Search

Background/aims: Although hepatocellular carcinoma patients who meet the Milan criteria are optimal candidates for liver transplantation, most such patients in Japan have been treated without liver transplantation., Methodology: In this retrospective analysis, the patient selection criteria were (1) admission between 1992 and 2005, (2) fulfillment of the Milan criteria, (3) classification within the Barcelona Clinic Liver Cancer stages A1-A4, and (4) no previous anticancer treatment., Results: Of 451 patients who met the selection criteria, 162 underwent hepatectomy. The proportion of patients who underwent hepatectomy was 58% of 106 with stage A1 and 29% of 345 with stages A2-A4. For patients with stages A2-A4, the survival probability after hepatectomy at 3, 5, and 7 years was 89%, 70%, and 61%, respectively. There were no significant differences in survival time between stages A1 and stages A2-A4 after hepatectomy. Among patients with Child-Pugh scores of 5 and 6 in stages A2-A4, 51% and 29% underwent hepatectomy, respectively., Conclusions: Hepatectomy may be an appropriate first-line treatment option for patients with stages A2-A4 who meet the Milan criteria, when they have a good hepatic reserve and a long waiting time for liver transplantation.

We report a patient with gallbladder cancer and multiple liver and lymph node metastases which completely responded to S-1, an oral fluoropyrimidine anticancer drug. The patient was enrolled in the "Late phase II study of S-1 in patients with advanced biliary tract cancer". A partial response was confirmed after the first two courses of S-1 monotherapy, and anti-tumor efficacy was finally evaluated as a complete response after the 5th course. S-1 was discontinued after the 18th course because of long CR, and the patient has been alive and disease-free for more than 3 years. Grade 2 anorexia, diarrhea, hand-foot skin reaction, and tasty disturbance were observed, but no grade 3 or more toxicities were noted. S-1 appeared to be effective against advanced gallbladder cancer and showed excellent tolerability.

Background: Few studies have compared the prognostic impact of reduction hepatectomy (RH) for advanced hepatocellular carcinoma (HCC) with that of non-surgical treatment or curative hepatectomy., Patients and Methods: The treatment outcome of 30 RH patients was compared to that of two control groups: control group A, including 71 patients who underwent curative hepatectomy; control group B, including 106 patients who did not receive definitive local therapy or best supportive care., Results: In patients with tumor extension in >50% of the liver, 1-year survival rates for patients according to treatment (RH, control A and B) were 50%, 90% and 11% and 3-year survival rates were 42%, 60% and 0%, respectively. There was no significant difference between RH and control A, but there was a significant difference between RH and control B., Conclusion: RH could be recommended to patients with multiple advanced HCC extending to >50% of the whole liver.

Background: The prognosis of advanced hepatocellular carcinoma (HCC) remains poor, particularly in patients with tumor thrombi (TT) in the major vessels., Patients and Methods: From July 1992 to October 2004, 161 patients diagnosed as having advanced HCC with major vascular involvement were seen consecutively at our hospital. Among these patients, 32 (20%) underwent surgical resection [16 complete resection (CR), 16 reductive resection (RR)]. Eighteen patients (11%) received radiotherapy (RT), 73 (45%) underwent transcatheter arterial chemoembolization (TACE) or transcatheter arterial infusion chemotherapy (TAI), 8 (5%) with distant metastases received systemic chemotherapy, and 30 (19%) received palliative therapy., Results: Excluding the CR group, the patients in the RR group had a higher 1-year survival rate than the other treatment groups. However, there was no significant difference in the overall survival rates of the RR, RT, and TACE/TAI groups. When we evaluated prognostic factors to clarify the indications for RR in the multidisciplinary treatment of patients with advanced HCC with TT, prothrombin activity (PA) was identified as a significant independent preoperative factor for overall survival in the RR group. The survival rate in patients with PA of < or = 78% was significantly lower than that of patients with PA of > 78% (P = 0.0004). The median survival time of patients with serum PA of > 78% who underwent RR was 13.9 months and that of patients who underwent CR was 9.1 months, with no survival difference between the groups., Conclusion: In advanced HCC with major vascular involvement, patients who had RR with PA of greater 78% achieved a similar survival to those who had CR. The surgeon should still proceed with RR in those patients with serum PA of > 78% if CR does not seem feasible on preoperative evaluation.

Background: Radiotherapy is often used to treat patients with unresectable advanced hepatocellular carcinoma (HCC). The present study examines the nature and frequency of adverse events with respect to liver function in such patients after radiotherapy., Methods: Forty-six patients with HCC who underwent radiotherapy were retrospectively examined. Radiotherapy was applied using coplanar 2-3-beam arrangements to a target dose of 50 Gy/5 weeks. The adverse hepatic events were evaluated according to the National Cancer Institute Common Toxicity Criteria and the Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer Late Radiation Morbidity Scoring Scheme during the acute phase and the late phase by following the patients for up to 1 year. The influence on survival by adverse hepatic events and other factors was analyzed., Results: The full irradiation dose of 50 Gy was given to 40 patients (87.0%). Grade 3 or 4 toxicity was observed in 18 (39.1%) within 3 months after radiotherapy and in 11 (33.3%) of 33 thereafter, respectively. The most frequent and serious adverse events were hyperbilirubinemia, hypoalbuminemia, and ascites. The independent adverse prognostic factors for survival were portal vein tumor thrombus (P = 0.0012), tumor response (P = 0.011), acute adverse hepatic event (P = 0.012), and late adverse hepatic event (P = 0.015)., Conclusions: Hypoalbuminemia, hyperbilirubinemia, and ascites were important hepatic adverse events that developed after applying radiotherapy to treat advanced HCC. These adverse events seriously affected survival.

