The Relevance of Biomarkers, Risk Factors and Gene– Environment Interactions in Disease: Scientific Developments and Therapeutic Approaches

PHILIP V. PEPLOW AND JAMES D. ADAMS JR

1.1 Introduction

An increasing prevalence of obesity, hypertension and cardiovascular disease, and diabetes has been reported for children and adults in many developed and developing countries. These conditions are interlinked and are associated with both immediate and long-term health effects.

Childhood obesity has more than doubled in children and tripled in adolescents in the past 30 years. In the United States of America the percentage of children aged 6–11 years who were obese increased from 7% in 1980 to nearly 18% in 2010. Likewise, the percentage of adolescents aged 12–19 years who were obese increased from 5% to 18% over the same time period. In 2010, more than one third of children and adolescents were overweight or obese. Overweightness and obesity are caused by caloric (energy) imbalance with too few calories expended for the amount of calories consumed, and are affected by various behavioral, genetic and environmental factors. Obese youth are more likely to have risk factors for cardiovascular disease such as high cholesterol or high blood pressure. Obese adolescents are more likely to have prediabetes, a condition in which blood glucose levels indicate a high risk of developing diabetes. Data from the National Health and Nutrition Examination Survey (NHANES) on the prevalence and control of hypertension in the USA between 1960 and 2008 have shown that the prevalence is higher in older individuals, non-Hispanic blacks, and women. It is unclear why non-Hispanic blacks are more likely to become hypertensive. Both genetic and environmental factors are probably important, one of which may be a greater likelihood of low birth weight, which appears to predict higher blood pressures in adulthood. NHANES data from 2007 to 2008 showed a 28 to 30% prevalence of hypertension in the 18 year and older population in the USA. This equates to approximately 65 million adult hypertensives in the USA, which is markedly higher than the estimated 43 million from the 1988–1991 NHANES survey. Thus, there has been a 50% increase in the number of adult hypertensives over a decade. With the increased prevalence of hypertension there has also been a marked increase in obesity and body mass index, and it has been estimated that one-half of the relative increased prevalence of hypertension may be due to an increased weight of the average individual. In surveys similar to the NHANES in Canada and Europe, the age- and sex-adjusted prevalence of hypertension ranged from 20 to 55%. Analysis of world-wide data suggested that 26% of the world adult population had hypertension in 2000. Rates of hypertension appear to be similar in developed and developing countries. Worldwide, hypertension is not adequately controlled, and control is worse in lower income countries.

There is increasing evidence that adult hypertension has its origins during childhood, as childhood blood pressure predicts adult blood pressure. Hypertension in childhood and adolescence may contribute to premature atherosclerosis and the early development of cardiovascular disease. Identifying children with hypertension and successfully treating their hypertension will have an important impact on long-term outcomes of cardiovascular disease. The definition of childhood hypertension is based on the normative distribution of blood pressure in healthy children. This contrasts with adult hypertension which is primarily defined by clinical outcomes (i.e. risk of cardiovascular disease and mortality) from large trials of antihypertensive therapy. This clinical definition cannot be applied to children because cardiovascular events other than left ventricular hypertrophy do not typically occur until adulthood. Body size is the most important determinant of blood pressure in children and adolescents.

In 2010, an estimated 19 million persons in the USA had diagnosed diabetes and another 7 million had undiagnosed diabetes. Since 1990, the prevalence of diagnosed diabetes in the USA has risen sharply among all age groups, both sexes, and all racial/ethnic groups for which data are available. The substantial increase in the prevalence of diagnosed diabetes is likely the result of improved survival of persons with diabetes and increasing diabetes incidence. Nationally representative data suggest that mortality among adults with diabetes in the USA declined substantially between 1997 and 2006, and at a faster rate than among adults without diabetes. This decline is paralleled by improvements in the health of persons with diabetes, including lower levels of risk factors for complications (e.g. hyperglycemia, uncontrolled blood pressure), decreased rates of complications associated with increased risk of death, and improvements in quality of care and medical treatments.

