240 Hepatocellular Carcinomas: Ultrasound Features, Tumor Size, Cytologic and Histologic Patterns, Serum Alpha-Fetoprotein and HBs Ag

Identification of Clinical Phenotypes and Related Survival in Patients with Large HCCs

BACKGROUND

Hepatocellular carcinoma (HCC) factors, especially maximum tumor diameter (MTD), tumor multifocality, portal vein thrombosis (PVT), and serum alpha-fetoprotein (AFP), influence survival.

AIM

To examine patterns of tumor factors in large HCC patients.

METHODS

A database of large HCC patients was examined.

RESULTS

A multiple Cox proportional hazard model on death identified low serum albumin levels and the presence of PVT and multifocality, with each having a hazard ratio ≥2.0. All combinations of these three parameters were examined in relation to survival. Using univariate Cox analysis, the combination of albumin >3.5 g/dL and the absence of both PVT and multifocality had the best survival rate, while all combinations that included the presence of PVT had poor survival and hazard ratios. We identified four clinical phenotypes, each with a distinct median survival: patients with or without PVT or multifocality plus serum albumin ≥3.5 (g/dL), with each subgroup displaying high (≥100 IU/mL) or low (<100 IU/mL) blood AFP levels. Across a range of MTDs, we identified only two significant trends, blood AFP and platelets.

CONCLUSIONS

Patients with large HCCs have distinct phenotypes and survival, as identified by the combination of PVT, multifocality, and blood albumin levels.

Pharmacokinetic imaging of 11C ethanol with PET in eight patients with hepatocellular carcinomas who were scheduled for treatment with percutaneous ethanol injection

PURPOSE

To evaluate the carbon 11 ethanol kinetics with positron emission tomography after intratumoral injection of the tracer and assess its redistribution and dilution in patients who have hepatocellular carcinomas and who were scheduled for treatment with percutaneous ethanol injection.

MATERIALS AND METHODS

The study included eight patients with hepatocellular carcinomas. 11C ethanol was administered via a puncture needle positioned with ultrasonographic guidance. Parametric images based on the Fourier transformation were created for further analysis of the local distribution patterns of the tracer. The ratio of the 45-minute postinjection standardized uptake value to the 5-minute postinjection standardized uptake value was used for the evaluation of ethanol dilution.

RESULTS

Five of eight tumors demonstrated almost constant uptake values after the initial distribution phase. In contrast, a rapid elimination of the 11C ethanol from the tumor was documented in three of eight tumors. The 45 minute-to-5 minute ratio was 0.18-0.67 (median value, 0.56) in the tumors. The time-activity curves of the normal liver parenchyma increased slowly but steadily with time owing to a low ethanol elimination from the tumor. Fourier transformation demonstrated inhomogeneous parts on the amplitude images in seven of eight tumors and random redistribution on the phase images in six of eight tumors.

CONCLUSION

Inhomogeneous drug distribution and drug dilution in the target area are likely to be the major limiting parameters for therapy response.

Etiologic factors and clinical presentation of hepatocellular carcinoma. Differences between cirrhotic and noncirrhotic Italian patients

BACKGROUND

It is not known whether the prevalence of hepatocarcinogenic factors differs between cirrhotic and noncirrhotic patients with hepatocellular carcinoma (HCC) or whether the clinical presentation of HCC in these two groups differs.

METHODS

The prevalence of the putative etiologic factors of HCC and its clinical presentation in 373 patients with cirrhosis and 102 without cirrhosis seen from 1981 to 1992 were evaluated.

RESULTS

Hepatitis C virus infection (76 vs. 48%, P = 0.003) and both current (22.5 vs. 10%, P = 0.007) and past (50.5 vs. 34.5%, P = 0.045) hepatitis B virus infections were more common in cirrhotic than in noncirrhotic patients with HCC. The absence of exposure to both viruses was much less frequent in the former (7 vs. 40%, P < 0.001). Heavy alcohol intake prevailed in patients with cirrhosis (30 vs. 16.5%, P = 0.01). Alpha-fetoprotein elevation was more common in cirrhotic patients (63% vs. 31%, P < 0.001); however, the prevalence of diagnostic (> 400 ng/ml) levels did not differ significantly (24 vs. 17%) between the two groups. Extrahepatic extension of HCC was more common in noncirrhotic patients (20.5 vs. 6.5%, P < 0.001) and its independent predictors were poor cancer differentiation and absence of cirrhosis. "Asymptomatic" cancers were more frequently encountered with cirrhosis. Abdominal pain was the most common presenting symptom in both groups. Signs of hepatic decompensation prevailed in cirrhotic patients, whereas a "toxic syndrome" dominated the clinical picture of the noncirrhotic patients.