Background: There is no standard treatment for patients with extrahepatic spread from hepatocellular carcinoma. The aim of this retrospective study was to identify candidates for aggressive intervention or new drug trial among such patients., Methods: Retrospective exploration was performed to extract a patient cohort step by step using univariate and multivariate analyses., Results: There were 201 subjects from the past 12 years. The 1-, 3- and 5-year survival rates were 31.0, 9.2 and 4.5%, respectively. The upper limit of the 95% confidence interval of median survival time did not reach 3 months in patients with a performance status of 2 or worse, or with Child-Pugh grades B or C. After excluding those, univariate and multivariate analyses were performed in the remaining 124 subjects. Three independent prognostic factors--brain metastasis, number of metastatic tumors and primary tumor status--were identified. The final cohort was set at 121 after excluding three with brain metastasis. Among these 121, there were 11 with a solitary metastatic tumor and T1/2 primary tumor. In this subgroup, 10 underwent complete removal of the metastatic tumor, and the median survival time exceeded 5 years., Conclusion: Candidates for anti-cancer treatment should meet the following requirements: a performance status of 0 or 1, a Child-Pugh grade A and no brain metastasis. Among these, challenging locoregional intervention was acceptable only for patients with solitary metastatic tumor and T1/2 primary tumor.

Background/aims: To examine the efficacy and safety of a combined modality therapy consisting of hepatic arterial infusion of 5-fluorouracil and external-beam radiotherapy in patients with advanced pancreatic carcinoma., Methodology: Hepatic arterial infusion chemotherapy consisted of 5-FU 1000mg/m2 administrated as a 5-hr continuous infusion once weekly. External-beam radiotherapy (total dose, 50Gy; 2Gy/day) was delivered to the pancreas tumor concurrently for 5-6 weeks. Seventeen patients with no distant metastases except to the liver were enrolled in this study., Results: Patients received a median of 13 cycles of chemotherapy. Sixteen of 17 patients received a total radiotherapy dose of 50Gy. In one patient, treatment was discontinued after 24Gy of radiotherapy and 2 cycles of chemotherapy because of progressive disease. Nausea and vomiting were the most common types of toxicity. Grade 3 or worse toxicity was observed in 2 patients. Four patients developed gastroduodenal ulcers. Of the 16 patients, 7 (41%) showed a partial response, and 9 (53%) showed no change. The median overall survival was 4.5 months and 1-yr overall survival of 11.8% was observed., Conclusions: The combined therapy is active and well tolerated, but results in a poorer prognosis, in spite of its high initial response rate.

Tumor implantation along the needle tract following percutaneous procedures under ultrasonographic guidance for hepatocellular carcinoma (HCC) and pancreatic carcinoma (PC) has been well documented. The purpose of the present study was to investigate the correlation between the procedure, the pathologic differentiation of the primary tumor, and the treatment after implantation. Between July 1992 and March 2000, HCC patients (n=372) who underwent biopsy, percutaneous ethanol injection (PEI) therapy and percutaneous microwave coagulation therapy (PMCT) and PC (n=73) patients who underwent biopsy were retrospectively studied. Needle tract implantation was found in six of the HCC patients (1.6%) and one of the PC patients (1.4%). The interval to diagnosis ranged from 5 to 25 months (mean +/- SD 11.2 +/- 7.6 months) in the HCC patients. The needle tract implantation was evident for all procedure types in these patients (two after PEI alone, two after both biopsy and PEI, and one after PMCT) and for each degree of pathologic differentiation of the primary tumors (well differentiated in one, moderately differentiated in two, and poorly differentiated in one). Each implanted tumor was surgically resected, with no recurrence at the focal lesion. These results suggest that needle tract implantation develops regardless of the procedure or the pathologic differentiation of the primary tumor, and that surgical resection might be effective for controlling these implanted lesions.

Background: Hepatic arterial infusion of zinostatin stimalamer and lipiodol emulsion shows a moderate activity against hepatocellular carcinoma. However, the anti-tumor activity of zinostatin stimalamer alone is uncertain., Methods: The primary endpoint was to evaluate the frequency of dose-limiting toxicity and determine the maximum-tolerated dose of zinostatin stimalamer when used by intra-arterial infusion. The candidates for this study were patients with hepatocellular carcinoma no longer amenable to established forms of treatment. Hepatic arterial infusion chemotherapy was performed by selectively introducing a catheter into the hepatic artery with zinostatin stimalamer alone. Treatment was repeated at 4-8-week intervals until disease progression or the appearance of unacceptable toxicity. The starting dose of zinostatin stimalamer was 3 mg/m(2), and doses were increased in 1 mg/m(2) increments in successive cohorts. At least three patients were treated at each dose level and three additional patients were treated in the presence of dose-limiting toxicity., Results: Twelve patients were entered into this trial. Dose-limiting toxicity was observed in one of six patients at 3 mg/m(2), and in two of six patients at 4 mg/m(2). The maximum-tolerated dose was judged to be 3 mg/m(2) with liver dysfunction and serum creatinine increase as the dose-limiting toxicity. There was one early death suggested to be related to the protocol treatment. None of the 12 patients achieved an objective tumor response., Conclusion: Hepatic arterial infusion with a zinostatin stimalamer of 3 mg/m(2) may be tolerated, but not active, in patients with far advanced hepatocellular carcinoma.

