expected to evolve, and this procedure has been proposed for patients with small HCCs who have normal Child-Pugh class A liver function.22,23 However, some factors should be considered in selecting patients for liver transplantation who have HCC without cirrhosis. There are problems related to transplantation, including opportunistic infection, graft failure, and development of other malignant neoplasms as a result of long-term immunosuppression. In addition, owing to the scarcity of organ donation, there is a high mortality rate of up to 25% among patients on the transplant waiting list.24 Furthermore, transplantation for HCC in a noncirrhotic liver has no additional benefit of improving the patient's long-term liver function.
The recent literature reported the 5-year survival rates for transplantation in patients with small HCCs and cirrhosis to range from 60% to 79% and the 5-year disease-free survival rates to range from 45% to 60%.24-26 Our results showed that hepatic resection of a noncirrhotic HCC can be performed safely, and the 5-year overall survival rate of 70.0% and 5-year disease-free survival rate of 54.8% in this cohort of patients with potentially transplantable disease is comparable to those of the reported series of transplantation for small HCCs. The small differences in these data should be considered before suggesting liver transplantation to patients with noncirrhotic HCC. In terms of overall survival results, at present hepatic resection can be justified as the first-line treatment for small HCC in patients without cirrhosis. In this scenario, liver transplantation can be reserved as salvage treatment for patients with HCC who have recurrent disease after hepatic resection, as reported recently by Poon et al26 and Hess et al.28
In early stage (TNM stage I and transplantable HCC) disease, patients without cirrhosis had significantly better survival rates than patients with cirrhosis. The differences in the background factors in relation to tumor progression and detection and surgical factors may be the main reason for the better prognosis. Early detection of a small HCC in the noncirrhotic liver is important. For small HCCs originating in noncirrhotic livers, hepatic resection is a reasonable primary treatment associated with favorable 5-year overall and disease-free survival rates.
Corresponding author and reprints: Gar-Yang Chau, MD, MPH, Division of General Surgery, Department of Surgery, Taipei Veterans General Hospital, 201 Shih-Pai Rd, Section 2, Taipei, Taiwan 11217 (e-mail: firstname.lastname@example.org).
Accepted for publication November 3, 2003.