Epicenter Department of Epidemiology Newsletter

Special Edition

David Hunter, MBBS, MPH, SCD, FRACPHM Vincent L. Gregory Professor in Cancer Prevention, Emeritus Richard Doll Professor of Epidemiology and Medicine Nuffield Department of Population Health, University of Oxford

168th Cutter Lecture on Preventive Medicine Adventures in the Environment & Genes December 12, 2018 Special Edition 168th Cutter Lecture Lecturer: David Hunter, MBBS, MPH, ScD, FRACPHM

Cutter Lecture on Preventive Medicine Since 1912, the Cutter Lecture on Preventive Medicine has been one of the most respected presentations, especially in the field of epidemiology. The lectures are administered by the Department of Epidemiology at the Harvard T.H. Chan School of Public Health according to the bequest from John Clarence Cutter, MD (1851-1909), a graduate of the Harvard Medical School. He specified that the lectures be delivered in Boston, free of charge to medical professionals and the press. Covering a range of public health topics, the lectures remain dedicated to enhancing the physical and social welfare of the world’s population.

David Hunter David Hunter studied medicine at the University of Sydney, before moving to Harvard University for 33 years where he was the Vincent L. Gregory Professor of Cancer Prevention. He is the Richard Doll Professor of Epidemiology and Medicine, and director of the Harvard-Oxford Program in Epidemiology. His early research was on HIV transmission in East Africa, and subsequently he was involved in collaborative studies of nutrition and HIV pathogenesis, while also studying diet and cancer etiology in large scale prospective studies and founding the Pooling Project of Prospective Studies of Diet and Cancer.

As Director of the Harvard Center for Cancer Prevention, he developed a sample handling and genotyping laboratory to explore genetic associations with cancer and gene-environment interactions. He founded the Program in Genetic Epidemiology and Statistical Genetics at Harvard. He was co-chair of the steering committee of the NCI Breast and Prostate Cancer Cohort Consoritum (BPC3) between 2003 and 2012, was co-director of the NCI Cancer Genetic Susceptibility Markers project focused on genome-wide association studies, and was an Eminent Scholar at the NCI between 2004 and 2009. From 2009-2016 he was Dean for Academic Affairs at the Harvard TH Chan School of Public Health, and in 2015-2016 he was Acting Dean. He is one of about 3000 “highly cited researchers” worldwide according to Thomson-Reuters.

Adventures in the Environment and Genes

Dr. David Hunter, the Richard Doll Professor of Epidemiology and Medicine at the University of Oxford, presented the 168th Cutter Lecture on December 12, 2018, at the Harvard T.H. Chan School of Public Health. Hunter fell in love with epidemiology here 34 years ago, completed his doctorate at Harvard, and worked at HSPH for the next three decades.

Hunter guided Cutter Lecture attendees on a brief journey through time to reflect on persistent epidemio- logical questions involving how much environmental and genetic factors influence incidence and outcomes of common diseases.

2 2 Hunter became interested in medical research through studying the health effects of exposure to high altitude among climbers. His first professional adventure involved a study on HIV, specifically the relationship between HIV acquisition and illness. Early on in the epidemic, scientists and medical professionals believed HIV, similar to other infectious diseases, would lead to only a fraction of those with the virus developing full-blown AIDS. But in a collaboration with Victor De Gruttola, published in The Lancet in 1986, Hunter showed a rapid and alarming rate of progression to the onset of AIDS among those who had received blood from donors infected with HIV. Subsequent studies conducted by researchers in Africa in that late 1980s suggested a higher incidence of HIV infection among women using oral contraceptives, complicating the daunting challenges faced by the continent in the midst of battling the spread of HIV and an increasingly devastating AIDS epidemic.

However, follow-up studies conducted by Hunter and colleagues in Africa documented that male partner sexual behavior, the presence of other sexually transmitted diseases, and social environment were stronger indicators for acquiring HIV than contraceptive methods.

Along with this early work on HIV, Hunter focused his doctoral thesis on skin cancer, laying the foundation for what would become a primary field of study throughout his career—gene-environment interaction in disease. He and colleagues collaborating on the Nurses’ Health Study looked at risk factors for basal-cell carcinoma in relation to number of severe sunburns (environmental) in people with different hair color (genetic). They discovered that the risk of cancer was definitively higher among those with red hair who also had experienced numerous sunburns. Diet Is Important but Does Not Cause Cancer

The next phase of Hunter’s adventure involved large-scale studies of the relationship between diet and cancer. At the time, a prevailing hypothesis was that breast cancer was associated with total fat intake. But numerous prospective studies showed very little connection between any particular food groups and breast cancer. Founding the Pooling Project of Prospective Studies of Diet and Cancer, now led by Stephanie Smith-Warner, a series of papers examined the relationships between diet and cancer across the major cohort studies. The only robust association for breast cancer was increased alcohol intake and heightened risk of breast cancer. Several studies that concluded daily consumption of milk likely protects against colon cancer. Other studies showed connections between healthy eating and substantially reduced risk for cardiovascular disease and diabetes, but no consistent associations for cancer in midlife. In 2018, the World Cancer Foundation summarized evidence of convincing associations between alcohol intake, obesity, and height across several kinds of cancer, but limited evidence that diet in middle life alters risk of most cancers.

