Determining if implementation of the haptoglobin phenotype biomarker should be tested in type 2 diabetes treatment planning to prevent cardiovascular disease: a multi-methods study of risk prediction, etiological, and qualitative analyses

Abstract

Recent research indicated that people with type 2 diabetes (T2D) and the haptoglobin (Hp) 2-2 phenotype had reduced coronary artery disease (CAD) risk from intensive glycemic control. In contrast, those without the Hp2-2 phenotype (non-Hp phenotype) did not experience this cardiovascular benefit and instead faced increased mortality risk. These findings suggest the potential for using the Hp phenotype test to personalize glycemic targets in people with T2D. This dissertation aimed to establish multiple lines of evidence through a multi-methods approach to evaluate whether implementation of the Hp phenotype biomarker test should be evaluated in T2D treatment planning to prevent cardiovascular disease (CVD), particularly CAD. First, prediction models using data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial of intensive glycemic control were developed and internally validated to predict the likelihood of benefit (CAD and CVD prevention) and harm (total mortality (TM) and severe hypoglycemia) from intensive glycemic control. Hp phenotype was included in the risk prediction models for CAD and TM, indicating that Hp phenotype may contribute to score predictability. Second, an etiological analysis investigated whether the association between longitudinally achieved time-varying glycated hemoglobin (HbA1c) levels and CAD risk varied by the major Hp phenotype groups among participants in the Action for Health in Diabetes (Look AHEAD) trial of intensive lifestyle intervention. Lower HbA1c (<6.5%) appeared to offer protection against CAD in individuals with the non-Hp2-2 phenotype, particularly within certain subgroups. Higher HbA1c (≥8.0%) was associated with greater CAD risk in individuals with the Hp2-2 phenotype and prior CVD. Third, qualitative inquiry provided insight into how people with T2D value personalized blood glucose control and how they perceive the concept of Hp phenotype-guided personalization of glycemic control. Participants felt the test was acceptable, helpful, and not inconvenient, and were generally supportive of it as a concept. However, they also identified some potential barriers to implementation. The findings in this dissertation collectively indicate that implementation of the Hp phenotype test should be tested in further research.