Widespread use of ultrasonography (US), computed tomography (CT), nuclear magnetic resonance imaging, angiography, and α-fetoprotein (AFP) and des-γ-carboxy (abnormal) prothrombin measurements has allowed the increasingly earlier detection of hepatocellular carcinoma (HCC).1-4 Nevertheless, postoperative recurrence remains frequent.5-9 Generally, long term survival after hepatic resection in patients with HCC continues to be difficult because of the high incidence of postoperative recurrence.5-9 Therefore, the number of reports of long term surviving patients (e.g., ≥10 years) after hepatic resection is quite small. To our knowledge thus far only 1 report form China has referred to patients surviving ≥10 years after hepatic resection,10 whereas no other similar reports have been published. Therefore, the characteristics of such patients remain obscure.
The aim of this study was to clarify the clinicopathologic features of Japanese patients with HCC surviving ≥10 years after undergoing hepatic resection.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
Between January 1971 and April 1987, 142 patients underwent hepatic resection at the Department of Surgery II, Kyushu University. Thirty-nine patients who died of surgical morbidity (surgical mortality rate: 27.5%) were excluded from this study. In 103 patients who did not die of surgical mortality, the survival rates were 83% at 1 year, 41% at 3 years, and 29% at 5 years after curative surgery. Among the 103 patients who underwent hepatic resection between January 1971 and April 1987, 12 patients (11.7%) survived >10 years after hepatic resection (overall survival among 142 patients, including those who died of surgical mortality, was 7.7%). The mean age of the patients was 52 years (range, 41-72 years); 10 patients were male and 2 were female.
The follow-up of the patients after hepatectomy in our institute was as follows. All patients were followed closely after surgery at regular 1-month intervals. Each follow-up visit included a physical examination, blood chemistry tests, and measurement of AFP and des-γ-carboxy (abnormal) prothrombin. Between 1976 to 1978, liver scans using 99m sulphur colloid were performed. After US and CT became available in 1979, liver scans were performed every 3 months. When recurrence of HCC was suspected, the patients were readmitted for angiography and/or magnetic resonance imaging.
The 103 patients were divided into 2 groups: 12 patients who survived >10 years after hepatic resection (10-year survivors) and 91 patients who died within 10 years after hepatic resection (nonsurvivors). Using these patients, such factors as clinicopathologic background and surgical procedures were compared between the two groups. Hepatitis C virus (HCV) mRNA levels were determined by a branched DNA probe assay.11 The pathologic lists of extensive examination were available for 91 patients. In six patients, the tumor was completely necrotic, most likely due to preoperative therapy. Therefore, in these six patients, histologic differentiation could not be determined.
The recurrent status, therapeutic modalities for recurrent tumors, and the occurrence of any extrahepatic primary tumors in 10-year survivors were examined.
All data were expressed as the mean ± standard deviation. We analyzed the association between each of the variables studied and liver failure, using Student's t tests with continuous variables and the chi-square test of independence with categoric variables. A P value < 0.05 was considered significant.
- Top of page
- PATIENTS AND METHODS
The incidence of long term survivors was 6.4% (2 of 31) between January 1971 and December 1981, whereas it increased to 13.9% (10 of 72) between January 1982 and April 1987.
Table 1 compares the clinicopathologic factors between the 10-year survivors and nonsurvivors. Liver function, as indicated by the indocyanine green retention test at 15 minutes (ICGR15) and the serum albumin level, in 10-year survivors was significantly better than in nonsurvivors. With regard to virus markers, 5 of 12 patients (41.7%) were positive for hepatitis B surface antigen (HBs-Ag) and the incidence of hepatitis B virus (HBV)-related HCC was significantly higher than in nonsurvivors (P = 0.020). Of the 10-year survivors, 4 of 12 patients were positive for HCV antibody, whereas 1 patient showed no positivity for the HCV antibody or HBs-Ag (Table 2). Among the four patients with positivity for the HCV antibody, the HCV RNA concentration was measured by branched DNA probe assay. The HCV RNA concentration was < 0.5 in 3 patients and 1.72 in 1 patient; thus, the HCV RNA concentration in the 10-year survivors was low. No statistically significant risk factors for the pathologic tumor findings were identified.
