by ecancer reporter Clare Sansom

The respected non-profit organisation World Cancer Research Fund International (WCRFI) has updated its analysis of the scientific evidence covering lifestyle factors and the risk of developing liver cancer for 2015 in a comprehensive report.

WCRFI, which was founded in 1982, is a worldwide research-based charity and a leading authority on the relationship between diet, weight, physical activity and cancer risk.

It was the first charity to seriously promote the link between diet and cancer.

It provides an expert analysis of published research in this area for a range of common and fairly common types of cancer through its Continuous Update Project, and findings from this project are used to update its public guidance on cancer prevention.

The 2015 report is an update of the liver cancer section of the Second Expert Report, which was published in 2007.

This new report was based on peer reviewed studies of diet, weight, physical activity and liver cancer published worldwide, analysed by a team of researchers at Imperial College, London, UK and independently assessed by an expert panel.

The team at Imperial was led by Teresa Norat, an epidemiologist based at the college’s Department of Public Health.

They reviewed a total of 34 independent studies, which together analysed a population of 8.2 million adults of whom 24,500 were liver cancer patients.

The key findings of the report were that a high body mass index (BMI), high consumption of alcoholic drinks, and consumption of foods contaminated by aflatoxins increase the risk of developing liver cancer, and that drinking coffee is linked with a decreased risk of the disease.

The findings on body mass index and coffee were reported by WCRFI for the first time in this update; its other findings were unchanged from the Second Expert Report.

There are other known factors that increase the risk of liver cancer not covered by this report: cirrhosis of the liver, chronic viral hepatitis, smoking, and long-term use of the combined contraceptive pill.

The report began with a summary of liver cancer types and current trends in incidence and survival.

The most common type of liver cancer is hepatocellular carcinoma, which arises from the main liver cells, the hepatocytes, and accounts for about 90% of all cases.

Rare types of liver cancer include cholangiocarcinomas, which start in the small bile ducts in the liver; hepatoblastoma and angiosarcoma.

It is the sixth most commonly occurring tumour type worldwide and the second most common cause of cancer death.

It is more common in men than in women, and is particularly common in East Asia where the median age at diagnosis is lower than it is in Western countries.

The mean age-standardised five-year survival rate for all types of liver cancer is currently reported to be 15%.

Hepatitis B and hepatitis C viruses are recognised causes of liver cancer, and carriers of hepatitis B are about 100 times more likely to develop the disease than those who are free of this virus.

Many of the studies analysed for the WCRFI report adjusted their data for smoking status and some also adjusted it for hepatitis B and C carrier status; patients with a diagnosis of cirrhosis of the liver or a history of alcohol abuse were not included in the statistics.

Most of the data analysed concerned the most common liver cancer type, hepatocellular carcinoma; limited data on the rarer types seemed to suggest different patterns of risk factors.

The report presented detailed statistical evidence for the increased risks arising from aflatoxins, alcohol consumption and BMI, and for the decreased risk arising from coffee consumption.

Aflatoxins are carcinogenic compounds that are produced by fungi from the genus Aspergillus, and which are frequently found as contaminants in some cereals, nuts and other foodstuffs.

Contamination with aflatoxins occurs more often in tropical and sub-tropical conditions and when food has been inappropriately stored.

All the case-control and cohort studies analysed by the WCRFI team found correlations between biomarkers of aflatoxin exposure and an increased risk of liver cancer, and most of these attained statistical significance.

It was not possible to perform a full meta-analysis as the methods used to obtain the data were not comparable across all the studies.

Several meta-analyses of aflatoxin risk have been published, however; the most recent of these included nine studies from China, Taiwan and sub-Saharan Africa and recorded a significantly increased risk (relative risk (RR) 4.75; 95% confidence interval (CI) 2.78–8.11).

The report also highlighted a synergy between infection with the hepatitis B virus and exposure to the most carcinogenic aflatoxin, AFB1, which has been reported in epidemiological studies.

The authors proposed several possible mechanisms for this synergy, suggesting that the carcinogenic effect of the aflatoxin can be increased by cytochrome P450 enzymes that are induced by the viral infection and by the presence of viral oncoproteins.

In summary, the panel concluded that consumption of food contaminated by aflatoxins was a convincing risk factor for liver cancer, and that these toxins should be considered a worldwide public health issue despite mainly affecting countries in Asia and Africa.

The panel identified a total of 30 publications from 19 studies that reported on the relationship between alcohol intake and liver cancer risk, 14 of which were from new or updated studies.

None of these included any patients diagnosed with hepatic cirrhosis, which is a known risk factor for this disease.

A large majority of these studies showed an increase in liver cancer risk with increasing levels of alcohol intake.

