By ecancer reporter Clare Sansom

The World Cancer Research Fund (WCRF) is an international network of cancer charities that funds research into the relationship between diet, weight, physical activity and cancer with an aim of providing authoritative and up-to-date information on how to minimise the risk of developing the main types of cancer. It runs an ongoing Continuous Update Project, collating all authoritative epidemiological research on diet, exercise and cancer incidence for 17 different cancer types plus breast cancer survivors. This evidence is analysed and interpreted by a panel of international experts and published as a set of expert reports; the full body of this published evidence then feeds into updates of the WCRF’s comprehensive but concise Cancer Prevention Recommendations.

The WCRF has just published an extensive report summarising all published studies of the links between specific lifestyle factors and the risk of developing colorectal cancer. This report, which updates those published in 2007 and 2011 with the inclusion of newer research, collates results from 99 studies that together include over 29 million adult participants, over 247,000 of whom developed colorectal cancer. Its headline findings include, for the first time, authoritative evidence to suggest that a diet rich in wholegrains – and specifically, one involving consumption of these grains at most meals – will decrease the risk of this cancer. The most significant of the earlier findings that were confirmed in this report are that people who consume large quantities of processed meat or alcoholic drinks, or who are overweight, obese or tall, are at higher risk of developing colorectal cancers, and that physical activity is protective against cancers of the colon. Interestingly, however, this last finding does not hold for rectal cancers.

Colorectal cancer is the third most common form of the disease worldwide, and the fourth most common cause of death from cancer; about 700,000 people die from it every year. It is becoming more common in many countries: prevalence is higher in richer countries where the typical diet is ‘Western’, but the death rate is higher in low and middle-income countries, mainly due to later diagnosis. Advances in drug therapy in recent years have slightly improved the prognosis for individuals diagnosed with late stage colorectal cancer.

The detailed analysis of the published literature was carried out by a team of researchers and statisticians based at Imperial College, London, and reviewed by an independent panel of international experts. The selected studies were all randomised controlled trials, cohort and nested case-control studies of the association between colorectal cancer and a range of dietary and exercise parameters; they were all published before April 30, 2015. A total of 344 papers were analysed and these results pooled with those included in the earlier colorectal cancer reports. Meta-analyses were performed for incidence only, with the exception of some studies relating to alcohol consumption that also considered mortality. Where possible, separate meta-analyses were carried out for different subtypes (particularly separating cancers of the colon from those of the rectum); for men and women; and by geographical area. Possible mechanisms through which exposure to various nutrients, or physical exercise, could promote or prevent tumour development were also reviewed and included in the study.

In order to produce the headline conclusions, the experts graded each association as ‘convincing’, ‘probable’, ‘limited – suggestive’, ‘limited – no conclusion’ and ‘substantive effect on risk unlikely’. Associations that were graded as ‘convincing’ were those that were considered robust enough to be unlikely to be changed as new evidence emerges, and where the conclusions justified adding to global cancer prevention recommendations. These required consistent evidence from multiple high-quality studies of different types, the presence of a dose-response or similar relationship and plausible evidence for a mechanistic link. Those that were judged as ‘probable’, including that suggesting the protective effect of a diet rich in wholegrains, required generally consistent evidence from independent case-control or cohort studies.

The meta-analyses that led to the most significant conclusions published in the report can be summarised as follows:

Wholegrains

The panel evidence for a ‘probable’ protective link between wholegrain consumption and colorectal cancer came from six studies pooled in a dose-response meta-analysis involving over 8,000 subjects. All studies showed some protection; when the analysis was stratified by site the combined relative risk (RR) was 0.82 (95% confidence interval, CI 0.73-0.92) for colon cancer, but the findings for rectal cancer were not statistically significant. Wholegrains are a rich source of nutrients and of dietary fibre, which may protect against cancer through promoting synthesis of short-chain fatty acids or reducing gut transit time.

