Throughout my life I have enjoyed the privilege of having significant and formative mentors. The first was my biology teacher in high school who instilled in me a love of science and convinced me to become a science teacher like him. By chance, one evening in the UK in 1972 I attended a lecture by Denis Burkitt. Quoting from his obituary, “Denis had two claims to fame – as the discoverer of the first cancer to be caused by a virus, and as the man who forced the world to take dietary fibre seriously. These achievements are made more extraordinary by the fact that he had no training in scientific research, in nutrition or in communication. He was the archetype of the unselfconscious, gifted amateur.” His talk on the geographical distribution and correlates of “Burkitt’s lymphoma” in Africa motivated me to return to university to undertake a research master’s degree in medical geography. These studies formed my introduction to epidemiology, which has grown to become my lifelong fascination. I went on to take my PhD at the University of Tasmania in 1980 on the epidemiology of asthma. Towards the end of this time I met my next significant mentor, Norelle Lickiss, who was Professor of Community Medicine at the Medical School in Hobart. She recruited me as a post-doctoral fellow and introduced me to cancer epidemiology by putting me to work on her large population-based study of blood cancers in Tasmania.
In 1983, one of Norelle’s friends, Dr Nigel Gray, the Director of the Cancer Council of Victoria, approached her to see if she knew of any potential candidates to work on the Victorian Cancer Registry. Legislation recently passed in Victoria at the time obliged medical institutions to report all cancer cases and deaths, thus providing the registry with a complete picture of cancer incidence throughout the state. In August that year I moved to the Cancer Council in Melbourne to take up the position of Director of the Victorian Cancer Registry. My mentors there were Nigel Gray and Professor Richard (Dick) Lovell; both epidemiology enthusiasts, who recognised the paramount importance of epidemiology to cancer control. With their support and guidance, and some discipline, in 1986 I became the Director of the Cancer Epidemiology Centre which subsequently grew into the present Cancer Epidemiology and Intelligence Division.
My initial task at the Cancer Epidemiology Centre was to identify where to invest efforts on causal factors associated with lifestyle and, therefore, potentially prevent them. Our knowledge of the causes of cancer at that time was fairly primitive, but smoking was well established, as was sun exposure, some occupational exposures and viruses. Diet and nutrition were considered to explain between 30-50% of all cancers but detailed information adequate to inform prevention was scant. In order to close these gaps in knowledge about diet and cancer, we devised the Melbourne Collaborative Cohort Study (Health 2020) which recruited over 41,000 Melbournians between 1990 and 1994, a quarter of whom were Greek and Italian migrants, since we wanted to have a wider range of dietary items and intakes in the cohort. We chose a prospective design (in which you collect information on people’s lifestyle factors before cancer diagnosis), over a retrospective design (in which you ask people who have already developed cancer about their lifestyle in the past). We also collected blood samples from virtually every participant. These samples were collected to measure markers in the blood that would give us objective evidence of nutritional status; we did not anticipate the rapid advances in genomics that made the blood samples extremely valuable as a source of DNA. This treasure trove of DNA can be used to examine genetic risk factors and their potential interaction with lifestyle-related exposures, which has been our main focus for the past decade or so.
No-one else in Australia has a study of Health 2020’s size and variety, with blood samples from all participants. We carried out the recruitment, follow-up and analysis of the data, and formed collaborations with other researchers who can use Health 2020 data to study not just cancers but also other common diseases. Health 2020 has contributed to over 800 scientific papers, an achievement I’m extremely proud of. Among the many findings that emerged from Health 2020, I would like to emphasise the importance of waist circumference measurement in determining obesity-related cancer risks: we found that waist circumference is more strongly related to risk of cancer than the more commonly used BMI and that it is associated with several cancers. These findings suggested that distribution of body fat, rather than fat itself, is important in influencing cancer risk.
30 years on, the ABC Study is the next chapter in this story, building on the knowledge gained from Health 2020 to further improve cancer prediction and outcomes. Advancing from Health 2020’s primary focus on lifestyle and particularly diet, the ABC Study is aiming for breakthroughs that will enable more individualised cancer prevention and screening to occur. Currently, most public health interventions are based on a one-size-fits-all approach. By more accurately predicting cancer risk based on an individual's genetic profile and lifestyle, we hope to be able to deliver more customised health advice and better-targeted public health messages. Importantly, prevention strategies such as screening will be aimed only at those who will benefit from them, sparing many people from unnecessary investigations.