The purpose of the study was to assess contrast enhancement patterns of hepatic tumours during the vascular phase using contrast-enhanced ultrasound and Levovist to differentiate hepatocellular carcinoma from other hepatic tumours. 89 hepatic tumours in 82 consecutive patients were evaluated using coded harmonic ultrasound imaging. The procedure used a phase inversion harmonic technique and coded technology. We observed images for 2 min from the beginning of the administration as the vascular phase using continuous transmission and intermittent transmissions of 1 s or 2 s. The contrast agent Levovist was administered intravenously as a bolus infusion of 2.5 g. Tumour vessels with flow spreading into the tumour and/or homogeneously stained hyperechoic images were observed in 34 of the 41 hepatocellular carcinomas (sensitivity, 82.9%; specificity, 93.8%). Peripheral enhancements were characteristic of intrahepatic cholangiocarcinoma and metastatic hepatic tumours (sensitivity, 60.0% and 83.3%; specificity, 65.5% and 76.4%, respectively). Pooling at the periphery or throughout the tumour was apparent only in haemangioma (sensitivity, 76.5%; specificity, 100%). A tortuous feeding artery and spoke-like vascularization were evident only in the two focal nodular hyperplasias. Contrast-enhanced ultrasound using coded harmonic ultrasound imaging and Levovist provided detailed information about tumour vascularity and contrast enhancement patterns in hepatic tumours.

We prospectively evaluated the efficacy and safety of transcatheter arterial chemoembolization (TACE) with microembolization material, degradable starch microspheres (DSMs), and epirubicin, for treatment of multifocal hepatocellular carcinoma (HCC). Seventeen patients with multifocal HCC were treated. At the first treatment, DSMs were injected alone to determine the dose for embolization of the hepatic artery in each patient. After 4 weeks, TACE was performed every 4 to 6 weeks with a mixture of DSMs and epirubicin at a dose of 40 mg/m2. A necrotic area of more than 50% was produced in 6 patients by DSMs alone, and in 11 patients by TACE. The overall response rate was 52.9% (2 complete and 7 partial responses). The duration of the responses ranged from 4 to 21 months (median: 9 months). Common toxicities were transient abdominal pain, nausea/vomiting, fever, and leukopenia. In four patients, grade III or IV toxicity was observed as gamma-glutamyl transpeptidase elevation. TACE with DSMs had tumor necrosis efficacy with acceptable toxicity. The median survival time was 21.7 months, and the 2-year survival rate was 45.3%. Further investigation of the effects of DSM treatment on survival should be carried out.

Background: This article describes the operative procedures, varying difficulties, and required instrumentation for performing laparoscopic hepatectomy (LH) on the basis of a lesion's extrahepatic growing (EG) index, as calculated by computed tomography (CT)., Methods: Laparoscopic partial hepatectomy cases were divided into the following two groups: an EG tumor group (n = 11) and an intrahepatic tumor group (n = 8). The surgical procedures, operative results, and laparoscopic instrumentation for these two groups were compared based on the EG index (/cm2; maximum diameter of tumor pedicle/maximum vertical diameter of tumor/area of the tumor)., Results: The mean operative time was significantly shorter and the mean blood loss was significantly less in the EG tumor group than in the intrahepatic tumor group. In addition, in the EG tumor group, there were significant differences in mean operative time and mean blood loss related to the values associated with the EG index (p <0.05, P <0.01). The selection of laparoscopic instruments was based on the EG index, as follows: (a) lesions with an EG index >5/cm2 underwent resection in combination with a microwave tissue coagulator and an ultrasonic surgical aspirator (13 cases, including intrahepatic tumor cases); (b) lesions with an EG index of 5-15/cm2 underwent resection in combination with a microwave tissue coagulator and laparosonic coagulating shears (four cases); (c) lesions with an EG index <%15/cm2 underwent resection with a laparoscopic linear stapler (two cases)., Conclusion: Our preliminary experience leads us to believe that it is useful to calculate the EG index by CT scan before formulating the technical strategy for a subsequent LH procedure.

A 70-year-old male was admitted to hospital for a liver tumor. We diagnosed that hepatocellular carcinoma (HCC), 1.5 cm in diameter, by several examinations. We performed ultrasonically guided radio-frequency ablation (RFA). Ultrasonography (US) 9 days post-therapy revealed a cystic lesion 1.5 cm in diameter. Color Doppler US revealed color imaging throughout the lesion and dynamic computed tomography (CT) demonstrated an area of hyperattenuation. Intrahepatic pseudoaneurysm was diagnosed. On 14 days post-therapy, no color signals on color Doppler US or enhanced areas on CT were seen. We submit that pseudoaneurysm should be recognized as a complication of radio-frequency ablation.

Lymph node metastasis is reported to occur only rarely in patients with hepatocellular carcinoma (HCC). However, we have encountered patients with HCC with extensive lymph node metastases. Here we report the clinical characteristics of HCC associated with extensive lymph node metastasis at diagnosis. Ten patients with HCC in whom extensive lymph node metastases were observed at the initial medical examination were studied. The degree of disease progression was documented with ultrasonography and dynamic computed tomography. Primary liver lesions were classified in the following three types according to imaging characteristics: type A, massive type with portal vein tumor thrombus; type B, multinodular, nonencapsulated type; and type C: multinodular, encapsulated type. In patients with types A and B HCC, a large number of lymph node metastases was observed, whereas a small number of isolated metastases was observed in patients with type C. All patients with types A and B HCC died within 7 months (median survival, 4 months), whereas those with type C survived for 4 years or more after treatment with transcatheter arterial chemoembolization and surgery. A relationship exists between the type of primary HCC lesions and the pattern of lymph node metastasis. Long-term survival may be expected for patients with isolated lymph node metastases.

