The burden of liver cancer and cirrhosis caused by hepatitis C is likely to continue to grow in the United States despite curative treatment for hepatitis C, and people who have cirrhosis at the time they are cured of hepatitis C will require long-term monitoring for liver cancer, studies presented this week at the 2015 AASLD Liver Meeting in San Francisco show.
However, curing hepatitis C results in a two-thirds reduction in the subsequent risk of liver cancer for people who have cirrhosis at the time they are cured. People aged 55 or over and those with diabetes or a high body mass index are likely to be at higher risk of liver cancer after a cure.
Sustained virologic response after treatment for hepatitis C has been defined as a cure, and is now the outcome of direct-acting antiviral treatment for the vast majority of people who receive the newest interferon-free treatment combinations.
However, many people who are being treated have suffered many years of liver damage as a result of hepatitis C and remain at elevated risk of hepatocellular carcinoma (liver cancer). What impact is curative treatment having on the incidence of liver cancer?
Until now most of the available data has come from small cohort studies, and in order to give a more precise estimate of the risk, researchers used the VA Hospitals Database to look at the risk of developing liver cancer in all VA patients cured of hepatitis C between 1999 and 2009 – a population of 10,817 people. (These people were treated with pegylated interferon or interferon and ribavirin, prior to the introduction of direct-acting antivirals, but there is no evidence that the type of treatment used to cure hepatitis C affects the subsequent likelihood of liver cancer.)
This population was compared to 11,380 people who underwent hepatitis C treatment but who did not achieve a sustained virologic response.
10,738 of those cured were defined as being free of HCC at the end of treatment and eligible for analysis, and of these 100 developed liver cancer.
The investigators found an incidence of liver cancer of 3.27 cases per 1000 person-years of follow up in those who achieved SVR12, compared to an incidence of 13.2 per 1000 person-years of follow up in those who were not cured, a risk reduction of 64.2%.
Liver cancer occurred much more frequently in people who had cirrhosis at the time of cure: there was an annual incidence of 1.54 cases per 1000 patient-years in people with cirrhosis, compared to 0.28 per 1000 patient-years in those without cirrhosis. Multivariable analysis showed that people with cirrhosis had around four-and-a-half times the risk of developing liver cancer, regardless of other risk factors, when compared to those without cirrhosis at the time of cure (hazard ratio 4.45, 95% confidence interval 2.53-7.82, p < 0.0001).
Diabetes was also associated with an elevated incidence of liver cancer, as was genotype 3 hepatitis C infection. The incidence of liver cancer was greatly elevated in those aged 65 and over, particularly when compared to under-45s and those aged 45-54 (0.95 cases per 1000 py compared to 0.07 and 0.21 cases per 1000 py respectively). People aged 65 and over were four-and-a-half times more likely to develop liver cancer than those under 55, a multivariable analysis showed (hazard ratio 4.69, 95% confidence interval 2.03 – 10.78, p = 0.0003).
AASLD President Gyongyi Szabo, speaking at an opening press conference, warned that although “we can eradicate virus, which makes patients clinically more stable, we haven't eliminated cirrhosis, [so patients with cirrhosis] should be candidates for screening [for HCC]. Direct-acting antivirals will reduce HCC risk, and the way they will do that is by treating patients before end-stage cirrhosis, before getting cirrhotic. If we treat patients who already have cirrhosis, we can make the liver a little better, but the risk of liver cancer will remain.”
Burden of liver cancer and cirrhosis still rising in the United States
These results need to be considered in the context of an overall rise in the burden of both cirrhosis and liver cancer among people living with hepatitis C. A second study of the Veteran’s Affairs population, conducted by Lauren Beste of the University of Washington School of Medicine and colleagues, found that the incidence of both cirrhosis and hepatocellular carcinoma are likely to continue rising in the United States until 2021, even taking into account the impact of treatment.
This analysis sought to separate the contributions of infectious causes and non-infectious causes of cirrhosis (such as non-alcoholic fatty liver disease) to the overall burden of cirrhosis, and to determine the accuracy of previous forecasts of the evolving burden of liver disease in the US population. Lauren Beste noted that whereas all-cause mortality among adults aged 45-54 has declined by around one-third in France, the United Kingdom, Canada and Germany over the past 25 years, it has remained above 400 deaths per 100,000 among US Whites and the death rate appears to be rising as a consequence of substance use, suicide and chronic liver disease and cirrhosis. In contrast the death rate among the US Hispanic population of the same age has followed the Canadian and Western European pattern.
The investigators looked at all cases of liver cancer or cirrhosis diagnosed in Veterans Affairs patients between 2001 and 2013 and categorised people to causal categories based on the presence of absence of hepatitis C, and in the absence of hepatitis C to categories on the basis of diagnosed alcohol use disorder, diabetes or clinical conditions such as primary biliary cirrhosis.
Forty-eight per cent of all diagnoses of cirrhosis occurred in people with hepatitis C, of which approximately two-thirds also had an alcohol use disorder. Alcoholic liver disease was the second biggest cause of cirrhosis, accounting for 30% of cases. Non-alcoholic fatty liver disease accounted for 15% of cases, hepatitis B for only 2.1% of cases in this population. The proportion of cirrhosis attributable to hepatitis C had risen from under 40% in 2001 to almost 50% by 2013, with a corresponding decline in alcohol-related cirrhosis.
Although the incidence of cirrhosis has not grown, the prevalence of cirrhosis attributable to hepatitis C doubled in the period 2001-2013. If current trends continue, the researchers estimate, the prevalence of cirrhosis is likely to peak in 2021.
Liver cancer has followed a similar pattern, but the effect of hepatitis C is even more pronounced. 68% of liver cancer diagnoses in 2013 occurred in people with hepatitis C, compared with just over 40% in 2001, and overall incidence rose two-and-half-fold between 2001 and 2013. Mortality due to liver cancer tripled in the same period, and the increase was overwhelmingly attributable to hepatitis C.
Lauren Beste warned that although the major cause of liver cancer and cirrhosis is curable, US health systems must prepare for further increases in both conditions.
A third study, this one assessing changes in the prevalence of cirrhosis among people with hepatitis C in the National Health and Nutrition Survey cohort, a very large sample of the general population in the United States, found that the prevalence of cirrhosis had increased across three time periods since 1988. This change was driven by ageing, but also by an increasing prevalence of metabolic factors such as diabetes and high body mass index (> 30). Based on the progression of liver disease observed among the NHANES cohort since 1988, investigators from Stanford University estimate that the size of the US population of people with hepatitis C who have cirrhosis has increased from approximately 170,000 in the period 1988-1994 to 370,000 in the period 2007-2012.
The investigators warned that due to the lack of symptoms in earlier stages of cirrhosis, many people may be unaware that they have hepatitis and advanced liver disease, and doctors should be particularly alert to the risk of cirrhosis in people with hepatitis C who are older, those with diabetes and those with high body mass index.
Additional reporting by Liz Highleyman, hivandhepatitis.com
Beste L et al. Trends in the burden of cirrhosis and hepatocellular carcinoma by underlying liver disease in US Veterans from 2001-2013. AASLD Liver Meeting, San Francisco, abstract 90, 2015.
El-Serag H et al. Incidence and predictors of hepatocellular carcinoma following sustained virological response: a national cohort study. AASLD Liver Meeting, San Francisco, abstract 90, 2015.
Udompap P et al. Increasing prevalence of cirrhosis among US adults with chronic HCV infection:
Results from NHANES 1988-1994 and 1999-2012. AASLD Liver Meeting, San Francisco, abstract 88, 2015.