The main driver of the increase in diabetes prevalence is the increase in the incidence of diabetes in the USA since 1990. Increasing incidence could be the result of many factors, including changes in diagnostic criteria, enhanced detection of undiagnosed diabetes, demographic changes in the population (e.g. aging of the population and growth of minority populations who are at greater risk for diabetes), and an increase in the prevalence of risk factors for the development of diabetes (e.g. obesity and sedentary lifestyle). The increase in diabetes prevalence coincides with the increase in obesity prevalence in the USA. Continued surveillance of diabetes prevalence and incidence, its risk factors, and prevention efforts is important to measure progress of prevention efforts. Strategies that target the entire population and high-risk groups are needed to reverse the trend of increasing diabetes prevalence. Many governments and public health planners still remain largely unaware of the current magnitude and future potential for increases in diabetes and its serious complications in their own countries. In addition to diabetes, the condition of impaired glucose tolerance is a major public health concern, both because of its association with diabetes incidence and its own association with an increased risk of cardiovascular disease.

From the above it is clear that an urgent need exists to identify as early as possible those children and adults at risk of developing obesity, hypertension and cardiovascular disease, and diabetes and to discover new improved therapies for treating these conditions and their associated complications. Complications from diabetes, which include coronary artery and peripheral vascular disease, stroke, diabetic neuropathy, amputations, renal failure, and blindness, are resulting in increasing disability, reduced life expectancy, and enormous health costs for society.

1.2 Biomarkers of Disease

In 2001, the National Institute of Health defined a biomarker as a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention or other health care intervention. The biomarker is either produced by the diseased organ (e.g. tumour) or by the body in response to disease. Before diagnosis, biomarkers could be used for screening and risk assessment. During diagnosis, biomarkers can determine staging, grading, and selection of initial therapy. Later, they can be used to monitor therapy, select additional therapy, or monitor recurrent diseases. Thus, identifying biomarkers include all diagnostic tests, imaging technologies, and any other objective measures of a person's health status. Genetics, genomics, proteomics, and modern imaging techniques and other high-throughput technologies allow more biomarkers than before to be measured. In addition, a greater understanding of disease pathways, the targets of intervention, and the pharmacologic consequences of medicine is achieved.

If a biomarker is to be used as a diagnostic test, it should be sensitive and specific and have a high predictive value. A highly sensitive test will be positive in nearly all patients with the disease, but it may also be positive in many patients without the disease. To be of clinical value, a test with high sensitivity should also have high specificity, so that most patients without the disease should have negative test results. For predicting the likelihood of disease on the basis of the test result, rather than the converse, the appropriate measures are positive and negative predictive values. However, the positive predictive value falls as the prevalence of the disease falls, so that tests for rare conditions have many more false positive results than true positive results.

Most biological markers are not simply present or absent but have wide ranges of values that overlap in persons with a disease and in those without it. The risk typically increases progressively with increasing levels; few biomarkers have a threshold at which the risk suddenly rises, so various cut-off points must be evaluated for their ability to detect disease. Cut-off points with high sensitivity producing few false negative results are used when the consequences of missing a potential case are severe, whereas highly specific cut-off points producing few false positive results are used to avoid mislabeling a person who is actually free of the disease. Sensitivity and specificity calculated at various cut-off points generate a receiver operating characteristic curve, which ideally will be highly sensitive throughout the range of specificity. The most useful clinical tests are those with the largest area under the curve.

The C-statistic, or area under the receiver operating characteristic curve, is a method used to test model discrimination. C-statistic for a multivariable model represents the probability that a case has a higher measure or risk score (or a shorter time to event in survival analyses) than a comparable control. The C-statistic measures the concordance of the score and disease state. The value of the C-statistic ranges from 0.5 (no discrimination) to 1.0 (perfect discrimination) and for the Framingham coronary heart disease risk score, the C-statistic is approximately 0.76.

When considering the efficacy of novel biomarkers in risk stratification, one approach is to determine to what extent entering the candidate biomarker into standard risk prediction models will actually increase the model's C-statistic. The value of a biomarker may also be assessed by studying how biomarker information may lead to a reclassification of individuals in low-, medium- and high-risk categories based on traditional risk factors. The ultimate goal of this approach is to refine risk stratification, and it has been emphasized when considering biomarker information that would serve to shift individuals who are in the intermediate-risk groups upwards into the high-risk category or downwards into the low-risk category. Some screening methods (e.g. Pap smears, colonoscopy) have success- fully reduced mortality. However the field of early detection is affected by problems of overdiagnosis (e.g. prostate specific antigen (PSA), mammographic screening), inadequate specificity of individual markers (e.g. carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP)), low compliance (colonoscopy), and a lack of analytical methods for discovering new diagnostic markers. High-throughput technologies have been used to assess genomic data, transcriptional data, and proteomic data and have resulted in DNA, RNA and protein biomarkers in clinical use.

