Adiposity and kidney disease: observations from large blood-based prospective cohorts across a wide range of different populations

The Kidney Research UK 2013 patient survey identified 'prevention of kidney disease' and 'cause of specific kidney diseases' as the second and third priority for renal research after 'finding a cure for kidney disease'. This fellowship focus demonstrably aligns with these priorities. Work I have recently completed using data of 1.4 million adults in England showed that being overweight and obese is a key cause of kidney disease in the UK. Tracking these people for 7 years, nearly 1-in-100 developed permanent serious kidney damage and 1-in-1000 required dialysis or a kidney transplant. Compared to lean people, those who were overweight were at moderately increased risk of permanent serious kidney damage (their risk was increased by about 30- 40%), but risk was doubled among those who were moderately obese (ie, those who had a body-mass index [BMI] between 30-35 kg/m2) and was tripled among those with severe obesity (BMI above 35kg/m2). Diabetes is a common cause of serious kidney damage in the UK and worldwide, and is one of the mechanisms by which obesity may cause permanent kidney damage. Nevertheless, this study found that excess weight increased the risk of kidney damage by a similar amount in people with or without diabetes, and also in people who had 'well-controlled' blood pressure. Taking into account how common overweight and obesity are in the UK, these data suggested that at least one-third of all permanent serious kidney damage in adults in the UK could be due to being overweight or obese, and that despite modern treatments for diabetes and raised blood pressure, obesity remains a major cause of serious kidney disease.

My overall research aim is therefore to study by how much being overweight or obese and deposition of fat in the abdomen (ie, central obesity) increase the risk of developing temporary, permanent and/or progressive kidney damage across a wide range of populations, and then explore the potential causal mechanisms. This will be done using some of the world's largest population studies which have measured people's weight and body fat and collected blood for genetics, then followed them for many years to see if and how kidney problems develop. The following steps in the research process are necessary:

1. Completing the identifying and confirming of people who develop new kidney disease in over 1 million people assessed and then followed in 3 large population studies (China-Kadoorie Biobank, UK Biobank and the Mexico City Prospective Study).

2. Assessing how body weight and central obesity predict the risk of developing different kidney diseases.

3. Confirming adiposity is a causal risk factor for kidney disease, and then focusing on potential mechanisms, where there is some controversy. This work will include use of novel gene-based computational methods in the studies.

Long-term, these studies will be capable of relating how different types of obesity and kidney disease affect risk of other diseases thought to be associated with both conditions, including heart disease, infection and cancer.

Key applications and benefits of this fellowship, beyond personal development and the formation of a new research program focused on obesity, diabetes and kidney disease, include providing:

1. Extending the expertise of a kidney doctor who already has knowledge of how to design and conduct large trials, and providing him with the ability to use specialist genetic analyses to test whether risk factors for kidney disease represent a probable cause, rather than an association with kidney disease. This ensures future clinical trials will test the best questions;

2. Defining reliable kidney outcomes for UK Biobank, China-Kadoorie Biobank and Mexico City Prospective Study users;

3. Providing new and major opportunities for the pharmaceutical industry to test whether biological pathways may reduce or increase risk of kidney diseases.

1. TO DERIVE RELIABLE OUTCOMES (RENAL AND NON-RENAL) FOR THE MEXICO CITY PROSPECTIVE STUDY AND CKD/ACUTE KIDNEY INJURY (AKI) OUTCOMES FOR 2 OTHER LARGE BLOOD-BASED PROSPECTIVE COHORTS (UK BIOBANK [UKB] & CHINA-KADOORIE BIOBANK [CKB])
For the Mexico City Prospective Study, fieldwork is necessary to confirm successful linkage of participants to hospitalisation records, and verify clinical diagnoses. For CKB, a traditional hospital notes based adjudication process will be undertaken in about 5000 recorded outcomes, including hospitals admissions recorded as CKD, 'acute renal failure', 'unspecified renal failure'. In UKB, data from secondary and primary care will be combined to identify CKD stages and hospitalisation for AKI and these algorithms tested using the baseline measured kidney function.

