Implication of Frequent Local Ablation Therapy for Intrahepatic Recurrence in Prolonged Survival of Patients With Hepatocellular Carcinoma Undergoing Hepatic Resection

Abstract

Objective:

By comparing cohorts in 2 exclusive time frames, the factors that affected the surgical outcomes of patients with hepatocellular carcinoma (HCC) are presented.

Summary Background Data:

Reportedly, survival results of patients with HCC who underwent hepatectomy have improved in recent years. However, the major factors contributing to these favorable outcomes have not been fully explained.

Methods:

Between January 1985 and December 2000, 610 patients with HCC underwent liver resections as a primary and curative resection. They were categorized into 2 groups according to the year in which the surgeries were performed: before 1990 (n = 212; early group); and after 1991 (n = 398; late group). Clinicopathologic data, survival data, type of recurrence, and treatment of intrahepatic recurrence were compared between the 2 groups.

Results:

Clinicopathologic data were almost identical between the groups except for age, blood loss, and duration of surgery. The overall survival rate was significantly better in the late group compared with the early group (58.0% vs. 39.1% at 5 years, P < 0.0001). By contrast, disease-free survival remained unchanged (27.8% vs. 26.2% at 5 years, P = 0.2887). The most common type of recurrence was intrahepatic relapse, and there was no difference in the rate and the type of recurrence between the 2 groups. The 5-year survival rate after recurrence was increased in the late group (21.8% vs. 11.6%, P = 0.0002). Stratified analysis by the type of initial recurrence revealed that better survival in the late group was achieved only in solitary intrahepatic recurrences, not in multiple intrahepatic or extrahepatic recurrences. Changes in modality of treatment of recurrence were observed only in the management of solitary intrahepatic recurrences, where percutaneous ablation therapies were more frequently applied with new ablation techniques. Patients that had undergone ablation therapies in the late group had better postrecurrent survival than those in the early group. Multivariate analysis showed that presence of local ablation therapies was an independent favorable prognostic factor only in the late group.

Conclusions:

Significant improvements in outcomes were achieved in patients with HCC who underwent curative liver resections. Percutaneous ablation therapy for intrahepatic recurrence was considered to be a major contributory factor for improving survival after recurrence, as well as for overall survival.

Recently, notable advances have been made in the surgical management of patients with hepatocellular carcinoma (HCC). Several studies have reported improved outcomes of patients with HCC who have undergone liver resections. The improved outcomes include not only decreases in operative mortality and morbidity, but also favorable long-term results.^1–4

Various factors might have contributed to these improved outcomes, including early detection of subclinical HCCs through screening programs for patients at high-risk for HCC.^5–8 The development of imaging tools such as ultrasonography (US),⁹ computed tomography (CT),^10,11 and magnetic resonance imaging^12–14 have also contributed to early detection. The establishment of operative guidelines for patients with poor liver function, advances in surgical techniques, and improved perioperative management have reduced the risk of postoperative mortality.^15,16 Even after tumors recur, rehepatectomy¹⁷ or nonsurgical treatments such as transarterial chemoembolization (TACE),¹⁸ or percutaneous ablation therapy¹⁹ have presumably helped with long-term survival.

Although all the factors mentioned above appear to have influenced positive outcomes to a certain extent, it is yet unclear which factors have had the greatest impact on long-term mortality and morbidity. This is due in part to the limited number of patients in a single institution or medical center receiving these therapies, but also to the evolving differences in the criteria for surgery between institutions and medical centers over time. The present study was designed to describe the results of a series of liver resections for HCC over a period of 16 years in a single center specializing in hepatobiliary surgery and to discuss major factors that influenced the long-term outcomes of patients with HCC.