Processed Meat

The strong evidence for a positive correlation between consumption of processed red meats such as bacon and salami and colorectal cancer risk came from 13 studies of over 10,000 individuals. Ten studies were included in a dose-response meta-analysis, which concluded that the risk of developing colorectal cancer was increased by 16% per 50g processed meat per day (RR 1.16; 95% CI 1.08-1.26). Several previously published meta-analyses also reached similar conclusions. It is thought likely that several mechanisms contribute to the increase in risk, among them the high proportion of fat found in processed meat, the high temperatures that it is cooked at, and the presence of carcinogenic N-nitroso compounds.

Alcohol

The link between alcohol consumption and colorectal cancer has been well studied; the researchers located 11 new studies to be added to eight previously analysed ones. A dose-response meta-analysis of sixteen of the 19 studies showed a 7% increase in cancer risk per 10g ethanol per day. However, the response was not linear; no increase in risk could be detected with low alcohol consumption, below about 20g per day. Similar results were obtained after stratification into cancers of the colon and rectum; into men and women; and for different types of alcohol (beer, wine and spirits). The panel concluded that the evidence for a positive if small correlation between moderate or high alcohol consumption and colorectal cancer risk was strong despite the fact that there is no one well-understood physiological mechanism.

Body Fat

The link between body fat, as measured by the proxy values of body mass index (BMI) and waist to hip ratio, and colorectal cancer incidence has been even better studied than that for alcohol. The researchers identified 24 new studies of the relationship between BMI and cancer risk to make a total of 57. Thirty-eight of these studies were included in a meta-analysis, and most of these showed a strong positive correlation between high BMI and cancer risk. Similar results were obtained when waist to hip ratio was used as the proxy for body fatness. Mechanistically, increases in body fat are associated with increases in insulin levels and with a stimulation of the inflammatory response, and both of these will promote carcinogenesis.

Height

A total of 20 studies of the relationship between adult height and colorectal cancer, nine of which were new or updated since the last report. Thirteen of these were included in a meta-analysis that showed a 5% increase in cancer risk for each 5cm increase in normal adult height (RR 1.05; 95% CI 1.02-1.07). This increased risk was more significant in women than in men and, oddly, appeared slightly more significant in North American populations. It is likely that the increased risk to tall people arises at least partly from greater exposure to growth factors and greater calorie consumption in early life.

Physical Activity

Studies relating physical activity to cancer risk were divided into those that considered all such activity and those that only considered recreational activity. In both cases, a dose-response meta-analysis could not be performed because each study grouped activity into different categories or bands; instead, the meta-analyses compared the highest and the lowest activity levels reported by each. Most of the papers reported results for colon cancer only, and the few published studies of rectal cancer showed no significant effect. However, the effect of exercise on colon cancer incidence was significant with an overall decreased risk of 20% for high total physical activity and of 16% for high recreational activity. The protective effects of exercise are likely to arise through reductions in body fat, and consequently in insulin resistance and inflammation, and through stimulating digestion.

Further analyses presented in this report confirmed probable links between the consumption of dietary fibre, dairy products and calcium supplements and reductions in cancer risk, and between consumption of unprocessed red meat (beef, pork, lamb and goat) and an increase in risk. There was limited evidence to support any reduction in risk caused by eating fish, non-starchy vegetables or fruit, or by taking vitamin D or multivitamin supplements.

In conclusion, this comprehensive, authoritative report, which summarises evidence from studies of cancer risk published up to April 2015, confirms what we have believed for many years: that colorectal cancer is that one of the most preventable of tumour types. Its findings support the general recommendations of the World Cancer Research Fund to eat healthily – and particularly, in this case, to favour wholegrain products and avoid processed meat and excess alcohol – to watch their weight and to exercise more. They will be fed into the next update to the Fund’s generalised recommendations, which is due to be published in 2018.

Reference:
World Cancer Research Fund International / American Institute for Cancer Research Continuous Update Project Report (2017). Diet, nutrition, physical activity and colorectal cancer. Available at http://wcrf.org/int/research-we-fund/continuous-update-project-findings-reports/colorectal-bowel-cancer. All CUP reports are available from http://wcrf.org/cupreports.