Crossing continents – The China Kadoorie Biobank
Biobanks are big news and big science. Built from huge cohorts of volunteers who are interviewed, measured and then monitored for decades, biobanks enable scientists to create a detailed picture of genetic, lifestyle and environmental influences on health.
Decades of collaboration between NDPH researchers and Chinese scientists have now grown into the China Kadoorie Biobank (CKB) – a long-term prospective study involving more than half a million participants recruited from ten diverse areas across China. This large study is now providing a wealth of high quality data to enable scientists from around the world to investigate the causes of common diseases that are relevant not just for China but elsewhere.
Going back to the roots
The history of the NDPH’s large-scale scientific collaboration with China starts with Professor Sir Richard Peto, one of the UK’s leading epidemiology researchers and co-director of the Clinical Trial Service Unit and Epidemiological Studies Unit (CTSU) at the NDPH. He started visiting China in the early 1980s, establishing research collaborations with Chinese scientists who wanted to understand more about the patterns of common diseases in their rapidly-changing country and how best to treat them.
Starting with relatively short-term population studies and clinical trials, for example looking for correlations between lifestyle factors and health across different regions of China, their ambitions grew as the collaborative relationships strengthened. The first major findings were published at the end of the 1990s, including a paper in the British Medical Journal looking back over the causes of death of around a million Chinese people.
One of the key players in the collaboration was Zhengming Chen, who first met Richard Peto and Rory Collins in China in 1987. He trained as a doctor in Shanghai and is now professor of epidemiology at the NDPH but back in 1987 he came to Oxford as a new graduate student in the CTSU, working with a cohort of only 10,000 Chinese factory workers. This grew into further collaborative projects in China, including big clinical trials and even bigger epidemiological studies.
One of their main ambitions was always to establish a huge, open-ended study, gathering data and samples from many hundreds of thousands of Chinese people and following them into the future to see who died early from particular diseases. It was a challenging concept at first – unlike most large-scale studies at the time, Chen and Peto weren’t planning on investigating a specific condition such as lung cancer or heart disease, but were just interested in gathering the data and seeing what stories they could reveal.
Building the biobank
Throughout the 1990s, it was impossible to find any funding agencies that shared this bold vision. After many rounds of failed attempts, Chen and Peto approached the Kadoorie Charitable Foundation (KCF) in Hong Kong in 2001, through the internationally renowned tobacco activist Dr Judith Mackay. The Foundation agreed to provide one-off funding to set up the study, on condition that it carried its name.
This marked the birth of the China Kadoorie Biobank (CKB) after a decade-long effort. At the time, the researchers had already had close collaborations with the Chinese Centre for Disease Control and Prevention (China CDC) and the Chinese Academy of Medical Sciences (CAMS) on various other projects, but this would be their most ambitious, challenging but invaluable collaborative study.
It may seem strange for a UK research team to take an interest in the health of the Chinese population, but the reasoning makes sense. Previous large-scale population studies have focused solely on predominantly Caucasian populations in Europe and the UK. Although many of the conditions affecting Chinese people are becoming increasingly similar to those in the West, the proportion of the population affected can be quite different.
For example, strokes (particularly the haemorrhagic kind) are much more common in Chinese people. Similarly, throat and stomach cancers are also more common in China than in the UK or US, which have a higher rate of breast and bowel cancers. Moreover, even within China there are well-documented varying rates for many common conditions across different regions, for reasons that are still not fully understood. For example, high blood pressure is the most important determinant of stroke, but there is little difference in average blood pressure between people in the UK and China, or between populations in different regions of China, even though the rates of stroke vary by more than 10-fold.
The large differences in lifestyle and environment between China and the West allows important biological details to emerge. For example, average Body Mass Index (BMI) is lower in China than in the West, so it’s possible to pick out disease risks in leaner people. Similarly, levels of so-called ‘bad’ LDL cholesterol (LDL-C) are much lower in China, especially those living in rural areas, providing unique opportunity to assess biological optimal levels of LDL-C that are associated with least risks of cardiovascular disease and potential ways to prevent and treat it. With 10 carefully selected study areas across China, CKB can shed important light on the risks and causes of disease in Chinese populations.
To create the biobank, researchers gathered a huge amount of information from each participant. This ranged from physical measurements (such as height, weight, blood pressure, lung function, ECG) to detailed questionnaires about diet (including preferences for particular foods and spices) smoking, alcohol and tea drinking (even their preferred temperature for drinking it!), physical activity, exposure to household air pollution, personal and family medical history, sleep habits, reproductive history, physical and mental illnesses.