CONCLUSIONS

Hepatitis viruses are associated more with carcinogenesis of the cirrhotic than of the noncirrhotic liver. Alpha-fetoprotein is not a sensitive neoplastic marker, particularly in noncirrhotic patients. In the latter, HCC appears more advanced at diagnosis and symptoms of neoplastic toxicity are prominent.

Gross pathologic types of hepatocellular carcinoma in Italian patients. Relationship with demographic, environmental, and clinical factors

BACKGROUND

The prevalence of the different hepatocellular carcinoma (HCC) macroscopic types, and the association between these types and age, gender, blood group, alcohol and coffee intake, smoking habit, hepatitis virus markers, underlying cirrhosis, and cancer histologic type were retrospectively assessed in 416 unselected patients (321 with cirrhosis).

METHODS

The gross pathologic types of HCC were assessed by ultrasonography combined, in most cases, with computed tomography and angiography.

RESULTS

Solitary HCC was the most common cancer type (54.8%), followed by the multinodular (31%), diffuse (7.7%), and massive (6.5%) types. Cirrhosis and blood group other than O were independent risk factors for multinodular HCC (relative risk [RR] 1.6, P < 0.05; and RR 1.7, P < 0.005, respectively); the absence of cirrhosis and a heavy smoking habit were risk factors for the massive type (RR 4.9, P < 0.001; and RR 3.3, P < 0.01, respectively); and blood group O for the solitary type (RR 1.4, P < 0.001). The prevalence of highly undifferentiated cells increased as the tumor size did, so that grade IV cell atypia was associated with massive size of the carcinoma (P < 0.05). In cirrhotic patients, advanced liver dysfunction was associated with diffuse HCC (P < 0.05). As far as solitary HCC is concerned, the tumor size was greater in noncirrhotic than in cirrhotic patients (7 +/- 0.4 cm versus 4.8 +/- 0.15, P < 0.001).

CONCLUSIONS

In Italian patients, HCC presents most frequently as a solitary nodule. The presence or absence of cirrhosis, blood group, and smoking habit can influence the likelihood of developing certain HCC shapes. The probability of harboring highly undifferentiated cells increases as the cancer increases in size. In cirrhosis, advanced liver dysfunction may predispose to diffuse HCC.

Depressed immune function in patients with cirrhosis before emergence of hepatocellular carcinoma

Hepatocellular carcinomas 1 cm in diameter with high or low echogenicity can be detected on ultrasonography and confirmed on fine-needle biopsy, but it is still very difficult to detect small hepatocellular carcinomas with isoechogenicity. In this study, we assessed lymphokine-activated killer cell activity and interferon-gamma production prospectively every 1 to 3 mo for 23 +/- 4 mo (mean +/- 1 S.D.) in 227 patients with cirrhosis. Transient depression of lymphokine-activated killer activity was detected in 43 patients (defective lymphokine-activated killer group), and hepatocellular carcinoma was detected in 24 cases before the end of the 18-mo follow-up. Twenty-one (87.5%) of the 24 hepatocellular carcinoma patients were included in the defective lymphokine-activated killer group. Defective lymphokine-activated killer activity was detected more than 6 mo before detection of a space occupying lesion in the liver or elevation of alpha-fetoprotein level above 400 ng/ml. Serum alpha-fetoprotein level was elevated above 400 ng/ml in only five cases in which hepatocellular carcinoma was detected as a space-occupying lesion. Our results indicate that sequential assessment of lymphokine-activated killer activity may be a predictor of hepatocellular carcinoma and that the depression of immune function in cirrhotic patients is a serious risk factor for hepatocellular carcinoma emergence.