Table 1. A Comparison of Clinicopathologic Features between 10-Year Survivors and Nonsurvivors after Hepatic Resection
|Variables||10-year survivors (n = 12)||Nonsurvivors (n = 91)||P values|
|Host|| || || |
|Age (yrs)||53 ± 9||56 ± 8||NS|
|Gender (M:F)||11 :2||70 :13||NS|
|HBs Ag (+)||5 (41.7%)||11 (12.1%)||0.02|
|Platelet (104/μL)||12.4 ± 3.3||14.0 ± 8.2||NS|
|Alb (g/dL)||4.0 ± 0.3||3.7 ± 0.5||0.04|
|T bili (mg/dL)||0.8 ± 0.2||0.9 ± 0.5||NS|
|sGPT (IU/L)||66 ± 36||93 ± 79||NS|
|ICGR15 (%)||12.8 ± 6.4||20.4 ± 11.6||0.02|
|Cirrhosis||9 (75.0%)||63 (79.7%)||NS|
|Tumor|| || || |
|Size (cm)||3.1 ± 1.9||4.4 ± 2.9||NS|
|fc||10 (83.3%)||76 (83.5%)||NS|
|fc-inf||7 (70.0%)||60 (78.9%)||NS|
|im||2 (15.3%)||23 (29%)||NS|
|Histologic differentiation (well/moderate/poor)||1 :10:1||11 :48:17||NS|
|AFP (ng/mL)||1578 ± 4912||6309 ± 33065||NS|
Table 2. Viral Status in 12 10-Year Survivors
|Viral status|| || |
|HBs Ag (+)||5 (38.5%)|| |
|HCV (+)||4 (30.8%)||DNA probe < 0.5 3|
| || ||≥ 0.5 1|
|NANBa||2 (15.4%)|| |
|HBs Ag (-), HCV (-)||1 (7.7%)|| |
Figure 1 (40K) shows the clinical course of the 10-year survivors. In 4 of 12 patients (33.3%), no recurrence was observed after hepatic resection (Cases 2, 6, 7, and 8). Solitary recurrence was observed in five patients, whereas multiple recurrence was observed in three patients. The mean disease free period in patients with recurrence was 66 months (range, 26-105 months) after the first hepatectomy. At last follow-up the longest survivor was still alive without disease 20 years and 1 month after the first hepatic resection (Case 1). Five of the eight patients with a solitary recurrence underwent a second resection. Among the five patients who underwent hepatic resection, four patients had no recurrence after the second resection (Cases 1, 4, 5, and 10). In one patient who underwent a second hepatic resection, a solitary recurrence again occurred and thus a third hepatic resection was performed (Case 11). In the patients with multiple recurrences, one patient underwent lipiodolization and not recurrence had occurred at last follow-up (Case 9), one patient refused further therapy because of her advanced age and died of disease 12 years and 2 months after the first hepatic resection (Case 3), and the third patient underwent lipiodolization (Case 12).
Figure 1. Clinical course of the 12 10-year surviving patients. SR: second resection; TR: third resection; LPD: lipiodolization; +: positive; -: negative.
Download figure to PowerPoint
In four patients, extrahepatic primary tumors were observed, including gastric carcinoma, urinary bladder carcinoma, rectal carcinoma, and breast carcinoma; all tumors appeared after hepatic resection.
- Top of page
- PATIENTS AND METHODS
The overall incidence rate in the 10-year survivors after hepatic resection was 7.7% of the curatively resected cases. The incidence of long term survivors has increased in the later observation period. This may be because the preoperative and intraoperative diagnoses of small intrahepatic metastases have improved,2, 12 and because the number of patients with occult remnant tumors at the time of surgery also have decreased. Furthermore, the incidence of surgical mortality was 27.5% in this study, which is extremely high. Nevertheless, this figure improved substantially during this 10-year period.13 In our clinic the surgical mortality was 29% between September 1981 and March 1985, whereas it was 1.9% between April 1985 and December 1988. Therefore, the incidence of 10-year survivors is expected to increase in the next 10-year period.