This was just the beginning. More recently, the CKB team is starting to analyse stored blood samples using novel, high-throughput technologies, looking at the levels of certain chemicals, proteins and small molecule metabolites that are important indicators of health and nutrition. They are also studying the participants’ DNA – data that contains hugely valuable information about the links between genes and disease risk.
So far they’ve analysed the genetic makeup of 100,000 participants, with the aim of genotyping the full half a million participants within the next two to three years. At the moment the researchers are looking at variations around 20 million specific sites in the genome (known as SNPs, or single nucleotide polymorphisms), but ultimately they plan to sequence the whole genome, which will yield 3 billion genetic data points for each participant.
The team also collected information from medical insurance and health records, as well as information about drug prescriptions and hospital admission data, enabling them to build a detailed picture of the kinds of diseases the participants are suffering or eventually dying from. They’re going back every few years to do what’s known as resampling – revisiting five per cent of the participants to repeat the same measurements and questions as the baseline survey, and to take new measurements (such as bone density, carotid artery health, wearable devices for physical activity and exposure to air pollution).
Setting up a large collaboration that reaches across borders and cultures is a complex challenge. However, the study came together remarkably quickly and smoothly, taking just four years to recruit more than half a million middle-aged Chinese men and women from ten diverse urban and rural regions, from the frozen north to the tropical south and from the rich coast cities to the poor rural villages in the west. These people are then being followed into the future for as long as possible.
Although Chen’s leadership of the research project and careful planning and piloting, were crucial, he explains that one of the key elements to the project’s success actually lay in the behind-the-scenes IT systems – the first three posts he recruited for the study in Oxford were the project manager and two software programmers.
In total, Chen’s team in Oxford has developed over 60 bespoke IT systems, enabling the study to be entirely paperless – a first for any large-scale population study worldwide – which has improved data quality and study management. The techniques, methods and tools developed by the CKB team have also informed the development of other biobanks around the world, including the 500,000-strong UK Biobank run by scientists at the University of Oxford.
Thanks to support from the Chinese Government, and rapid development of research facilities in China, the CKB team has been able to analyse DNA samples in China’s major new genetics facilities, such as the BGI in Shenzhen. With special permission, the CKB team has also managed to store half of the plasma samples in Oxford, which will enable the research team take advantages of novel, high-throughput technologies in the UK. The data generated will then be made available to researchers in China, further enhancing collaboration and capacity development.
Expanding collaboration and revealing relationships
After a decade of follow-up, the CKB is an unparalleled, rich data set which is contributing to a wide range of projects. Since the early years, the team’s collaborations have expanded to involve many universities and institutes in China and elsewhere around the world, and the data are gradually being made available through open access for researchers everywhere to sift through.
Over the past five years, the research teams in Oxford and China have published many important findings that will inform our understanding about causes, prediction and prevention of many common diseases. Some of the early results have focused on some well-researched topics such as diet, but revealing many novel findings. For example, the CKB showed that fresh fruit consumption can cut the chances of developing diabetes or cardiovascular disease, spicy foods reduce the risk of dying prematurely, and even discovered that daily tea drinking is linked to a lower risk of ischaemic heart disease.
Returning to the original roots of Richard Peto’s work on the hazards of smoking, the CKB has continued to build on earlier studies bringing together the case against tobacco. An important paper published in The Lancet in 2015 showed that tobacco was responsible for 1 million premature deaths in China in 2010, and that figure is set to rise still further over the coming decades unless there are widespread initiatives to cut smoking rates.
Other papers have focused on the roles that body weight, blood pressure, physical activity, alcohol drinking, and household air pollution play in the risk of serious health conditions, while a recent publication showed that Chinese women who breastfeed their babies have a ten per cent lower risk of developing cardiovascular disease later in life. The team also carried out a major analysis of the burden of diabetes across the Chinese population, finding that although the condition is more common in urban areas, it is more likely to cause an early death in rural parts of the country due to poor management of the condition. This led to a major editorial by the then-Director General of the World Health Organisation, Dr Margaret Chang, calling for major policy changes to tackle the problem.
Using genetic data, the team has also developed a range of research projects focusing on certain genetic variations that are specific to East Asians, which alter the activity or levels of particular proteins that might be potential drug targets. These natural experiments in human populations can improve the drug development process, by prioritising targets that are most likely to be effective, improving safety, and identifying other conditions that might also be treatable. This approach has already proven useful for identifying potential drug targets for tackling heart disease, according to two recent papers from the CKB team.
Finally, it’s not just NDPH researchers that are digging through the CKB data in search of clues about health and disease. The team recently launched a dedicated data platform, encouraging the scientific community from China and around the world to apply use this incredible resource to help answer their own research questions and find new ways to improve global health.
Find out more on The China Kadoorie Biobank website