The 2018 publication of Diet, Nutrition, Physical Activity and Cancer: a Global Perspective is a landmark in the science of cancer prevention. In the Third Expert Report from World Cancer Research Fund (WCRF) and the American Institute of Cancer Research (AICR), there are important changes from the 2007 Second Expert Report:

  • an important shift in emphasis to a more holistic focus
  • fine-tuning the approach of assessing and interpreting evidence
  • emerging evidence of note

Emphasising a more holistic approach

The Cancer Prevention Recommendations in the Third Expert Report are similar to those in the 2007 Second Expert Report. However, they incorporate an important shift in emphasis in the Expert Panel’s interpretation of the evidence.

Through the years, the Continuous Update Project (CUP), and its predecessors the First and Second Expert Reports, have identified many specific foods (such as processed meat) and components of foods (such as alcohol) in the human diet that increase or decrease the risk of one or more particular cancers.

However, it appears increasingly unlikely that specific foods, nutrients or other components of foods are themselves important singular factors in causing or protecting against cancer: rather, different patterns of diet and physical activity combine to create a metabolic state that is more, or less, conducive to the acquisition of the genetic and epigenetic alterations that lead to the phenotypical structural and functional alterations in cells described by the Hallmarks of Cancer.

In humans, as with all organisms, the normal physiological and metabolic state is subject to external and endogenous challenges (stresses). Nutrition is an important component of the body’s capacity to withstand these stresses and avoid the development of diseases; in the absence of frank nutritional deficiency, this resilience is not dependent on the singular effect of specific nutrients. A more holistic focus on the determinants of resilience to external and endogenous challenge may be more fruitful than a continuing search for specific dietary factors that may cause or protect against cancer.

Furthermore, studies evaluating the impact of adherence to the Cancer Prevention Recommendations from 2007 have shown that the more people adhere to those recommendations, the greater the reductions in the risk of specific cancers, of cancer as a whole and of death from any cause.

For all these reasons, therefore, the Expert Panel emphasise the importance of recognising that, while following each individual Recommendation is expected to offer cancer protection, the most benefit is to be gained by treating them as an integrated pattern of behaviours relating to diet and physical activity, and other factors, that can be considered as a single overarching ‘package’ or way of life.

The consistency in the Recommendations since 2007 increases confidence in the evidence base and in the advice given to policymakers, the scientific community, health professionals and the public.

Cancer Prevention Recommendations

Fine-tuning how we assess evidence

There has been an increase in the overall amount of evidence since 2007. This has enabled the Expert Panel to fine-tune its approach to assessing and interpreting evidence:

  • Growth in the number of cohort studies and the number of cases in existing cohorts, as well as improvement in the quality of these studies, has enabled the Expert Panel to concentrate more on evidence from cohort studies, which are considered the best source of evidence on cancer prevention. This growth has also provided greater confidence in the accumulated evidence.
  • Results from pooled analyses of cohort studies have been particularly helpful in adding evidence for subgroup analyses.
  • Where possible when reviewing evidence on diet and nutrition, the Expert Panel has increasingly considered the effects of dietary patterns. This is important because people do not eat foods in isolation but in combination, to form an overall diet or eating pattern, which itself is related to other health-linked behaviours such as smoking or physical activity. While sophisticated epidemiological or statistical techniques may help to minimise the inevitable effect of the resulting confounding, confidence in the nature of the truly causal exposure must always be greater for the aggregated set of dietary factors and other behaviours than for any single food, nutrient or other behavioural marker. Therefore, even where there is no direct evidence on dietary patterns the Expert Panel has aimed to interpret evidence on specific foods in relation to dietary patterns.
  • There is more evidence on subtypes of cancer now, such as oesophageal cancer, and therefore more conclusions on the effect of diet, physical activity or body fatness on differential risks of these subtypes. Evidence on subtypes is still emerging. This is important because different exposures might influence the risk of different subtypes in different ways.
  • It has been possible to use non-linear analyses more to identify thresholds, or plateaus. For example, there may be a threshold in the level of exposure below which there is no association with the risk of cancer and above which there is. This has proved important, for example, when making Recommendations on the level of consumption of alcoholic drinks.
  • Stratified analyses have provided valuable insights, for example, when considering the effect of exposures stratified by smoking status.

Emerging evidence of note

Emerging evidence that is particularly noteworthy includes:

  • Growth in the amount of high-quality data that is available has allowed more sophisticated analyses of how effects on cancer risk change with the level of exposure – for instance on the shape of dose-response associations. For fruit and vegetables, for example, emerging evidence suggests it’s the people who consume the least, who eat very little or none of these foods, who are most at risk of developing certain cancers. It may therefore be more important for these people to increase their consumption levels than for people who already eat more than one or two portions per day.
  • The influence of height on cancer risk is becoming more apparent. However, height itself is unlikely to be the actual cause of cancer. It is most likely a marker for developmental factors related to growth and metabolism operating from the earliest stages of life to influence cancer susceptibility. More research is needed to build understanding of precisely how this might happen.
  • The importance of the life course in general is emerging more strongly. There is evidence that greater adult height predicts higher risk of several cancers; and for breast cancer specifically, that greater birthweight is associated with higher risk, while greater body fatness in young adulthood predicts lower risk. However, more research is required to help further understand the mechanisms.
  • Evidence on cancer survivors is accumulating, though is still at an early stage, and more is needed, particularly from well-conducted trials (some of which are under way).