Current issues of ACP Journal Club are published in Annals of Internal Medicine


Chronic hepatatis C virus infection was independently associated with increased risk for hepatocellular carcinoma

ACP J Club. 1992 May-June;116:91. doi:10.7326/ACPJC-1992-116-3-091

Source Citation

Simonetti RG, Cammà C, Fiorello F, et al. Hepatitis C virus infection as a risk factor for hepatocellular carcinoma in patients with cirrhosis. A case-control study. Ann Intern Med. 1992 Jan 15;116:97-102.



To determine whether chronic hepatitis C virus (HCV) infection is an independent risk factor for hepatocellular carcinoma and whether infection with HCV increases the cirrhosis-related risk for hepatocellular carcinoma.


2 case-control studies of matched pairs.


A referral-based hospital in Italy.


Study 1 included 212 patients (mean age, 62 years; 76% men) with hepatocellular carcinoma who were consecutively admitted to the center between June 1982 and December 1988. Diagnosis was confirmed by biopsy or by elevated α-fetoprotein and compatible findings on ultrasonography or computed tomography. The control group included 212 concurrently hospitalized patients with chronic nonhepatic diseases who were matched by sex and age to case patients. Study 2 included the subset of 197 patients (mean age, 62 years; 75% men) from study 1 who had hepatocellular carcinoma and cirrhosis diagnosed clinically or by biopsy findings. The control group was 197 patients with cirrhosis who were matched by sex and age to case patients.

Assessment of risk factors

Stored serum samples were assayed for anti-HCV by enzyme-linked immunosorbent assay (ELISA) to determine exposure to HCV. All patients were also tested for hepatitis B surface antigen (HBsAg) and antibody to hepatitis core antigen (anti-HBc). Alcohol intake was assessed from medical records for patients with carcinoma or cirrhosis.

Main outcome measures

Relation of potential risk factors to the development of hepatocellular carcinoma.

Main results

In study 1, 151 patients (71%) with hepatocellular carcinoma were anti-HCV positive compared with 11 controls (5%) with nonhepatic diseases (odds ratio [OR] 42, 95% CI 22 to 95). In multivariate analysis, anti-HCV was associated independently with hepatocellular carcinoma (OR 69, CI 15 to 308), as were HBsAg (OR 8.7, CI 1.5 to 50) and anti-HBc (OR 4.2, CI 1.7 to 10.7). In study 2, 146 patients (74%) with hepatocellular carcinoma and cirrhosis were anti-HCV positive compared with 122 patients (62%) with cirrhosis alone (OR 1.8, CI 1.1 to 2.8). Multivariate analysis confirmed that anti-HCV (OR 2.0, CI 1.3 to 3.2) and HBsAg (OR 2.0, CI 1.0 to 4.2) were independently associated with hepatocellular carcinoma.


HCV infection was independently associated with hepatocellular carcinoma. The difference in the strength of association in comparisons with noncirrhotic and cirrhotic control patients suggests that the association may be causally linked through cirrhosis.

Source of funding: Not stated.

Address for article reprint: Dr. R.G. Simonetti, Divisione di Medicina, Ospedale Cervello, via Trabucco 180, 90146 Palermo, Italy.


Until recently, suspicion of an association between exposure to non-A non-B (NANB) hepatitis virus and primary hepatocellular carcinoma was based on anecdotal clinical reports of patients who developed hepatocellular carcinoma after chronic liver disease attributed to post-transfusion hepatitis. With improved diagnostic techniques, recognition of specific antibodies has been simplified, and almost a score of centers have now reported a high prevalence of anti-HCV, the predominant NANB virus, in patients with hepatocellular carcinoma.

The findings of Simonetti and colleagues suggest that cirrhosis may be an intermediate step between HCV exposure and the development of hepatocellular carcinoma. To determine whether cirrhosis is an obligatory intermediate stage for HCV-associated hepatocellular carcinoma in some patients, we will need more information on HCV exposure in patients who develop hepatocellular carcinoma without evidence of cirrhosis.

We should be cautious in making causal inferences from case-control studies about the observed associations. A causal link between HCV and hepatocellular carcinoma is consistent with a similar association involving hepatitis B, but confirmatory studies for HCV need to be done. The use of hospitalized controls in this study would tend to underestimate the true association, but future case-control studies should attempt to control for the duration of cirrhosis as well. Doing this may be difficult in retrospective studies because malignant transformation may occur in chronic HCV infection a decade after cirrhosis (1).

The questions of whether HCV causes hepatocellular carcinoma and of the role of cirrhosis can also be addressed using laboratory tools such as the polymerase chain reaction to probe for the presence of viral DNA sequences in the liver of cirrhotic patients and in hepatocellular carcinoma tumor tissue. Clinical epidemiologic studies like this one combined with powerful molecular tools should soon provide a definitive answer.

Robert H. Resnick, MD
Harvard UniversityBoston, Massachusetts, USA


1. Kiyosawa K, Sodeyma T, Tenaka E, et al. Interrelationship of blood transfusion, non-A non-B hepatitis and hepatocellular carcinoma: analysis by detection of antibody to hepatitis C virus. Hepatology. 1990;4:671-5.