We report the successful treatment of hepatocellular carcinoma (HCC) associated with an intraductal tumor thrombus in a 67-year-old male. Abdominal ultrasonography (US) and computed tomography (CT) revealed intrahepatic biliary dilatation in the left hepatic lobe and an intraductal tumor thrombus. The main tumor lesion was not clearly visualized on abdominal US, dynamic CT, and hepatic angiography. We biopsied the intraductal tumor thrombus under US guidance. Histologically the biopsy specimen was a poorly differentiated HCC We thus diagnosed HCC with intraductal tumor thrombus. The total serum bilirubin level gradually rose to 3.1 mg/dl. This tumor was inoperable because of severe hepatic dysfunction. We chose to treat the patient with radiotherapy aimed only at the intraductal tumor thrombus because the main tumor was unclear. A percutaneous transhepatic biliary drainage (PTBD) tube was inserted into the common bile duct beyond the tumor thrombus and the tube was dilated. Once total serum bilirubin had reached the normal range, a combination of external beam radiation therapy (EBRT) plus an intraluminal brachytherapy, 192Ir boost was administered. The intraductal tumor thrombus was found to have vanished and the PTBD tube was removed. After this treatment, transcatheter hepatic arterial embolization was performed at the point of tumor appearance. This patient had a relatively long survival, approximately 30 months, with no clinical evidence of recurrent disease and biliary drainage was not necessary.

Background/aims: Liver metastasis is a common progression of pancreatic carcinoma, but an effective chemotherapy has not been established. The purpose of this study was to examine the efficacy and safety of a hepatic arterial infusion of 5-FU in patients with liver metastasis from pancreatic carcinoma., Methodology: Thirteen patients were enrolled in a pilot study of a hepatic arterial infusion of 5-FU therapy. They received 5-FU for 5 days at a dose of 500 mg/m2/day by continuous hepatic arterial infusion every 4 weeks., Results: One patient showed a partial response, while 6 showed no change. Of these 6 patients, 2 showed a minor response. The overall response rate was 8% (95% confidence interval: 0-22%). Nausea and vomiting were the most common types of toxicity. Three patients (23%) had hepatic arterial occlusion. There were no life-threatening toxicities or complications. The overall median survival time was 15.9 weeks., Conclusions: Hepatic arterial infusion of 5-FU in patients with liver metastasis from pancreatic carcinoma is tolerable but is minimally effective at this dose and schedule. The schedule of administration should be modified.

Purpose: A B-flow sonographic technique was recently developed to provide direct visualization of blood flow with gray-scale sonography. Compared with color Doppler sonography, B-flow imaging has wideband resolution and a high frame rate. The purpose of this study was to evaluate the usefulness of B-flow sonography for visualizing blood flow in hepatic vessels and tumor vascularity in patients with liver cirrhosis or hepatocellular carcinoma (HCC)., Methods: Twenty-five patients with liver cirrhosis, including 15 with HCC, were studied by B-flow and color Doppler sonography. Blood-flow detection rates in portal veins and hepatic arteries and tumor vascularity in HCC were analyzed, and the 2 methods were compared., Results: Using B-flow, blood flow was visualized in the portal vein in 23 (92%) of 25 patients and was visualized in the hepatic artery separately from the portal vein in 9 (36%) of 25 patients. The blood-flow signals were visualized only within vessels, never "bleeding" outside the vessel's lumen. Blood flow in the portal vein was observed with color Doppler sonography in all 25 patients, but the hepatic artery was never clearly separated from the portal vein. Vascularity within the HCC tumor was detected in 9 (60%) of 15 nodules with B-flow imaging, and fine arteries flowing into the tumor were observed in 6 nodules. Color Doppler sonography detected blood flow in 13 (87%) of the 15 HCC nodules., Conclusions: Blood flow in hepatic vessels and tumor vessels of HCC were visualized with B-flow sonography. B-flow sonography is a potentially useful technique for the evaluation of liver vascularity and intratumoral vessels., (Copyright 2000 John Wiley & Sons, Inc.)

Hepatocellular carcinoma (HCC) is characterized by tumor vascularization from the hepatic artery. The objective of our work was to compare color Doppler ultrasonography (CDU), including power Doppler ultrasonography (PDU) with radiographic imagings with contrast medium in regard to the detection of the arterial tumor vascularity of small hepatocellular carcinomas (HCC). We examined 42 small HCC lesions 2 cm or less in diameter in 37 patients for arterial tumor vascularity by conventional CDU, PDU, dynamic computed tomography (dCT), digital subtraction angiography (DSA), and CT hepatic arteriography (CTA). Color images were detected in 25 (59.5%) and 28 (66.7%) of the 42 lesions with conventional CDU and PDU, respectively, and tumor vascularity was detected in 26 (61.9%) by dCT, 23 (54.8%) by DSA, and 29 (69.0%) by CTA. Tumor vascularity could be detected in 51.9% by PDU and CTA, more than by conventional CDU, dCT, and DSA (44.4, 44.4, and 33.3%, respectively) in well-differentiated HCC, whereas the detection rates by these techniques were almost equal (86.7% by CDU, 93.3% by PDU, dCT, and DSA, 100% by CTA, respectively) in moderately and poorly differentiated HCC. PDU is superior to CDU, dCT and DSA and equal to CTA for the detection of tumor vascularity in small HCC, particularly in well-differentiated HCC.