Ten out of 11 of the studies that reported on liver cancer incidence and five of the six that reported on mortality showed this increase in risk, and half of these were statistically significant; one study of mortality found an increase in risk that was significant only in men.

The researchers performed a dose-response meta-analysis of fourteen of the studies and concluded that there was a statistically significant increase in liver cancer incidence of 4% per 10 grams of alcohol per day and a smaller but still significant increase in mortality.

This effect only reached statistical significance in studies from Asia, where the number of studies and cases was much larger; it was, however, significant in both men and women.

There was some evidence for a non-linear relationship between alcohol intake and cancer risk, with a significant cause-effect relationship only becoming clear for intakes above about 45g per day.

These results were consistent both with the findings reported in the Second Expert Report and with other previously published meta-analyses.

The authors discussed possible mechanisms through which alcohol is carcinogenic to the liver; these are complex and not yet fully understood, but are closely related to the factors through which it leads to hepatitis, liver fibrosis and eventually cirrhosis.

In conclusion, the report authors propose that consumption of alcohol should be considered a convincing cause of liver cancer, at least at levels above 45g/day.

Insufficient physical activity is now known to contribute to a number of chronic diseases, including cancer, and the WCRFI researchers identified four studies linking physical activity and liver cancer risk.

These studies were not completely comparable, as they assessed slightly different parameters – total physical activity, leisure-time physical activity, walking and vigorous activity – so it was not possible to perform a meta-analysis.

Each of these studies showed that the group of subjects with the highest activity levels were at lower risk of liver cancer than those with the lowest, although the magnitude of the effects differed and some did not reach statistical significance.

Statistically significant differences between the highest and lowest activity groups were observed in three cohort studies: one of leisure-time physical activity, one of walking (in Japan) and one of vigorous activity.

The researchers considered that the beneficial effects of regular physical activity were likely to arise through its effect on reducing chronic inflammation, improving insulin sensitivity and preventing weight gain.

They concluded that physical activity is likely to have some protective effect against developing liver cancer, but that the evidence was limited, largely because only a few studies have considered this factor.

High levels of body fat are also known to lead to an increased risk of developing many cancers and other diseases.

The panel of researchers chose to use body mass index (BMI) as a measure of fat levels, while recognising that this is imperfect as it does not distinguish between lean and fat mass.

They identified a total of 22 publications from 15 studies linking body mass index and liver cancer, of which 14 studies (18 publications) were new or had been updated since the 2007 report.

Nine of 11 studies of liver cancer incidence reported a positive correlation between that and BMI, and six of these were statistically significant in both men and women.

There were only three studies of the relationship between BMI and liver cancer mortality, but all these reported positive associations between mortality and BMI.

The researchers performed a dose-response meta-analysis of twelve of the studies and derived a value for the increase in risk of 30% for a rise in BMI of 5 mg/kg2, with a steeper increase in risk at the highest BMI levels.

This was statistically significant for incidence but not for mortality, for both men and women and in all geographical areas, although it seemed to be highest in Europeans.

Mechanisms linking body fat and carcinogenesis in the liver are thought to include changes to circulating hormone levels; increased stimulation of the inflammatory response; and the development of type 2 diabetes, which is itself a risk factor for liver cancer.

The panel concluded that a high BMI should be regarded as a convincing cause of liver cancer.

The researchers also investigated a possible link between coffee consumption and the development of liver cancer.

They identified eight studies and 11 publications covering this area, of which six were new or had been updated.

Six of the seven studies that looked at liver cancer incidence reported an inverse correlation, with a higher coffee intake associated with a lower cancer risk; two of these reported a risk that was statistically significant in men but not in women.

The single study of liver cancer mortality also reported an inverse correlation, but again this was statistically significant only in men.

A dose-response meta-analysis calculated from six of the studies showed a decrease in risk of 14% per cup of coffee per day, but when the meta-analysis was stratified by sex the result was again significant only in men.

Several published meta-analyses have also shown inverse correlations between coffee consumption and liver cancer risk.

Animal studies have suggested several possible mechanisms through which coffee might provide some protection against liver cancer, linking it to a decrease in inflammatory markers, a decrease in DNA damage and the induction of apoptosis.

The researchers concluded that high consumption of coffee probably provides some protective effect against liver cancer, more significantly in men than in women.

A summary of the complete report can be found at http://wcrf.org/int/research-we-fund/continuous-update-project-findings-reports/liver-cancer

Reference

World Cancer Research Fund International/American Institute for Cancer Research, 'Continuous Update Project Report: Diet, Nutrition, Physical Activity and Liver Cancer.' 2015. Available at http://wcrf.org/sites/default/files/Liver-Cancer-2015-Report.pdf