Chue-Shue Lee

Growth rate of asymptomatic hepatocellular carcinoma and its clinical implications.

The growth rate of 31 asymptomatic hepatocellular carcinomas (diameter less than or equal to 5 cm) discovered in 28 patients by a prospective screening program was determined by real-time ultrasonography over 36-860 days. Except for one tumor that shrank on follow-up, the doubling time ranged from 29 to 398 days, with a median of 117 days, an arithmetic mean of 136 days, and a geometric mean of 110 days. In 17 tumors with more than two measurements, the growth rate remained exponential in nine, declined in growth in seven, and showed an initial lag period in one. Doubling time correlated with initial tumor diameter but was independent of the patients age, sex, hepatitis B surface antigen status, tumor location, liver function tests, stage of liver cirrhosis, histologic type, or grade of malignancy. Although initial alpha-fetoprotein levels did not correlate well with growth rate, in 14 patients with an exponential increase of serum alpha-fetoprotein, the alpha-fetoprotein doubling time was closely related to the tumor doubling time. Based on the above data, the median detectable subclinical period of hepatocellular carcinoma was deduced to be 3.2 yr, and the suitable screening interval for its early detection in our area was 4-5 mo.

Early detection of hepatocellular carcinoma by real-time ultrasonography. A prospective study

For early detection of hepatocellular carcinoma (HCC), real‐time ultrasonography (US) was performed prospectively in 528 patients, including 236 with cirrhosis, 81 with chronic hepatitis, 168 asymptomatic hepatitis B surface antigen carriers, and 43 with a family history of HCC. Simultaneous measurement of serum alpha‐fetoprotein (AFP) level was also done. In addition, 233 patients had regular controls at 3‐ to 6‐month intervals, with an average follow‐up period of 1.4 years. On initial screening, a total of 17 patients were found to have HCC: 13 in the cirrhotic group, 3 in the HCC family group, and 1 in the asymptomatic carriers. Of these HCCs, 7 were smaller than 3 cm, 6 were between 3 to 5 cm, and 4 were larger than 5 cm. In patients with tumors smaller than 5 cm, the AFP levels were normal in 46.2%, between 20 to 400 ng/ml in another 46.2%, and only 7.6% were over 400 ng/ml. On follow‐up, another seven patients, all in the cirrhotic group, were found to have HCCs varying from 1.6 to 4.7 cm; three of them had normal serum AFP level. The authors conclude that real‐time US is more sensitive than AFP assay in early detection of HCC, and the high‐risk subjects should receive this procedure at regular intervals.

Serum α-Fetoprotein in the Early Stage of Human Hepatocellular Carcinoma

The serum alpha-fetoprotein levels of 17 patients with cancers less than or equal to 3 cm in size were studied using radioimmunoassay to determine alpha-fetoprotein response in the early stage of human hepatocellular carcinoma. The levels were normal in 6 patients (35%), and elevated to 645 +/- 1140 micrograms/L (mean +/- SD) in the remaining patients. The levels were not correlated with tumor size. In 10 surgically resected patients, 5 had elevated levels of alpha-fetoprotein that returned to normal after surgery. The levels and tumor sizes were serially observed for 3-26 mo in the remaining 7 patients not surgically treated. In 1 patient alpha-fetoprotein levels were persistently normal and in the other 6 patients, the levels tended to increase with a median doubling time of 60-75 days (range 30-223 days). Despite continued tumor growth, a spontaneous fall in serum alpha-fetoprotein levels was encountered in 4 patients; in 2 patients the levels even fell within the normal range. After this fall, the levels increased drastically in 4 patients. We concluded that in the early stage of hepatocellular carcinoma, serum alpha-fetoprotein level is frequently normal, and thus determination of serum alpha-fetoprotein levels only is not a reliable indicator in the early detection of human hepatocellular carcinoma. A spontaneous fall in the level of alpha-fetoprotein is not an uncommon finding in the early stages of this cancer and cannot be used to rule out the diagnosis of hepatocellular carcinoma.

Prognostic histologic features of resected small hepatocellular carcinoma (HCC) in Taiwan: a comparison with resected large HCC

The morphologic features and their prognostic significance were analyzed in 83 surgically resected hepatocellular carcinomas (HCC): 44 of them were of ⩽5 cm in diameter and 39 were larger. This study demonstrated a high prevalence of tumor capsule in both small and large HCCs, 86.4% and 84.6%, respectively. In small HCC, the capsule formation was significantly higher in the liver with cirrhosis (96.9%, or 31/32) than without (58.3%, or 7/12) (P < 0.003). In both small and large HCCs, the most important histologic parameter influencing the tumor recurrence was the liver invasion. In the small but not in the large HCC, the invasion through the capsule also correlated well with recurrence (P < 0.05). None of the 19 patients whose HCC were confined by a tumor capsule and did not invade the liver, had a recurrence during the 10 months or longer follow‐up period. The significantly more favorable outcome of smaller HCC was related to the lower frequencies of liver invasion (P < 0.001), portal vein involvement (P < 0.01), and satellite formation (P < 0.01). Tumor recurrence did not correlate with the tumor grading, presence of clear cells, liver cell dysplasia, or host inflammatory response. The HBsAg positivity in tumor cells was significantly higher in small (40.9%, or 18/44) than in large HCC (10.3%, or 4/39), suggestive of a gradual loss of the cytoplasmic expression of HBsAg in the tumor cells during the growth of HCC.