METHODS

Patients

Study subjects are 610 patients with HCC who underwent liver resections as an initial treatment in the Department of Gastroenterological Surgery at Kyoto University Hospital, Kyoto, Japan, between January 1985 and December 2000. Patients with intrahepatic metastases who were treated with ethanol injection, microwave coagulation therapy (MCT), or radiofrequency ablation (RFA) during surgery were excluded from the study. Inpatient hospital deaths were also excluded. Histologic diagnoses of HCC were confirmed in all 610 patients. Patients were then categorized into 2 groups according to when they underwent hepatectomy; the early group (from January 1985 to December 1990; n = 212); and the late group (from January 1991 to December 2000; n = 398). These time intervals were chosen because they represent the period of time before and after the introduction of more sophisticated operative techniques developed for living donor liver transplantation and the onset of screening programs for hepatitis C antibody positive patients.

Preoperative Evaluations

The preoperative diagnoses of HCC were based mainly on US, CT, and serum alpha-fetoprotein and protein induced by vitamin K absence or antagonist-II (PIVKA II) (available only after 1990) levels. Tumor stage, liver damage, and surgical procedures (Hr) were defined according to the General Rules for the Clinical and Pathologic Study of Primary Liver Cancer[r], second edition.²⁰

Follow-up Strategies and Treatments for Recurrent Disease

All patients received follow-up at Kyoto University Hospital or other hospitals. Serum alpha-fetoprotein and PIVKA II (if available) levels were measured monthly and CT or US were performed at least every 3 months. If tumors recurred, appropriate treatment was initiated according to the number, size, and location of the tumors. Basically, percutaneous local ablation therapy, including percutaneous ethanol injection therapy (PEIT), MCT, or RFA, was chosen for small tumors (up to 2 cm) and for small numbers (3 or less) of tumors. TACE was chosen mainly for multiple (3 or more) recurrences. Rehepatectomy was considered for large tumors or tumors difficult to control by nonsurgical treatments.

Statistical Analysis

Continuous variables were expressed as the mean ± the standard deviation and compared by an unpaired t test. Categorical variables were compared using a χ² test. Survival curves were generated using the Kaplan-Meier method and compared by log-rank test. Cox proportional hazard model was used for multivariate analysis. A P value of less than 0.05 was considered statistically significant.

RESULTS

Patients' Characteristics

Characteristics of patients in this study are summarized in Table 1. Patients were older in the late group than in the early group (59.7 ± 8.6 years of age in the early group; 61.9 ± 9.5 years of age in the late group; P = 0.0076). There were no significant differences in gender and the number of patients who had positive hepatitis B surface antigen in either group. Hepatitis C antibody was positive in approximately 60% of the late group patients; however, it was evaluated only after 1990 and therefore was not compared between the groups. Liver function was comparable between the groups, as determined by the results of an indocyanine green clearance test or by liver damage. There was no significant difference in surgical procedures (P = 0.6168) between the groups. However, the operative time was shortened and the intraoperative blood loss was decreased in the late group (P = 0.0013 and 0.0395, respectively). Tumor characteristics, including preoperative alpha-fetoprotein levels, tumor size, the number of tumors, the presence of vascular invasion, tumor stage, and the proportion of poorly differentiated tumors, were all comparable between the 2 groups.

TABLE 1. Clinicopathological and Operative Data of Patients Who Underwent Liver Resection for HCC

Overall Survival

The overall survival curves for both groups are depicted in Figure 1. In the early group, the median survival time (MST) was 3.9 years, with 1-, 3-, and 5-year survival rates of 84.1%, 58.0%, and 39.1%, respectively. In the late group, the MST was 6.0 years, with 1-, 3-, and 5-year survival rates of 90.5%, 72.1%, and 58.0%, respectively. There was a statistically significant difference between the survival rates of the 2 groups (P < 0.0001).

FIGURE 1. Overall survival curves of patients with hepatocellular carcinoma who underwent curative hepatic resection between 1985 and 2000. The differences in overall survival rates were statistically significant between the early group and the late group (P < 0.0001).

