Changing aetiological factors of hepatocellular carcinoma and their potential impact on the effectiveness of surveillance.

BACKGROUND: The aetiological factors of hepatocellular carcinoma may vary over time. AIMS: The study assessed the potential impact of the aetiological factors on the effectiveness of surveillance in real-world patients. METHODS: Multicentre, cross-sectional study enrolling consecutive hepatocellular carcinoma cases during a six month period. RESULTS: 1733 cases (1311 prevalent and 422 incident) were recruited (mean age 68.6 years; 46.1% cases over 70 years; 73.9% males; 95.3% with cirrhosis); 63.0% were hepatitis C virus positive and 23.7% were virus negative. Amongst incident HCCs, 34.5% were single ≤3cm and 54.4% met the Milan criteria; 61.6% were diagnosed during surveillance; virus negative patients showed the lowest rate of surveillance (51.0%). Surveillance was an independent predictor of detecting single HCCs ≤2cm (O.R.=5.4; 95% C.I.=2.4-12.4) or HCCs meeting the Milan criteria (O.R.=3.1; 95% C.I.=1.9-5.2). Compared with an earlier Italian survey, there was a higher proportion of elderly subjects (P<0.01), Child-Pugh class A cases (P<0.01), of virus-negative patients (P<0.01) and with single tumours ≤3cm (P<0.01) and a lower prevalence of hepatitis C virus positive individuals (P<0.01). CONCLUSION: HCC is characterised by a growing prevalence of elderly patients and cases unrelated to hepatitis virus infections. The application of surveillance must be implemented, particularly amongst non-viral patients.

Is it time to reconsider the BCLC/AASLD therapeutic flow-chart?

PURPOSE: Recommendations of the Barcelona Clinic Liver Cancer (BCLC) therapeutic flow-chart, endorsed by the American Association for the Study of Liver Diseases (AASLD), are the most applied worldwide. Over recent years, however, several referral centers have questioned some of the BCLC treatment allocations and proposed alternative strategies. The present study plans to review and discuss these suggestions, with the aim to evaluate whether there are well-grounded reasons to reconsider some of the BCLC/AASLD recommendations. METHODS: A search was made into the MEDLINE database, focusing on randomized controlled trials, meta-analysis reviews, case-control studies, concordant clinical trials on novel therapies and studies reporting the opinion of respected experts. Their results and conclusions were compared stage by stage with BCLC/AASLD recommendations. RESULTS: In stage 0 (very early, or single <2 cm, or carcinoma in situ, Child A) radiofrequency should replace resection. In stage A (early, or single or three nodules up to 3 cm, Child A-B) radiofrequency and resection should expand their indications. In stage B (intermediate, or multinodular, Child A-B) resection and transplantation should expand their indications, while intra-arterial therapies are changing from conventional to selective treatments. In stage C (advanced, portal invasion or extrahepatic disease, Child A-B) systemic therapies should offer previously unknown promising options. CONCLUSION: In our opinion, so much evidence leads to suggest it is time to reconsider several BCLC/AASLD recommendations. Some treatments are comparable in results but vary in costs, local availability, or complication rates.

Prediction of asymptomatic cirrhosis in chronic hepatitis C patients: accuracy of artificial neural networks compared with logistic regression models.

OBJECTIVE: Models based on logistic regression analysis are proposed as noninvasive tools to predict cirrhosis in chronic hepatitis C (CHC) patients. However, none showed to be sufficiently accurate to replace liver biopsy. Artificial neural networks (ANNs), providing a prediction based on nonlinear algorithms, can improve the diagnosis of cirrhosis, a syndrome characterized by complex, nonlinear biological alterations. We compared ANNs with two logistic regression analysis-based models in predicting CHC histologically proven cirrhosis. METHODS: Liver biopsy was obtained in CHC patients of two different cohorts (an internal cohort including 244 patients and an external cohort including 220 patients). One hundred and forty-four patients from the internal cohort served as a training set to construct ANNs and a logistic regression model (LOGIT). These two models and the aspartate aminotransferase-to-platelet ratio index (APRI) were tested in the remaining 100 patients (internal validation set) and in the external cohort (external validation set). Diagnostic performances were evaluated by standard indices of accuracy. RESULTS: In the internal validation set, ANNs, LOGIT, and APRI showed similar discrimination powers (0.88, 0.87, and 0.87 respectively). However, ANNs showed the best positive predictive value (0.86 vs. 0.67 and 0.56) and positive likelihood ratio (40.2 vs. 13.4 and 8.4). In the external validation set, the discrimination power of ANNs (0.76) was significantly higher than those of LOGIT (0.67) and APRI (0.67). CONCLUSION: Compared to conventional models, ANNs performance in predicting CHC cirrhosis is slightly better and more reproducible.

Diagnostic value of HSP70, glypican 3, and glutamine synthetase in hepatocellular nodules in cirrhosis.