Purpose: We report the sonographic appearances and pathologic findings for hepatic nodules 2 cm or smaller that were detected by sonography but that did not produce a tumor stain on constrast-enhanced helical CT or digital subtraction angiography., Methods: Sixty-six nodules 2 cm or less were found by sonography in 39 patients. Sonographically guided needle biopsies were done on all lesions., Results: Twenty-seven nodules were benign nodules, 9 were borderline lesions, and 30 were hepatocellular carcinomas. Benign nodules were significantly smaller than hepatocellular carcinomas (mean size, 1.0 cm versus 1.4 cm, respectively; p <0.00001). Nodules 1 cm or smaller were more likely to be diagnosed as benign nodules (68%) than as either of the other types (32%; p = 0.01). A significantly greater percentage of hepatocellular carcinomas showed a heterogeneous pattern (64%) on sonography compared with benign nodules (25%) and borderline lesions (11%; p = 0.04). In nodules 1 cm or smaller, a homogeneous pattern (68%) was more common than a heterogeneous pattern (32%; p = 0.01)., Conclusions: The nature of small hepatic nodules cannot be determined with sonography. Thus, biopsy remains the only technique for obtaining a definitive diagnosis.

Purpose: To evaluate transcatheter arterial embolization in patients with hepatocellular carcinoma, portal vein tumor thrombus, and arterioportal shunts., Materials and Methods: Ten patients with hepatocellular carcinoma, portal vein tumor thrombus, and severe arterioportal shunting were identified; in these patients, portal blood flow before embolization was hepatofugal. Embolization of arterioportal shunts was performed with steel coils that were introduced through a catheter during arteriography. After embolization, changes in portal hemodynamics and clinical signs and performance status of patients were evaluated; survival rates of patients with and patients without severe arterioportal shunting were compared., Results: In all patients after embolization, angiography showed resolution of arterioportal shunting, and portography showed hepatopetal blood flow in the portal vein trunk. After embolization, performance status of five patients with initial scores of 2 or 3 improved. Ascites resolved in four patients and improved in four patients. One patient died of hepatic failure caused by rupture of esophageal varices 7 days after embolization. Median survival was 4.3 months, and the 6-month and 1-year survival rates were 45% and 12%, respectively. There were no significant differences between survival rates in patients with and patients without severe arterioportal shunting., Conclusion: Transcatheter arterial embolization of arterioportal shunts is a useful treatment for improving quality of life in patients with hepatocellular carcinoma.

We report a hepatocellular carcinoma that appeared to undergo spontaneous regression twice. On June 5, 1995, two hepatocellular carcinomas were detected by ultrasonography in a 72-year-old woman. On July 1, 1995, the tumor in the lateral segment had increased in size from 2.0 cm to 3.2 cm in 27 days, and the level of alpha-fetoprotein (AFP) had increased from 743 to 1300 ng/ml. On September 22, 1995, one tumor appeared as a typical hypervascular lesion on contrast-enhanced computed tomography. It was 3.9 cm in size and located in the posterior segment. The other was a 3.9-cm cystic mass located in the lateral segment. The tumor in the posterior segment was treated with an emulsion of epirubicin and lipiodol administered via the posterior branch of the right hepatic artery and percutaneous ethanol injection. The tumor in the lateral segment was kept under observation because it was avascular, had regressed to 2.0 cm in size, and the patient's AFP level had decreased to 237 ng/ml by December 5, 1995. By February 21, 1996, the AFP level had increased again to 105,340 ng/ml, and by March 22, 1996, the latter tumor had increased in size to 3.8 cm. However, it later regressed again to 1.6 cm, and the AFP level decreased spontaneously to 7 ng/ml. From the changes in the contrast-enhanced computed tomographic images, and the level of AFP, we cannot rule out the possibility that the spontaneous regression might have been caused by the tumor's rapid growth.

The outcome in 3225 patients with hepatocellular carcinoma (HCC) was studied in groups with equivalent prognosis treated with resection, transcatheter arterial embolization (TAE), and percutaneous transhepatic ethanol injection (PEI). Significant factors for better clinical background included a tumor diameter of < or = 30 mm, tumor number < or = 3 and (clinical) Stage I. In patients with Stage I disease having tumors of < or = 30 mm and < or = 3 in number, survival afer resection and PEI did not differ, while survival after TAE was significantly worse. In those patients with Stage II disease, survival after PEI was significantly better than after resection or TAE. In patients with Stage I or II disease having tumors > or = 31 in size and < or = 3 number, survival after resection was significantly better than after TAE. In patients with Stage I disease having tumors of > or = 31 mm and > or = 4 in number, survival after resection was significantly better than after TAE. Our conclusions are as follows. Firstly, resection or PEI is recommended for patients with Stage I disease having < or = 3 tumors all < or = 30 mm in size. Secondly, PEI is recommended for patients with Stage II disease having < or = 3 tumors all < or = 30 mm in size. Thirdly, for patients with Stage I disease having tumors 31 mm or larger in size, whatever the number of tumors, resection should be selected rather than TAE.