Survival data of patients stratified by predictive factors for outcomes are shown on Table 2. Varying degrees of survival improvement were observed in nearly all strata. Regarding tumor stage, survival advantages in the late group patients were apparent in moderately advanced disease. Differences in the 5-year survival rates between the early and the late groups were more than 20% in stage II (25.6%) and III (20.2%) patients but were limited to approximately 10% in stage I and IV-A. While differences of 5-year survival rates between the early and the late groups for patients with either liver damage A and B were significant, these differences were more prominent for patients with liver damage A (25.0%) than with liver damage B (8.8%). After stratifying patients by tumor size, number of tumors, presence of vascular invasion, or tumor differentiation, analysis revealed that survival advantages in the late group were significant except in patients with 1) tumors exceeding 10 cm, 2) tumors with vascular invasion, or 3) poorly differentiated tumors. Whether solitary or multiple tumors were present, survival was longer in the late group.

TABLE 2. Overall Survival of Patients Who Underwent Curative Resection for HCC Stratified by Predictive Factors for Survival

Disease-Free Survival

Figure 2 shows disease-free survival curves. In contrast to the improvement of overall survival, disease-free survival was similar in both groups. The disease-free survival rates at 1, 3, and 5 years after surgery were respectively 58.7%, 37.0%, and 26.2% in the early group and 67.6%, 38.7%, and 27.8% in the late group (P = 0.2887). Even after stratifying patients according to tumor stage or liver damage, there were no strata in which statistically significant improvements of disease-free survival were present (Table 3). Since the serological status of hepatitis virus is an important factor influencing disease recurrence after hepatectomy in patients with HCC,²¹ we examined whether disease-free survival rate differed by type of hepatitis virus (Fig. 3A). In patients with a non-B non-C or a positive hepatitis B surface antigen, the disease-free survival rates gradually decreased in the course of the follow-up period and plateaued 4 years after surgery. By contrast, disease-free survival rate in hepatitis C antibody-positive patients continued to fall 4 years after surgery. Even when patients only with positive hepatitis B surface antigen were analyzed, there were no statistically significant differences in the disease-free survival rates between the early and the late groups (Fig. 3B).

FIGURE 2. Disease-free survival curves of patients with hepatocellular carcinoma who underwent curative hepatic resection. The differences in disease-free survival rates were not statistically significant between the early group and the late group (P = 0.2887).

TABLE 3. Disease-Free Survival of Patients Who Underwent Curative Resection for HCC Stratified by Stage or Liver Damage

FIGURE 3. A, Disease-free survival rates depending on the serological status of hepatitis. Disease-free survival rates of all patients decreased until postoperative year 4. After postoperative year 4, disease-free survival rates of patients with hepatitis C continued to decrease while those of the other patients plateaued. HBV, patients with positive hepatitis B surface antigen; HCV, patients with positive hepatitis C antibody; nB nC, patients with neither hepatitis B virus nor hepatitis C virus. B, Comparison of disease-free survival rates of patients with hepatocellular carcinoma with positive hepatitis B surface antigen between the early group and the late group. The differences between the 2 groups were not statistically significant (P = 0.5947).

Recurrence Rate and Type of Initial Recurrence

There were 166 disease recurrences in the early group (78.3%) and 299 recurrences in the late group (75.1%). Type of initial recurrence was not different between the groups as summarized in Table 4. Recurrent tumors were localized in the remnant liver in 139 patients in the early group (83.7%) and 256 in the late group (85.6%), with similar proportion (P = 0.8222). Among these patients, the ratios of solitary to multiple recurrences were also similar between the groups (P = 0.1445).

TABLE 4. Type of Initial Recurrence

Survival After Recurrence

Concurrent with improved overall survival and unchanged disease-free survival, survival after recurrence was significantly longer in the late group; 1.7 years versus 2.5 years in MST and 11.6% versus 21.8% in 5-year survival rate (P = 0.0002) (Fig. 4A). Stratified analysis by type of initial recurrence revealed that significant improvement in survival rate after recurrence was obtained only in solitary intrahepatic recurrence (P = 0.0097), not in multiple intrahepatic recurrence (P = 0.0940) or extrahepatic recurrence (P = 0.2975) (Fig. 4B–D).