1.3 Risk Factors of Disease

Many factors influence the health of a person with some functioning on an individual level (e.g. health behaviours, genetic make-up) while others act at a broader societal level (e.g. availability of health services, vaccination programs, a clean and healthy environment). All these influencing factors are known collectively as determinants of health. Health determinants can influence health of an individual in either a positive or negative way. Those determinants affecting health in a negative way are referred to as risk factors. They can increase the likelihood of developing disease or adversely affect the management of existing conditions (e.g. hypertension can increase the likelihood of developing cardiovascular disease). Positive determinants of health are known as protective factors and include high-quality nutrition, safe sexual behaviour, maintaining a healthy body weight and exercise.

The prevalence of many chronic diseases is increasing in many countries. There are many reasons for this, and they include early detection and improved treatments for diseases that previously caused premature death, lifestyle behaviours such as smoking or low-nutritional diet, and an aging population. Chronic diseases impact heavily on the use of health services and contribute to major funding pressures on the health care system.

Risk factors affect health with varied levels of severity and measuring this can be quite complex. Burden of disease studies attempt to quantify the impact of selected risk factors, individually and in combination, on individual and population health. Burden of disease is a measure used by epidemiologists, statisticians and health economists to assess and compare the relative impact of different diseases and injuries on people or populations. Risk factors carry different levels of burden and the benefits of reducing them vary for individuals and the community.

The development of chronic diseases is strongly associated with the behavioural risk factors of smoking, physical inactivity, low-nutritional diet, and the harmful use of alcohol. These behaviours can contribute to the development of biomedical risk factors such as hypertension, obesity, and hypercholesterolemia. The top five risk factors for males and females in Australia in 2003 were smoking, hypertension, high body weight, hyper-cholesterolemia, and physical inactivity.23 Some chronic conditions themselves are considered to be risk factors for other conditions, e.g. diabetes, where there are known links between diabetes and increased likelihood of cardiovascular, eye and kidney disease.

A recent study conducted in Australia revealed that almost all individuals aged 15 years or over have at least one risk factor (99%) and most persons have three. More than one-third of persons have at least one of the risk factors of daily smoking, risky alcohol consumption or obesity. A higher proportion of males than females reported having five or more risk factors. Persons living in the most socioeconomic disadvantaged areas have greater numbers of risk factors than those living in less disadvantaged areas. Persons living in the major cities have fewer risk factors than those living elsewhere. As the number of risk factors increases, so does the likelihood of reporting certain chronic diseases. Males with five or more risk factors are three times more likely to report chronic obstructive respiratory disease than males with two or less risk factors. Females with five or more risk factors are three times more likely to report stroke, and two-and-half times more likely to report depression than females with two or less risk factors. Studying risk factor combinations that are prevalent in the community complements information about the prevalence of individual risk factors and the number of risk factors. Knowledge about patterns in multiple risk factors can assist health professionals treating persons with chronic disease to develop management regimens and indicate where interventions might best be targeted. Having multiple risk factors can affect the speed at which a condition progresses and develops into a new condition. The combination of hypertension, dyslipidemia and diabetes accelerates atherogenesis, thereby leading to blockage of the artery. Controlling those risk factors can stabilize lesions and slow progression. Some risk factors can be the result of others e.g. low-nutritional diet or inadequate physical activity can contribute to obesity.

1.4 Gene–Environment Interactions in Disease

Studies of gene–environment interactions aim to describe how genetic and environmental factors jointly influence the risk of developing a disease. Information is required on both elements of the relationship. Genetic pre-disposition can be inferred from family history, phenotype (e.g. skin colour) or direct analysis of DNA sequence. Environmental and lifestyle factors are measured in epidemiologic studies using self-reported information. This information is obtained by interview or questionnaire, from records or direct measures in individuals (e.g. anthropometry) or biomarker-based inference on environmental exposures. Until recently, many studies of genetic predisposition obtained little information on environment and lifestyle. Similarly, many epidemiologic studies of unrelated individuals (known as association studies) did not obtain blood samples or other sources of DNA that would allow direct assessment of genetic vvariation. More recently, in both family-based and association studies, collection of both genetic and environmental data is becoming more common so that the interaction between the two can be examined.