UNLABELLED: Hepatocellular nodules in cirrhosis include regenerative (large regenerative, LRN) and dysplastic (low and high grade, LGDN and HGDN) nodules, early and grade 1 HCC (eHCC-G1), and overt HCC. The differential diagnosis may be particularly difficult when lesions such as HGDN and eHCC-G1 are involved. We investigated the diagnostic yield of a panel of 3 putative markers of hepatocellular malignancy such as HSP70, glypican 3 (GPC3), and glutamine synthetase (GS). We selected 52 surgically removed nonmalignant nodules (15 LRNs, 15 LGDNs, 22 HGDNs) and 53 HCCs (10 early, 22 grade 1, and 21 grade 2-3) and immunostained them for HSP70, GPC3, and GS. The sensitivity and specificity of the individual markers for the detection of eHCC-G1 were 59% and 86% for GS, 69% and 91% for GPC3, and 78% and 95% for HSP70. We identified 2 main phenotypes: (1) all negative, seen in 100% LRN and LGDN, 73% HGDN and 3% eHCC-G1; (2) all positive, a feature detected in less than half the eHCC-G1. Using a 3-marker panel, when at least 2 of them, regardless which, were positive, the sensitivity and specificity for the detection of eHCC-G1 were respectively 72% and 100%; the most sensitive combination was HSP70+/GPC3+ (59%) when a 2-marker panel was used. CONCLUSION: The adopted panel of 3 markers is very helpful in distinguishing eHCC-G1 from dysplastic nodules arising in cirrhosis.

Association between heterozygosity for HFE gene mutations and hepatitis viruses in hepatocellular carcinoma.

Hepatitis B virus (HBV) and hepatitis C virus (HCV) infections are strong and independent risk factors for hepatocellular carcinoma (HCC) development. Patients with hereditary hemochromatosis (HH) are considered at risk of developing cancer. However, the interaction between HFE gene mutations and hepatitis viruses for HCC development has not been systematically searched for. To assess the interaction between HFE gene mutations and exogenous risk factors in the risk of HCC occurrence, a case-only approach, in which just a series of patients is enrolled, was used. Three hundred three cirrhotic patients (231 males, 72 females) from five liver units in different geographic areas of Italy, who developed HCC during regular follow-up between January 1999 and March 2003, and whose blood DNA was available, were analyzed. In all subjects, hepatitis B surface antigen (HBsAg), anti-HCV and HFE gene mutations were assayed; alcohol intake was recorded by history. The interaction between HFE genotypes and hepatitis viruses for HCC was estimated by multivariate analysis adjusting for the confounding effect of alcohol intake, area of residence and months of follow-up. Of the 303 HCC cases, 12 (4.0%) were heterozygous for the C282Y mutation, 93 (30.7%) for the H63D, and 198 (65.3%) homozygous for the wild allele. Multivariate analysis showed that C282Y heterozygous males were 3.8-fold (95% CI=1.0-15.2) more likely to be HBV positive and that H63D heterozygous females were 6.0-fold (95% CI=1.2-113.8) more likely to be HCV positive than wild type subjects. In conclusion, given the association between C282Y mutation and HBV infection in male patients with HCC, a careful evaluation and follow-up should be considered in the C282Y-positive subjects with hepatitis B virus related liver disease. The interaction between the H63D mutation and HCV, observed only in women, may reflect a higher sensitivity to H63D-induced iron metabolism abnormalities and a reduced antioxidant capability in the presence of an even minor increase of iron which may occur as a consequence of the coexistence of hepatitis C infection and heterozygosity for HH.

Methylation framework of cell cycle gene inhibitors in cirrhosis and associated hepatocellular carcinoma.

One of the main regulatory pathways reported to be altered in hepatocellular carcinoma (HCC) is that of cell cycle control involving RB1 gene-related cell inhibitors. We investigated p14(ARF), p15(INK4B), p16(INK4A), p18(INK4C), and RB1 genes in a series of HCCs and associated cirrhosis with the goal of ascertaining their pattern of inactivation by gene methylation. Thirty-three HCCs, adjacent nonneoplastic cirrhotic tissues, and 6 HCC cell lines were studied. Cirrhoses (25 of 33, 76%), HCCs (31 of 33, 94%), and 3 of 6 (50%) cell lines showed 1 or more methylated genes. Cirrhoses (17 of 33, 51%) had more frequently than HCCs (11 of 33, 33%, P =.01) only 1 methylated gene. With the exception of p18(INK4C) the genes under study showed promoter methylation with frequency ranging from 82% (p16(INK4A) in HCC) to 33% and 39% (p15(INK4B) and p16(INK4A) in cirrhoses). In cases with only 1 methylated gene, p15(INK4B) in cirrhosis (8 of 17, 47%) and p16(INK4A) in HCC (10 of 11, 91%) were the more frequently altered. An optimal correlation was found between p15 and p16 gene methylation and complete protein loss in HCC detected by immunocytochemistry, whereas a partial loss of the same proteins was a feature of methylated cirrhoses. Inactivation by DNA methylation of several genes of the RB1 pathway is common to cirrhosis and HCC. An early pattern of methylatory events (1 methylated gene) is a feature of cirrhosis rather than HCC, whereas an advanced one (> or = 3 methylated genes) is characteristic of malignancy. Early methylation changes seem to involve p15(INK4B) and p16(INK4A) in cirrhosis and p16(INK4A) in HCC. In conclusion, a stepwise progression of methylating events is a feature of the sequence cirrhosis-HCC and contributes to the process of hepatic carcinogenesis with potential clinical implications.