FIGURE 4. Survival rates after recurrence of patients with hepatocellular carcinoma by type of initial recurrence. A, Overall. B, Solitary intrahepatic recurrence. C, Multiple intrahepatic recurrence. D, Extrahepatic recurrence. The differences in survival rates after recurrence were statistically significant between the 2 groups in overall (P = 0.0002) and solitary intrahepatic recurrence (P = 0.0097).

Treatment of Intrahepatic Recurrence

The most commonly used treatments for intrahepatic recurrence were rehepatectomy, TACE, and percutaneous ablation therapy including PEIT, MCT, and RFA. Table 5 shows how these therapies were applied in each type of recurrence. The number of patients who received each therapy is shown in the left column and the total number of cases in the right column. Rehepatectomy was performed on 18 patients (10.8%) in the early group and 37 patients (12.4%) in the late group, with almost equal frequency (P = 0.6242). TACE was performed with similar frequency in both groups: 289 times on 115 (69.3%) patients in the early group and 562 times on 204 (68.2%) patients in the late group, respectively. The mode of TACE changed over time as indicated by increased frequency of selective TACE in the late group. On the other hand, percutaneous ablation therapy was used more frequently in the late group than in the early group (152 times on 43 [25.9%] patients vs. 513 times on 146 [48.8%] patients, P < 0.0001). It was not only used on more patients more frequently, but new types of ablation therapy were used as well. MCT and RFA were performed almost exclusively on the late group patients. When categorized by the type of initial recurrence, differences between the 2 groups in therapeutic interventions were only seen in the treatment of solitary intrahepatic recurrences. There was no change in the type of treatment of multiple recurrence or extrahepatic recurrence. These results clearly indicate that treatment strategies for solitary intrahepatic recurrence have been significantly altered between early and late periods.

TABLE 5. Treatment for Recurrence by Type of Initial Recurrence

We further examined whether these changes in the way of treatment of recurrent disease truly impacted on survival after recurrence. A total of 465 patients who had recurrence were divided by the presence or absence of ablation therapy and survival rates after recurrence were compared between the early and the late groups in each subset (Fig. 5). In the subset that had no ablation therapy, there was no improvement of survival, suggesting that technical advances in TACE or rehepatectomy, if any, had no or little impact on survival (Fig. 5A, P = 0.5512). On the other hand, there was clear improvement in the subset that had undergone ablation therapy (Fig. 5B, P = 0.0045), indicating that progress in ablation therapy produced favorable survival results in the late group. Finally, multivariate analysis was performed to determine the significance of variances that potentially affected postrecurrent survival in the early and the late group (Table 6). Presence of rehepatectomy was an independent prognostic factor in both groups, as well as type of initial recurrence and interval between surgery and initial recurrence. On the other hand, presence of local ablation therapies was an independent positive prognostic factor only in the late group (P < 0.0001 in the late group and P = 0.1346 in the early group).

FIGURE 5. Survival rates after recurrence of patients with hepatocellular carcinoma who (A) have not undergone and (B) have undergone local ablation therapies for the treatment of recurrence. Significant improvement of postrecurrent survival was achieved only in the patients who have undergone ablation therapies.

