Abstract and Introduction
Genetic advances in Type 2 diabetes (T2D) have led to the discovery and validation of multiple markers for this complex disease. Despite low predictive value of current T2D markers beyond clinical risk factors and family history, researchers are exploring the clinical utility and outcomes of implementation in practice, and testing is available via direct-to-consumer markets. Clinical utility research demonstrates high hypothetical utility to patients for motivating behavior change and potentially reducing risk. However, trials to date have not demonstrated improvements in behavioral and clinical outcomes over and above counseling based on traditional risk factors. Ongoing research in T2D genetics and associated risk-prediction models is necessary to refine genetic risk pathways, algorithms for risk prediction and use of this information in clinical care. Further research is also needed to explore care models and support interventions that address the needs of personalized risk information and sustainable preventive behaviors to reduce the rising prevalence of T2D.
Advances in genome-wide association studies (GWAS) and validation of findings have increased our knowledge of genetic markers of chronic disease quite rapidly. Currently, approximately 65 SNPs have been associated with risk of Type 2 diabetes (T2D), and research is ongoing to determine their specific functions in the pathways to T2D and associated risk factors such as obesity.[1,2] In addition, risk testing for SNPs associated with T2D is already available to the public through direct-to-consumer testing and has the potential to enter clinical practice more frequently as advances such as whole-genome sequencing (WGS) become more pervasive. Yet, much caution has been expressed in the literature regarding application of advancements in T2D genomics 'before their time'. Unlike Mendelian or single-gene traits, T2D is a multifactorial chronic disease with many environmental and clinical risk factors in addition to, or overlapping with, genomic risk. Environmental and clinical factors have traditionally been utilized in practice to assess and counsel patients regarding risk and prediabetes diagnosis. Yet, the question remains as to whether the technological genomic developments in risk testing for T2D will add to current practices in other ways, such as guiding clinical care strategies and counseling. Therefore, in this paper, we review the literature to determine the current state of the science exploring the clinical and personal utility of genomic testing for T2D risk in primary care.
When evaluating tests, several characteristics are considered: analytic validity (the accuracy of the test in measuring its intended target), clinical validity (the accuracy of the test in predicting clinical outcomes – in this case, the onset of diabetes), clinical utility (whether the test results will change the clinician's recommendations/treatment plan), personal utility (whether the test results change patients' behaviors or sense of psychological well-being) and clinical outcomes (i.e., weight and insulin resistance).[4,5] In the case of T2D genomic testing, analytic validity is well accepted. However, clinical validity is a source of controversy with one of the most prominent cautions being that T2D genomic tests have not added significantly to the accuracy or predictive value of traditional risk factors or epidemiological algorithms.[6–9] Family history of T2D has been shown to confer an odds ratio of about 5, while the odds ratio for genetic markers, even in combination, is less than 2.[10,11] This leads to uncertainty as to the incremental value and clinical validity of current genomic risk markers for T2D. Despite these concerns, there is still the potential for even weakly predictive genetic markers in a complex disease such as T2D to have clinical utility (i.e., guide prevention strategies) and/or personal utility (i.e., add to the risk discussion between patients and providers, motivate healthy behaviors or encourage adherence to preventive therapies).[12,13]
Personal utility is a particularly important aspect of complex conditions such as T2D, where patient behavior can dramatically alter clinical course, and the novelty of genetic risk results may provide a stronger motivational force than traditional risk results.[12,14] The Common Sense Model of self-regulation of health and illness, adapted by Marteau and Weinman,[15,16] provides a theoretical framework for explaining personal utility by hypothesizing that perceived risk increases motivation to engage in preventive health behaviors through self-regulation, a dynamic process that leads to efforts to reduce the discrepancy between one's current status (e.g., presence of health threat) and desired status (e.g., reduction in threat of disease).
A key issue in assessing the clinical and personal utility of genetic risk testing for a chronic disease such as T2D is defining the expected outcomes that are measurable and would define 'utility.' These may include health behaviors or lifestyle change, particularly diet and physical activity, given that obesity is a well-known risk factor for the development of T2D. Medication choices and adherence may also be considered relevant outcomes for reducing risk. Clinical measures, based on T2D risk calculations and validated risk factors, include fasting glucose or HbA1c to define T2D risk status. Weight or BMI, lipid levels, waist circumference and insulin resistance (homeostatic model assessment – insulin resistance) are also potential relevant outcomes. In reviewing the literature, we identified relevant studies that addressed two different areas of exploration in the clinical utility of genetic risk testing for T2D: hypothetical clinical utility and evaluation of actual outcomes related to genetic risk testing for T2D. This is a young area of the science in genetic testing; therefore, we included the hypothetical utility research, even though it is often weakly predictive of actual outcomes.
Personalized Medicine. 2013;10(6):549-563. © 2013 Future Medicine Ltd.