Dr Lewis is a senior lecturer in Genetic Epidemiology at Bristol Medical School (University of Bristol).
It is clear that many dietary factors, body weight and physical activity influence cancer risk. While some risk factors have been identified, the evidence for others is not certain. In addition, the mechanisms for how various risk factors influence cancer risk are not clear. This is why we put a team together to develop a novel method for reviewing the evidence from human, animal and cell studies so that we can improve our understanding of how lifestyle risk factors affect cancer risk and survival.
What we did
We set out to develop methods to:
- identify potential biological pathways between diet, physical activity, body fatness and cancer
- collate all the existing evidence from human, animal and cell studies on a specific biological pathway
We were a multidisciplinary team of more than 20 scientists with expertise in statistics, large genetics, bioinformatics, cancer biology and nutrition. We held five one-day workshops over 18 months to brainstorm ideas. At these workshops we presented data, carried out exercises in small groups and took part in discussions. In the intervening periods, we carried out searches and applied our methods to a case study to test our ideas.
The challenges we faced
We faced many challenges in this work:
- how to identify the biological pathway by which a particular exposure causes cancer
- how to cope with the enormous wealth of data generated in our search for studies
- how to determine the relevance of animal studies to human disease
- how best to integrate all the evidence
We adapted different approaches to tackle each of the challenges – one approach was to enlist Dr Clinton and Dr Hurstings from the US to help with determining the relevance of animal studies to humans.
We also developed a new online tool for prioritising biological pathways for further investigation and we developed a new graph for displaying data from studies that are different but address a common question.
We have tested our new methods in a case study of milk and prostate cancer, which has already been published, and our methods have been independently tested by two groups, one in Germany and one in the Netherlands.
The methodology and tools we have developed could identify interesting biological pathways which may explain how a lifestyle or dietary factor causes cancer and assess the weight of the evidence underlying a particular biological mechanism. This will allow us to understand more about the causes of cancer and how we can prevent it more than ever before.