TABLE 6. Multivariate Analysis of Variables Potentially Predictive of Survival After Recurrence in HCC Patients

DISCUSSION

This study demonstrates that the long-term prognosis for patients with HCC who have undergone hepatic resection has improved in the last decade. Overall survival was significantly better for patients in the late group than in the early group. Generally, there are several potential factors that might influence the medical outcome: 1) selection of patients, 2) operative mortality, 3) recurrence rate, 4) postoperative surveillance, and 5) management for recurrence. To focus the analysis on long-term results, all postoperative inpatient deaths were excluded from our study so that the decrease in operative mortality from the late group did not contribute to the improved long-term survival rate of the late group. Several studies have demonstrated that the increased proportion of HCC detected in early stage led to favorable outcomes;^2,4 however, this was not the case with the present study. Patient demographics, medical history, and tumor characteristics did not differ between the early and the late groups as determined by serological status of hepatitis, liver function, stage, size, multiplicity of tumors, presence of vascular invasion, and tumor differentiation. Thus, the improvement of overall survival in our series is substantial and is neither due to the selection of patients nor to a decrease in operative mortality.

In contrast to overall survival, disease-free survival rates were almost equal between the groups. Thus, in our series, improved long-term survival could not be attributable to a decrease in the recurrence rate. While these results are disappointing, they are predictable because removal of tumors by surgical resection is only a local therapy; it neither cures the occult metastasis nor prevents metachronous carcinogenesis. In contrast to our study, however, a study by Poon et al reported that disease-free survival rate after surgical intervention had improved.³ Different characteristics between their study patients (80% had hepatitis B) and our study patients (60% had hepatitis C) might account for the contrast because HCCs related to hepatitis C are more prone to recur due to frequent multicentric metachronous carcinogenesis.²¹ Indeed, in our study, recurrence patterns were different between patients with hepatitis B and those with hepatitis C. However, even when only hepatitis B patients were analyzed, the disease-free survival rates did not differ between the early and the late groups in our study. Disease-free survival curves in Poon's study showed a clear difference between groups only in the initial few months after surgery. The early difference in the disease-free survival rates in their study might reflect the presence of small intrahepatic metastases missed at the time of surgery or early postoperative deaths in the early group instead of a true decrease in the recurrence rate in the late group.

Patients in both early and late groups had similar presentations and initial surgical interventions, and also had similar types of recurrences; however, survival time after recurrence was better in the late group. As HCCs recur mostly in the liver²² and more than 80% of recurrences were localized in the liver in the present series, it is reasonable to hypothesize that changes in treatment of intrahepatic recurrence resulted in more positive outcomes in the late group. Notably, both quantity and type of percutaneous ablation therapy changed dramatically as indicated by the increase in the number of patients who received it and by the advent of MCT or RFA in the latter period. Technical advances in percutaneous ablation therapy such as CT-guided ablation might account for its more frequent use in the late group.^23,24 In addition, lower local recurrence rate and better long-term survival rate provided by MCT or RFA presumably increased its application in preference to other treatment.^25,26 Notably, the more frequent application of ablation therapies was present almost exclusively in the treatment of solitary recurrences; little changed in the management of multiple recurrences or extrahepatic recurrences. The fact that improved survival was achieved only in cases of solitary recurrence strongly implies a direct causal relationship between advances in therapy and prolonged survival after recurrence. This relationship is supported by our analysis on patients who had undergone ablation therapy (Fig. 5B), demonstrating that patients in the late group had superior survival than in the early group. It is further supported by our multivariate analysis (Table 6) in which presence of ablation therapy was an independent positive prognostic factor only in the late group. On the other hand, technical advances in TACE seemed to have had no or little impact on the improved survival in the late group as evidenced by the analysis on patients who did not undergo ablation therapy as a treatment of recurrence (Fig. 5A).

This study demonstrates that, although overall survival after surgery has improved significantly since 1991, disease-free survival has not changed in patients who underwent hepatic resection for HCC. The prevention of multicentric metachronous recurrence in patients with virus-induced hepatitis by antiviral therapy may represent our best hope to decrease the recurrence rate and further improve overall survival rate.^27,28 Survival results after solitary intrahepatic recurrence have improved mainly due to the more frequent application of percutaneous ablation therapies. On the other hand, our study indicates that the treatment strategy and poor prognosis for multiple or extrahepatic recurrences of HCC has remained largely unchanged. Identifying improved treatment of these types of recurrence is required.