Research Methodologies Fuel Ongoing Controversy

The debate continues about the best methods of exposure measurement in observational epidemiological studies. Many say people lie or do not know the answers to questions asked on questionnaires, and only “objective” measurements can be relied on. Physical activity, for example, is difficult to measure, and wearables such as accelerometers are proposed as a solution. Hunter says both sources of information are necessary. For large studies, most self-reported data is reliable for behaviors people can tell scientists about, but other measures are needed for exposures (e.g. genetic variants) they are not able to report to researchers. In addition, objective measures can be just as prone to confounding. A better measurement of physical activity does not solve the problem that active people are different from sedentary people. Some experts recommend Mendelian randomization to address the confounding problem, but this approach does not always provide the right answers. However, in the Breast and Prostate Cancer Cohort Consortium looking at body mass index and breast cancer risk, if scientists had relied only on Mendelian Randomization analysis, they would have mistakenly advised women to gain weight after menopause!

No technique is bulletproof, says Hunter. A combination of approaches is required, and many studies are just too small, he noted. A good big study is always more reliable than a good small study.

3 3 Special Edition - 168th Cutter Lecture: David Hunter

4 4 Special Edition - 168th Cutter Lecture: David Hunter

5 5

Genetic Risk Scores Deliver Valuable Information

We know a lot more about disease risk than they did fifteen years ago. Scientists are now able to examine nearly two hundred common genetic variants associated with breast cancer risk. While any one of these variants alone does not increase a woman’s risk of breast cancer, taken together, they can be combined into powerful and robust polygenic risk scores, and the more of these reproducible adverse associations that are present, the higher the risk.

At the high end of the risk distribution, scientists predict that about 1 percent of women have a 30 percent lifetime risk of breast cancer, while at the other end of the spectrum, about 1 percent of women have an extremely low risk of ever having the disease. Other analyses that consider all risk factors together estimates that nearly 20 percent of women over 60 years old account for approximately 40 percent of breast cancer cases.

These studies, which are still underway, suggest possible recommendations for less frequent screening requirements for women at lower risk, or earlier starting age for mammography for high-risk women. Knowledge of an increased risk of acquiring a common but preventable or treatable disease is valuable and actionable information for healthcare consumers. The ability to provide this kind of data represents a significant step forward in addressing common diseases that impact millions of people.

A 2016 study concluded that women at higher preexisting genetic risk for breast cancer can benefit by minimizing or eliminating just one modifiable lifestyle behavior (e.g. drinking or weight gain). Similar findings have been reported in lifestyle impacts on those with genetic risk for coronary heart disease. So, knowing your genotype is helpful, and favorable lifestyle changes will positively impact people at risk for common diseases. Genes and the environment both influence common disease risk. We must comprehensively understand both for effective prediction of individual risk.

“All of the genetic textbooks need to be rewritten. The genetic architecture is much more complex than one gene, one protein, one enzyme paradigm. Thosands of variants across the genome contribute to the risks associated with common diseases.” -David Hunter

Challenges for future epidemiological studies include gathering prospective data on less common diseases and disease subtypes. As part of the ongoing effort to analyze larger populations, cases must be aggregated in multi-study consortia, increasing sample sizes from thousands to tens of thousands or more. Researchers must leverage digital technologies for information capture and remote automated outcome assessments.

Committed to facing these kinds of challenges head on, Hunter is currently involved in a new program in the UK called SUCCEED, which will track five million healthy Britons in an attempt to accelerate earlier detection of disease. The intent is to shift healthcare from treating illness in late stages to pre-empting disease in very early stages. Scientists will test and validate new diagnostic and predictive tools at scale, and discover entirely new indicators of disease. The program aims to deliver information back to the population, helping UK citizens and medical professionals effectively leverage data like polygenic risk scores in order to improve health outcomes.

Hunter says the most rewarding years in epidemiology are still ahead. “We have the best job in the world,” he claims, “because we never know what’s coming next. There are always more mountains to climb, more to discover. The adventure continues.”