Tumor invasiveness and prognosis in resected hepatocellular carcinoma. Clinical and pathogenetic implications

In order to elucidate the biologic behavior of hepatocellular carcinoma (HCC), the long‐term prognostic impact of the pathologic features of 143 surgically resected HCC were studied. Seventy‐four were smaller than 5 cm in diameter (small HCC), and 69 were larger (large HCC). This study confirmed that tumor size was an important but not the only determining prognostic factor in HCC. Although cirrhosis could cause hepatic failure, patient mortality was mainly attributable to tumor recurrence, which, in turn, was strongly correlated with the invasive nature of HCC. Tumor invasion to the liver and the intraportal spread were very frequent and particularly extensive in large HCC. In both small and large HCC the noninvasive groups not only had high 4‐year actuarial survival (84.6% and 90%, respectively), but there was also no patient mortality from intrahepatic tumor recurrence. Therefore, it was concluded that invasiveness of an HCC is the most crucial factor in determining the long‐term outcome for the patient, and that the clinical course of resected HCC is predictable in a great majority of the cases. In our small HCC series only 2.4% of HCC were regarded as having true multicentric origin. These findings suggest that the majority of HCC occur unicentrically, and that multiplicity and tumor recurrence result mostly from intrahepatic dissemination. In both small and large HCCs invasive tumors were accompanied by high patient mortality from tumor recurrence even when the tumor was small, indicating that intrahepatic spread may start very early during the growth of HCC.

Clinical Management of Recurrent Hepatocellular Carcinoma

ObjectiveThe aim of this study was to evaluate the long-term benefits of the aggressive treatments with resection or transarterial chemoembolization (TACE) for recurrent hepatocellular carcinoma (HCC). Summary Background DataPrimary HCC is one of the most fatal malignancies in Taiwan. The result of resection for HCC remains unsatisfactory, primarily due to the high recurrence rate. To improve surgical results, recurrent HCC must be treated with aggressive resection or TACE. MethodsThe authors evaluated the results of repeated hepatic resection among 25 patients with recurrent HCC and of TACE among 12 patients with resectable recurrent HCC. The outcomes of an additional 64 patients with unresectable recurrent HCC were also evaluated. ResultsDuring the follow-up period from 2–112 months, 52% (13/25) of patients receiving repeat resection (group 1) were alive, whereas 42% (5/12) of patients receiving TACE (group 2) were alive. No perioperative deaths within 30 days after surgery occurred in the repeated resection group. The cumulative survival rates at 1,2,3, and 5 years after the first operation were 92%, 84%, 71.6%, and 65.1 % in group 1 and 83.3%, 75%, 75%, and 22.5% in group 2. The survival rates at 6 months and at 1,2, and 3 years after recurrence were 92%, 72%, 64%, and 44.8% in group 1 and 83.3%, 75%, 66.7%, and 48% in group 2.The survival of patients with unresectable recurrent HCC was much worse: 1 -, 2-, 3-, and 5-year survival after surgery was 57.8%, 29.8%, 15.5%, and 0%; and 6-month and 1-, 2-, and 3-year survival after recurrence was 46.5%, 29.2%, 12.5% and 7.8%. ConclusionsMore aggressive treatment with repeated hepatic resection can prolong survival time after recurrence of HCC in selected patients. However, TAGE can also achieve good results although it is not thought of as curative.

Detection of hepatitis B virus DNA in hepatocellular carcinoma: methylation of integrated viral DNA.

In order to determine whether integrated hepatitis B virus DNA sequences in primary liver tumours are methylated we have analysed tumour DNA digested with either MspI or HpaII restriction endonuclease by Southern hybridization. Our results demonstrate methylation in 11 of 17 tumour DNA samples. Where possible, we have examined the tumour tissues for expression of HBsAg and HBcAg using the indirect immunoperoxidase technique. One tumour was positive for both HBsAg and HBcAg and a second was positive for HBsAg alone. Both of these tumours were in the group in which methylation of integrated HBV DNA sequences could not be detected. We postulate that methylation of integrated HBV DNA sequences may influence HBV gene expression in hepatocellular carcinoma.

Evolution of expression of hepatitis B surface and core antigens (HBsAg, HBcAg) in resected primary and recurrent hepatocellular carcinoma in HBsAg carriers in Taiwan. Correlation with local host immune response.

Studies were conducted on the evolution of hepatitis B virus (HBV) surface and core antigens (HBsAg and HBcAg) in the tumors of both primary and recurrent hepatocellular carcinoma (HCC) in 27 HBsAg carriers; these were followed for up to 8 years after the resection of the primary tumor. Twenty‐seven primary and 34 recurrent tumors were included. HBV antigens were detected in the tumor of the primary HCC in ten cases (37%): six (22.2%) had both antigens (Group I) and four (14.8%) had HBsAg alone (Group II). The remaining 17 cases were negative for both antigens (Group III). Intrahepatic tumor recurrence occurred in 17 cases; both HBcAg and HBsAg were found in the recurrent HCC in four of five HBcAg‐positive cases (Group I). In contrast, HBcAg was detected in none of the other 12 cases (Group II, 0 of one; Group III, 0 of 11), and HBsAg in only one (Group II, 0 of one; Group III, one of 11), p< 0.03 and p < 0.02, respectively. Groups I, II, and III had extrahepatic recurrence in two, four, and seven cases, respectively. HBcAg was detected in none, while HBsAg was found in only one case (7.7%). The frequent detection of both antigens in the primary HCC and even in the intrahepatic recurrences suggests that HBV replication in HCC may occur more commonly than previously perceived, especially in the small HCC. Failure to detect HBV antigens in the extrahepatic recurrences suggests that the switch‐off of the viral gene expression, particularly the core gene, may be an event related to the extrahepatic growth of HCC. HBV antigen expression in HCC is associated with more evident lymphocyte infiltration; this local host immune response may in turn result in a negative selection and expansion of the antigen‐negative HCC cell clones. This suggestion is in accord with the fact that HBV antigens, particularly HBcAg, are rarely detected in advanced HCC.