Such host factors as viral status and liver function appeared to be more important than tumor factors for 10-year survival. As a result, the 10-year survivors had better liver function than the nonsurvivors. Good liver function clearly is necessary for long term survival after hepatic resection. Furthermore, Adachi et al.14 reported that a serum albumin level of ≤3.7 g/dL was associated significantly with the postoperative recurrence of HCC after hepatectomy. Ikeda et al.15 also reported that a higher ICGR15 was associated independently with the appearance rate of HCC in cirrhotic patients. These findings thus indicate that both primary HCC and second primary tumors may arise in patients with advanced cirrhosis.
The incidence of HBV-related HCC in 10-year survivors was significantly higher than in nonsurvivors. Previously, the data from our clinic showed that the late recurrence rate of HBV-related HCC tended to be lower than that of HCV-related HCC.16 Adachi et al.14 recently reported that inflammatory activity in the liver parenchyma was associated independently with intrahepatic recurrence, especially multicentric occurrence of HCC, after liver resection. The hepatitis activity of the liver parenchyma in HCV-related HCC is significantly higher than in HBV-related HCC.16 Therefore, in patients with HBV-related HCC, multicentric occurrence of HCC17, 18 may occur rarely. In addition, in the 10-year survivors with the HCV antibody, serum HCV RNA levels tended to be low. The mechanism for this occurrence remains unknown. Several studies showed a progressive increase in the serum HCV RNA levels in patients with more advanced liver disease, such as chronic active hepatitis and cirrhosis.19-21 The low concentration of HCV RNA may be related to low hepatitis activity and the preservation of liver function even after a 10-year observation period. Another reason may be because the incidence of multicentric occurrence of HCC is lower in patients with low concentrations of serum HCV RNA. Mazzella et al.22 suggested that interferon therapy for patients with HCV-related cirrhosis lowered the incidence of HCC occurrence. In addition, several studies showed that the incidence of postoperative recurrence of HCC after hepatectomy is higher in patients with high hepatitis activity than in patients with low hepatitis activity.6, 23, 24 These facts may be related to the low concentration of serum HCV RNA in the 10-year survivors with the HCV antibody. However, further examination is needed.
In spite of the promising results for long term survivors reported in this study, the recurrence of HCC was frequent, 8 of the 12 patients had recurrence. As the optimal therapeutic procedure for recurrent tumors, our first choice for treatment is a repeat hepatectomy.25-28 We previously reported that a repeat resection could contribute to improved survival after hepatic resection. Therefore, in this study, a second hepatic resection was performed in five patients with a solitary recurrent HCC. A surprising finding was that four of five patients who underwent a second hepatic resection had no recurrence after the second resection. This suggests that HCC also could be cured by a second resection in cases of solitary recurrence. In patients with multiple recurrences, lipiodolization, a targeting chemotherapy for HCC,29 was performed. In one patient, recurrent tumors vanished after lipiodolization. As we reported previously,30 lipiodolization has prolonged survival rates in postoperative patients with recurrent HCC.
These results suggest that postoperative follow-up and early detection of recurrence may be important and thus may contribute greatly to an improved survival. We believe our strict follow-up system thus has made the early detection of recurrence possible. Another question is when clinical follow-up may be terminated. Because the longest disease free survival in patients in the current study with recurrence was 106 months, we believe that clinical follow-up after hepatic resection should continue for the remainder of the patient's life.
In 4 of 12 10-year survivors (33.3%), extrahepatic primary tumors also were observed. This incidence was extremely high, and most likely was due to the long term follow-up period. Recently, the data from our clinic have shown that extrahepatic primary tumors are not as rare as believed previously.31 Therefore, especially in long term surviving HCC patients, strict follow-up, as well as screening for the most common malignancies (such as gastric carcinoma) is strongly called for.
For patients with good liver function who are HBs-Ag positive and have low concentrations of serum HCV RNA levels, long term survival after hepatectomy can be expected. However, because tumor recurrence is common, postoperative follow-up is important. Finally, aggressive therapy for recurrence, including repeat resection when necessary, is recommended.