2. TO ASSESS ASSOCIATIONS BETWEEN MEASURES OF GENERAL/CENTRAL ADIPOSITY AND RISK OF CKD AND ACUTE KIDNEY INJURY ACROSS DIFFERENT POPULATIONS
To quantify the magnitude of associations between measures of general/central adiposity and risk of CKD/AKI across different populations using data from 3 blood-based cohorts (CKB, UKB, and the Mexico City Prospective Study) which have complete baseline adiposity measurements and a large number of prospectively collected outcomes.

3. TO EXPLORE MECHANISMS FOR THE ASSOCIATION BETWEEN INCREASED ADIPOSTY AND CKD RISK, INCLUDING MENDELIAN RANDOMIZATION EXPERIMENTS
Diabetes is a mechanism by which adiposity may cause kidney diseases, but increasing BMI is a CKD risk factor in people without diabetes. To investigate what is causing this association, genotyping data from SHARP, CKB, UKB and eventually the Mexico City Prospective Study will assess whether adiposity is likely to be a cause of non-diabetic kidney disease or not. Whether genetically elevated blood pressure is a mediator of CKD progression or incident CKD will then be assessed. Genotyping/biomarker data will then be used to explore new hypotheses.

1. Renal trials: The datasets (and my development) will help ensure future renal trials are developed using robust epidemiological evidence, including Mendelian randomization experiments to assess causality, combined with reliable classical epidemiological methods to predict realistic outcome event rates and treatment effect sizes.

2. Population Health Impact: Worldwide, it is estimated that 2 billion people are overweight or obese and 600 million are obese. In the UK, about two-thirds of adults are overweight or obese and 1-in-3 are obese. Levels of obesity are also particularly high in North America, Western Europe, Australasia, parts of Central and South America, the Middle East and some Pacific Islands. Obesity is of particular concern in the low and middle income countries where obesity levels have risen sharply in recent years and the countries have insufficient resources to treat the epidemic of diabetes and increased cardiovascular/infection risk which will result. For example, the obesity epidemic in Mexico and the consequent high prevalence of poorly controlled diabetes is estimated to be a direct or indirect cause of 1-in-3 adult deaths between the ages of 35-74 years between 1998 and 2014. Many of these deaths were attributed to renal disease and its consequences. Assuming causality, this recent work suggests that 1-in-3 cases of 'serious and irreversible' kidney disease in the UK is a direct or indirect result of being overweight or obese.

3. Policy makers: Much like the successes that population health has had in influencing tobacco policy worldwide, generating reliable estimates of the risks associated with increased adiposity from different populations and among different types of people will help inform policy makers' decisions worldwide of the importance of preventing and treating obesity. It will also inform them of the size of the benefits one might expected if the average weight of the population was reduced. Our direct links with the Mexican Ministry for Health and the influence the Mexico City Prospective Study has already had on diabetes policy in Mexico is an example of the major impact this type of research can have on policy.

4. Health professionals: There is lack of reliable prospective data on the optimum measure of obesity to inform public health policy and clinicians, so the National Institute for Clinical Excellence (NICE) is currently limited in the recommendations it offers with regard to measures for assessing obesity. The 2014 NICE guidelines stated that body-mass index (BMI) is "a practical estimate of adiposity in adults" and that clinicians should "think about using waist circumference, in addition to BMI, in people with a BMI less than 35 kg/m2." NICE's Public Health Interventions Advisory Committee also recently identified that there is "a lack of studies that directly compared the association between BMI and the differential health risks for people of different ethnic groups apart from for diabetes" and there is "a lack of studies on health outcomes using waist circumference as the explanatory variable." This research proposal directly addresses some of these uncertainties with respect to kidney disease (and so will in time be expanded to other diseases associated with obesity), and may help identify who should be targeted for particular interventions to reduce risk of kidney disease.

5. Epidemiologists: will benefit from the enhancement of UK Biobank and China-Kadoorie Biobank to include reliable follow-up for renal disease outcomes. More prospective doctorial students may also start to use these resources to study kidney disease and its complications.

6. Understanding of healthcare data in Mexico would not only support the prospective study, but could also improve the quality of future Mexican 'big data' healthcare research (analogous to the vision of the Farr Institute in the UK).