6 6 CUTTER LECTURES 1912-2018 1912 William H. Park • John F. Anderson 1913 George C. Whipple • Mark W. Richardson 1914 Charles V. Chapin • Cressy L. Wilbur 1915 Joseph Gold-berger • Victor C. Vaughan 1916 George W. McCoy • Simon Flexner 1917 Martin H. Fischer • Ludwig Hektoen 1918 Elmer V. McCollum 1919 Harry E. Mock • Alice Hamilton • T.M Legge 1920 Theobald Smith • Jules Bordet 1921 Alonzo Taylor • Clemens von Pirquet 1922 Charles Wardell Stiles • Alfred F. Hess 1923 Bella Schick • Sir Arnold Theiler 1924 Thorvald Madsen 1925 Watson S. Rankin 1926 George E. Vincent • Sir Arthur Newsholme • F. Neufeld 1928 Wade Hampton Frost • C. Levaditi 1929 Charles E. A. Winslow • Edwin O. Jordan 1930 Lafayette B. Mendel • William Mansfield Clark • M. Weinberg 1931 Karl F. Meyer 1932 Leslie T. Webster • Charles Armstrong 1933 Louise I. Dublin • Eugene L. Opie 1934 Karl Landsteiner • Anthony J. Lanza • Charles Sidney Burwell 1935 Milton J. Rosenau 1937 Thorvald Madsen 1938 Andrija Stampar 1939 Frederick F. Russell 1940 Ludwig Hektoen • James B. Murphy 1941 Hendrik Dam • Ernest W. Goodpasture 1942 Sir Wilson Jameson 1943 Lowell T. Coggeshall • Alfredo Sordelli 1944 Anatol A. Smorodintzev 1945 Sir Alexander Fleming • Donald Hunter 1946 Sir Lionel Whitby 1947 Thomas M. Rivers • Haven Emerson 1948 William N. Pickles 1949 James C. Spence 1950 Douglas H. K. Lee 1951 Hugh M. Sinclair • Dugald Baird 1953 A. Bradford Hill • Sir Gordan Covell 1954 Thomas Parran 1955 James V. Neel 1957 Charles M. Fletcher 1958 Sir Macfarlance Burnet 1959 Harold F. Dorn 1960 Benjamin Pasamanick • Thomas McKeown 1961 Archibald L. Cochrane 1962 Alexander D. Langmuir 1963 Paul M. Densen 1964 Robert E. Shank 1965 Einar Pederson 1966 Jerome Lejeune 1967 Sir Richard Doll 1969 Donald D. Reid 1970 Kung-pei Chen 1971 C. Ronald Lowe 1972 E. S. Anderson • Sir Richard Doll 1974 H. R. Neveanlinna 1975 E. George Knox 1976 Patricia Jacobs 1977 Sir Peter Medawar 1979 Gilbert Beebe 1982 Dimitrios Trichopoulos 1986 Peter Smith 1991 Hans-Olov Adami 1993 Olli Sakari Miettinen 1994 Lorenzo Tomatis 1996 Philip Cole 1997 David Schottenfeld • Lee N. Robins 1998 Timothy D. Noakes 2000 George Davey Smith • Eric S. Lander 2001 Alexander Walker • Joseph F. Fraumeni, Jr. 2002 Jonathan Samet • Jeffrey P. Koplan 2003 Malcom C. Pike • Jeffrey Sachs 2004 Nicolas Wald • R. Palmer Beasley 2005 Sir Richard Peto • Leslie Bernstein 2006 Nancy Mueller • K. Srinath Reddy 2007 Harvey Fineberg • George Klein 2008 Chien- Jen Chen 2009 Ming Tsuang • Harald Zur Hausen 2010 Norman Breslow 2011 Nubia Munoz • Mark McClellan 2012 Margaret Spitz•Alice Whittemore 2013 Moyses Szklo . Bruce Psaty 2014 Robert N. Hoover • Kenneth J. Rothman 2015 Dame Valerie Beral 2016 Duncan C. Thomas, Walter C. Willett 2017 Sir David Cox, Jeffrey M. Drazen 2018 Margaret A. Hamburg, David Hunter

7 Harvard T.H. Chan School of Public Health Department of Epidemiology 677 Huntington Ave, Kresge 9th Floor Boston, MA 02115 hsph.harvard.edu/epidemiology

How to Subscribe To subscribe, please send your email address to:

Scan this QR code Coppelia Liebenthal: [email protected] to access previous Newsletters

@HarvardEpi Contact The Department of Epidemiology hsph.harvard.edu/epidemiology Tel: 617.432.6477 Editorial Board: Eric DiGiovanni & Coppelia Liebenthal 8