We analyzed the clinical characteristics and survival of 185 patients with hepatitis B virus-related hepatocellular carcinoma (HBV group) and 1033 with hepatitis C virus-related hepatocellular carcinoma (HCV group) by multi center study. The patients in the HBV group (mean age 52.1 yr) were about 10 years younger than those in the HCV group (mean age 62.9 yr). Liver function, as measured by indocyanine green retention at 15 min, was better in the HBV group (17.5%) than in the HCV group (25.4%). A higher proportion of the HBV group (55%) than the HCV group (44%) had clinical stage I, T-factor differed significantly between the groups: 53% of the HBV group were T3-4 compared with 41% of the HCV group. Furthermore, a higher proportion of the HBV group were graded 2-3 for tumor thrombus in the portal vein (20.3%) and had poorly differentiated hepatocellular carcinoma (7%) compared with the HCV group (7.1% and 5% respectively). Univariate analysis identified poor prognostic factors for hepatocellular carcinoma as HBV, age < or = 50 yr, clinical stage II-III, a high AFP level, higher number of tumors, larger tumor size, tumor thrombus in the portal vein 2-3 and in the hepatic vein 2-3. On multivariate analysis, poor prognostic factors were a high AFP level, higher number of tumors, tumor thrombus in the portal vein 2-3 and in the hepatic vein 2-3, but not HBV, age, clinical stage or tumor size. These results suggest that HBV itself is not a stronger prognostic factor than HCV.

Small hepatic nodular lesions are frequently detected by ultrasonography in patients with liver cirrhosis during follow-up. However, hepatocellular carcinomas (HCCs) and non-HCC nodules are difficult to differentiate by ultrasonography because they have a similar appearance. We used color Doppler ultrasonography (CDU) to examine 29 HCCs and 26 non-HCC nodules less than 2 cm in diameter to determine whether CDU can be used to differentiate small hepatic nodules. There were no significant differences between the examined HCCs and non-HCC nodules with regard to ultrasonographic appearance, i.e., hypoechoicity or hyperechoicity. Blood flow signals were detected in a significantly higher percentage of HCCs (13 of 29, 44.8%) than in non-HCC nodules (2 of 26, 7.7%; P<0.005), although the sensitivity was low. Therefore, we classified nodules as hypoechoic or hyperechoic by B-mode ultrasonography and examined their blood flow signals. Among nodules that were hypoechoic, signals were detected in a significantly higher percentage of HCCs (12 of 18, 66.7%) than in non-HCC nodules (2 of 13, 15.4%; P<0.005). In contrast, among hyperechoic nodules, signals were detected in only 1 of 11 (9.1%) HCCs and in 0 of 13 non-HCC nodules. Our results suggest that examination of blood flow by CDU is useful for differentiating small hepatic nodules, especially hypoechoic nodules.

We examined serial changes of tumor images in a patient with hepatocellular carcinoma. The hepatocellular carcinoma was initially detected as a homogeneous low-echo area with unclear margins, which was not enhanced by contrast media on computed tomograms, and did not reveal any vascular abnormalities on hepatic angiography. About 11 months later, the tumor growth accelerated, with a parallel increase in serum alpha-fetoprotein levels, and the ultrasonographic features of the tumor changed from a homogeneous low-echo area to a mixed low- and high-echo area with a peripheral low-echo zone. Hepatic angiography revealed a hypervascular tumor at this time. The present case indicates that tumor growth and imaging patterns of hepatocellular carcinoma are closely related to vascularization of the tumor.

A 61-year-old man received systemic mitoxantrone chemotherapy following transarterial embolization of a large hepatocellular carcinoma with extensive intrahepatic, lung and bone metastases. His serum alpha-fetroprotein levels were 199,000 ng/ml prior to chemotherapy. He was given 10 mg mitoxantrone/m2 (14 mg/dose) intravenously every three weeks, and showed a rapid decrease in his serum levels of alpha-fetoprotein. There was almost complete resolution of the multiple bilateral metastatic lung nodules at five weeks and a marked decrease in the size of the intrahepatic metastatic nodules by seven weeks. The primary tumor, however, which had been previously treated by transarterial embolization, only underwent a 33% reduction in size according to WHO criteria. This occurred despite the disappearance, demonstrated by contrast enhanced computed tomography, of all viable tumor tissue. Ultrasonography also revealed only a minor regression, and could demonstrate no changes in the tumor echo pattern. The response pattern observed in this patient indicates the response to chemotherapy for hepatocellular carcinoma to be modified by prior transarterial embolization.

We describe the results of our study on the early detection of the development of hepatocellular carcinoma among patients with chronic liver disease. Over a period of 18 years, 33 patients were diagnosed as having hepatocellular carcinoma. From 1970 to 1978, we used serum alpha-fetoprotein determination, liver palpation, and radionuclide liver scans. In addition to alpha-fetoprotein determination, computed tomography and ultrasonography were introduced in 1979. However, we did not have any general guidelines for the use of these imaging modalities. From 1984 onwards, ultrasonography and serum alpha-fetoprotein determination have been performed every three months and computed tomography every year in patients whose right hepatic lobe had atrophied due to liver cirrhosis. On the basis of this screening program, 50% of the detected hepatocellular carcinoma (9/18) were found to be smaller than 1.9 cm in diameter, and tumor resection was performed in 11 out of 18 patients (61%).

Ultrasonographic features of tumor nodules in 64 patients with hepatocellular carcinoma and 87 with metastatic liver cancer were analyzed for differential diagnosis. Sonographic characteristics of hepatocellular carcinoma were the mosaic, posterior echo enhancement, and lateral shadow patterns, which were rarely observed in metastatic liver cancers. Small hepatocellular carcinomas, less than 3 cm in diameter, showed a hypoechoic pattern. Metastatic liver cancers originating from the gastrointestinal tract were frequently hyperechoic irrespective of their size and the bull's-eye-like pattern was specific for metastases from the lung. These results indicate that ultrasonography is useful for the differential diagnosis of hepatocellular carcinoma and metastatic liver cancer.

