This study demonstrated that tumor size influences the local efficacy of PEI for small HCC. Therefore, we recommend that a reasonable indication for PEI therapy is HCC lesions measuring less than 30 mm in greatest dimension.

A total of 71 consecutive patients with unresectable hepatocellular carcinoma were analyzed retrospectively to determine the significant prognostic factors. All the patients received systemic chemotherapy in a phase 2 study from 1980 to 1990, with no other anticancer treatment. Median survival time and 1-yr and 2-yr survival rates were 5.6 mo, 23% and 5%, respectively. By the univariate analysis, a performance status of 0-1 and tumor size less than 50% of the liver cross-sectional area were shown to be the factors most significantly favoring a better prognosis. By the multivariate analysis using the Cox proportional hazards model, a performance status of 0-1 (p less than 0.001), absence of tumor thrombus in the main portal trunk (p = 0.003) and age less than 60 yr (p = 0.036) were independent favorable prognostic factors. A prognostic index was calculated from these three factors according to the following equation: 1.8109 x (0 = performance status of 0-1 and 1 = performance status of 2-3) + 0.9322 x (0 = tumor thrombus absent in the main portal trunk and 1 = present) + 0.6996 x (0 = age less than 60 yr and 1 = age greater than or equal to 60 yr). This index was used to classify the patients into three groups with a good, intermediate and poor prognosis. The median survival times for these three groups were 9.8, 3.8 and 1.9 mo, respectively (p less than 0.01). The results of this study may be useful in the design and analysis of future clinical trials of systemic therapy for hepatocellular carcinoma.

Background. Mutations in the K‐ras oncogene at codon 12 are detected at a remarkably high frequency in pancreatic carcinomas and are believed to be a critical event in oncogenesis. The authors attempted to detect K‐ras mutations in DNA obtained from pure pancreatic juice collected endoscopically, as a novel diagnostic approach to pancreatic carcinoma. Methods. K‐ras mutations were examined using the two‐step polymerase chain reaction (PCR) combined with restriction enzyme digestion, followed by nonra‐dioisotopic single‐strand conformation polymorphism (SSCP) analysis. Results. Specific mutations of the K‐ras gene at codon 12 were found in six of nine (67%) duct cell carcinomas, all of which were negative by cytodiagnosis of the same pure pancreatic juice. K‐ras mutations were not detected in the pancreatic juice from 14 healthy control subjects, 10 patients with chronic pancreatitis, or 3 patients with islet cell tumors. Conclusions. Detection of K‐ras mutation at codon 12 in pancreatic juice is highly specific for diagnosing pancreatic duct cell carcinoma and may be a valuable diagnostic modality for pancreatic carcinoma and for differentiating chronic pancreatitis from carcinoma. Cancer 1994; 73:1589–94.

radiation course. Thereafter, weekly infusions (500 mg/m 2 ) were administered until disease progression. 1 Department of Internal Medicine, National RESULTS. Of the 20 patients, 17 (85%) completed the scheduled course of chemoraCancer Center Hospital, Tokyo, Japan. diotherapy. Grade 3 or worse toxicity, graded according to World Health Organiza-2 Department of Therapeutic Radiology, Nation criteria, was observed in 4 patients (20%). Two patients (10%) achieved partial tional Cancer Center Hospital, Tokyo, Japan.response, and disease remained stable in 16 patients (80%). After the completion of combined therapy, serum CA 19-9 levels were reduced by more than 50% in 10 of 12 patients (83%) who had pretreatment CA 19-9 levels of 100 U/mL or greater. The median progression free survival and 1-year progression free survival rate were 4.9 months and 29.5%, respectively. The median overall survival and 1-year overall survival rate were 10.3 months and 41.8%, respectively. T he prognosis of patients with pancreatic carcinoma is extremely poor because of difficulty in the early detection of this disease and the ineffectiveness of nonsurgical treatments. For patients with locally 5-FU is an antimetabolite with a very short plasma half-life, and its major cytotoxicity occurs during the S-phase. 4 Therefore, pro- CONCLUSIONS.

A phase 2 study of cisplatin was performed in 28 previously untreated patients with unresectable hepatocellular carcinoma. The drug was given intravenously at a dose of 80 mg/m²/day every 4 weeks. Of 26 patients evaluated, 4 (15.4%) showed partial responses lasting for > 3 months, while no patient achieved a complete response. Of 22 patients whose serum level of α-fetoprotein (AFP) was high ( > 400 ng/ml) before treatment, 6 (27.3%) showed a > 50% reduction in serum AFP levels after treatment. The current study indicates that cisplatin is an anticancer agent worthy of further testing in patients with this disease.