Four patients with massive cavernous hemangioma of the liver were treated with 6 Mev X-ray, and favorable results were obtained in 3. Patient 1 was treated with a total tumor dose of 940 rads without response. The 2nd patient was given a total tumor dose of 1900 rads over 54 days, with a daily dose of 50 to 100 rads. Although the hemangioma regressed after treatment, the response was transient. A total tumor dose of 2940 rads was given over 41 days with a daily dose of 100 to 200 rads as a second course, and the hemangioma regressed with no sign of reenlargement. Patients 3 and 4 were given 150 to 200 rads every other day to a total of about 3000 rads, with a satisfactory response. Follow-up studies showed the latter 3 patients to be well 8, 5, and 4 years after radiotherapy, respectively.

The accuracy of real-time ultrasonography in screening for hepatocellular carcinoma at the out-patient Liver Clinic was prospectively evaluated in 245 patients with chronic liver disease. There were 14 patients with hepatocellular carcinoma who had no specific symptoms related to it. Tumour was detected by ultrasonography in 12 patients, but it missed the tumour in two patients. There were also three false positive results. The sensitivity and specificity of ultrasonography in the diagnosis of hepatocellular carcinoma were 85.7% and 98.8%, respectively. These are superior to alpha-fetoprotein (AFP) measurement, especially in specificity (p less than 0.01) at 19 ng/ml cut-off. False negative results were compensated for by AFP and computed tomography. A decision on treatment schedule was easily made on an anatomical basis by ultrasonographic examination.

The clinical value of preoperative ultrasonography in screening for synchronous liver metastases was prospectively evaluated in 338 patients with colorectal cancer. Synchronous liver metastases were observed at laparotomy in 11.5% (39/338) of the patients. The liver metastases had been found by preoperative ultrasonography in 30 patients and missed in nine. The overall accuracy rate, sensitivity and specificity, and the positive and negative predictive values of this modality were 0.970, 0.769, 0.997, 0.968 and 0.971, respectively. In detecting liver metastases, the results were superior to those of biochemical blood tests and measurements of carcinoembryonic antigen serum levels. The accuracy of the ultrasonography was also superior to that of these other tests combined. The results indicate ultrasonography to be an indispensable preoperative examination for patients with colorectal cancer.

Twenty-four patients with hepatocellular carcinoma were treated with 80 mg etoposide/m2 for five consecutive days at 4-week intervals. Of the 21 patients evaluated, only one showed a partial response (of 8 weeks, duration). The dose-limiting factor was bone marrow suppression. Grades 3 and 4 white blood cell count toxicity occurred in five (24%) and two (10%) patients, respectively. Alopecia (grades 1 and 2) was observed in 18 patients (57%). The The present study indicates that etoposide has very limited activity against hepatocellular carcinoma.

Spontaneous regression of hepatocellular carcinoma was observed in a 67-year-old Japanese male with liver cirrhosis. Although the patient showed no histological evidence of hepatocellular carcinoma, findings obtained by hepatic angiography, computed tomography and echography strongly indicated that the tumor was a hepatocellular carcinoma. Moreover, the rate of growth of the tumor nodule paralleled an increase in serum alpha-fetoprotein. Following this accelerated increase in serum alpha-fetoprotein a sudden drop was observed with no specific treatment for cancer, and regression of the tumor was also observed. Six months later the serum alpha-fetoprotein level had normalized, and the tumor had disappeared completely. The patients is now alive and doing well more than 70 months after the initial diagnosis of hepatocellular carcinoma.

During a 7-year period in our hospital, 69 patients with inoperable hepatocellular carcinoma (HCC) underwent 111 courses of transcatheter hepatic artery embolization (TAE) and/or chemoinfusion with lipiodol. Patient survival was 0.5-37 months following therapy and the factors affecting prognosis were evaluated. Survival rates at 1, 2 and 3 years after TAE were 53%, 24% and 15%, respectively. Survival rates at 1, 2 and 3 years in relation to tumor size were 100%, 100% and 100% in 5 patients (tumor size less than 2 cm in diameter), 81%, 33% and 16% in 23 patients (2.1-5.0 cm), and 35%, 9% and 0% in 41 patients (greater than 5.1 cm). An analysis of prognostic factors showed that the size of the main tumor significantly influenced the prognosis following TAE (P less than 0.01), whereas the frequency of TAE, intrahepatic metastasis and the degree of liver dysfunction showed a slight correlation (P less than 0.1). These results suggest that TAE has a significant potential for becoming the first choice of treatment for patients with small multiple HCCs (less than 2 cm), provided that neither severe hepatic dysfunction nor a tumor thrombus in the main portal vein is present.

A combined hepatocellular and cholangiocellular carcinoma of diffuse type in a Japanese man is described. A small localized solitary tumor apparently grew rapidly into a diffuse-type carcinoma, and the liver weight increased about 4-fold during the last two months. The clinical course of this case was as expected for a diffuse type of hepatocellular carcinoma except that unusually high levels of serum carcinoembryonic antigen were found. The patient died of hepatic failure with systemic bleeding five months later. At autopsy, multiple small nodules were suspected to be intrahepatic metastatic foci because portal tumor thrombus was observed in the right antero-superior segment where the initial tumor was localized. Histologically, the tumor had components of both hepatocellular and mucin-producing cholangiocellular carcinoma. This is believed to be the first report on a diffuse type of combined hepatocellular and cholangiocellular carcinoma initially presented as a localized small nodule.

One hundred and twenty-three patients with hepatocellular carcinoma and 4 patients with hepatoblastoma were treated with systemic chemotherapy, and then one patient with hepatocellular carcinoma and 4 patients with hepatoblastoma underwent hepatectomy. Preoperative chemotherapy was evaluated to have been successful in the patient with hepatocellular carcinoma and 3 of the 4 patients with hepatoblastoma. In the former, marked tumor necrosis was observed in the resected specimen. In the latter, a rapid decrease of serum AFP level (T1/2 was less than 13 days) was observed in the effective cases. Preoperative chemotherapy is very useful if effective drugs are available. However, the most serious problem in this field is the lack of drugs active against hepatocellular carcinoma. Our recent results of phase II studies on hepatocellular carcinoma using various drugs were as follows adriamycin 6% (1/17), tegafur 7% (1/15), UFT 4% (1/26) and VP-16 5% (1/21). Development of a new active anticancer agent is thought to be essential if treatment of hepatocellular carcinoma is to advance.

Bone metastasis was observed in 16.1% or in 14 of 87 male autopsy cases of hepatocellular carcinoma. The primary tumor within the liver was located in the right lobe in all but one case. There were six patients who first presented with signs attributable to bone metastasis, and lung metastasis subsequently became evident in five of them. These 6 patients lived significantly longer as compared with 8 other patients with bone metastases and 73 patients without. The possible route by which hepatocellular carcinoma cells were carried to the bone is discussed.

Double primary liver cancer, i.e., hepatocellular carcinoma (HCC) and cholangiocellular carcinoma (CCC), was detected by ultrasound (US) during the follow-up of chronic liver disease, and was safely resected in two patients, one synchronously and the other metachronously. On admission, laboratory data were normal except for slightly abnormal liver function tests. One patient (case 1) had a 5.1-cm white firm CCC with several small daughter lesions around it which was enhanced late in computed tomography, and a 2.1-cm well-circumscribed HCC with a pseudocapsule; the two were situated very close to one another in the right posterior segment of the liver. Another patient (case 2) had a small 1.5-cm CCC in the anterior superior area, and 6 yr after the first resection, alpha-fetoprotein gradually increased and a 4-cm HCC was detected anew and resected; the preoperative diagnosis made by US and ethiadol-computed tomography was correct. The histopathological diagnosis of the noncancerous portion was chronic active hepatitis in both cases. However, it changed to precirrhosis in case 2 during the follow-up.

Twenty-four eligible and evaluable patients with measurable liver metastases from gastric cancer were treated with oral or rectal Tegafur. Objective responses were seen in 8 of the patients (33.3%), lasting between 1.5 and 15 months. The median survival period was 10 months for the responders and 2.6 months for non-responders. No complete response was observed. The survival of responders was significantly longer than that of non-responders (p less than 0.01). Toxicity was mild and consisted principally of gastrointestinal and hematological side effects. No central nervous system toxicity was seen.

We measured the size of 17 semispherical hepatocellular carcinoma tumor nodules in 13 patients repeatedly by either computed tomography or ultrasonography for 10 to 291 days. At the initial examination the diameter of these tumors was 30 +/- 15 mm (mean +/- SD). When the doubling time of these tumor volumes was calculated by the formula proposed by Schwartz, it was 119 +/- 96 days (mean +/- SD). There was no correlation between the doubling time of the tumors and the degree of impairment of liver function in any case. There was also no correlation between the doubling time of the tumors and the alpha-fetoprotein level. However, in patients with hepatitis B surface antigen, the doubling time was short (48 +/- 8 days), when compared with the antigen-negative cases (140 +/- 98 days). Based on the shortest doubling time of the smaller tumors, it was revealed that the diameter was duplicated in four months. Since the minimum tumor diameter that was detected by ultrasonography and/or computed tomography was considered to be 1 cm, we recommend that ultrasonography and/or computed tomography should be performed at least every four months to detect so-called "small hepatocellular carcinoma," which has a diameter of less than 2 cm, in patients at high risk for hepatocellular carcinoma.

A small hepatocellular carcinoma found and followed in a Japanese man is described. He died of diabetic nephropathy and bleeding esophageal varices seven years after surgical biopsy of the liver that revealed hepatocellular carcinoma. Clinical and autopsy findings demonstrated no appreciable change in tumor size during the seven-year period during which no specific treatment was given for cancer.

The relationship of tumor volume doubling time to length of patient survival was investigated for 15 patients with small hepatocellular carcinoma smaller than 4.5 cm in diameter. The mean tumor volume doubling time of these 15 nodules was 102 +/- 77 days (mean +/- SD; range 41 to 305 days) before the initiation of a specific treatment for cancer. These doubling times tended to correlate with mitotic indexes of the tumors and the patients could be divided into two groups according to the therapeutic modalities used. Patients in Group A received systemic chemotherapy without response or nonspecific treatments for cancer. In this group, there was a positive correlation between tumor volume doubling time and survival length (r = 0.8812; P less than 0.025). Patients in Group B either received hepatectomy after transarterial embolization or systemic chemotherapy or received hepatectomy alone. In this group, early death occurred in patients who had shorter tumor volume doubling times. Three surgically treated patients in Group B were evaluated as having survived for a significantly long period as assessed from their tumor volume doubling times. These results indicate that tumor volume doubling time is one of the determining factors of survival length in patients with hepatocellular carcinoma, and, therefore, can be used in the evaluation of therapeutic efficacy.