With Caroline A. Brorsson, PhD, and Michael Gonzalez-Campoy, MD, PhD
Three major subgroups of newly diagnosed patients with type 2 diabetes (T2D) experienced different rates of disease progression over 18 months,1 according to data presented at the 53rd annual meeting of the European Association for the Study of Diabetes in Lisbon, Portugal. The research was part of the Diabetes Research on Patient Stratification project (DIRECT) within the European Union Framework 7 Innovative Medicines Initiative.
Patients with type 2 diabetes are likely to present with varying degrees of insulin resistance and beta cell failure.1 Understanding the heterogeneity of a T2D presentation may lead to more effective treatment strategies for these patients. An underlying difference in pathophysiology may be indicative of a patient’s responsiveness to a prescribed treatment and have an anticipated effect on disease progression.1
Caroline Brorsson, PhD, a postdoctoral researcher at the Technical University of Denmark and colleagues used the detailed clinical phenotyping from the Diabetes Remission Clinical Trial (DIRECT) to identify and cluster subgroups of patients who were newly diagnosed with T2D.1,2 In the DIRECT study, detailed metabolic data were collected on patients newly diagnosed with either prediabetes or type 2 diabetes.2
“Using a very detailed clinical phenotyping methodology, we wanted to systematically capture disease heterogeneity in newly diagnosed diabetes patients using a data-driven approach to be able to investigate the effect of different patient subgroups on disease progression during the follow-up period,” Dr. Brorsson told EndocrineWeb.
The researchers enrolled 757 individuals in the study, and characterized the diabetes phenotype at the time of an initial diagnosis, then followed these patients for 18 months. The goal was to assess whether any difference in disease progression was evident within the subgroups of patients. In addition, the researchers evaluated the stability and movement of patients between subgroups based on changes in disease characteristics or factors influencing the progression of diabetes.1
Twenty variables were included in the clustering analysis, which was an unsupervised agglomerative hierarchical approach.1 The distance matrix calculations were done using Gower’s similarity method, clustering using Ward’s method, the cluster package in R, and variables with skewed distributions were log-transformed.1
The researchers identified three major patient subgroups that differed significantly in their baseline disease characteristics. 1The three subgroups were:
The main traits influencing clustering were differences in insulin sensitivity and insulin secretion. Interestingly, the insulin resistant subgroup was treated with more metformin at 18 months, had higher liver fat percentages, higher diastolic blood pressure, higher total GLP-1 and glucagon at baseline, had lower physical activity at baseline, and showed a faster rate of progression on hemoglobin A1c slopes over the 18 months. that these patients were followed.1
The co-primary endpoints were patients reporting freedom from the most bothersome migraine symptom and freedom from pain at 2 hours post-dose. The researchers found that ADAM zolmitriptan 3.8 mg resulted in significantly more patients being pain-free at 2 hours post-dose than placebo (41.5% vs. 14.3%; p = 0.0001) and MBS-free at 2 hours (68.3% vs. 42.9%; p = 0.0009).1 ADAM zolmitriptan 1.9 mg also had a significantly higher proportion of patients that were pain-free than placebo (27.7% vs. 14.3%; p = 0.0351), but the MBS results showed no significant difference (53% vs. 42.9%; p = 0.1649).1
“We demonstrated that subgroup membership was associated with how fast the individuals progressed in their diabetes over the 18 month follow-up period,” Dr. Brorsson said, “The individuals who were part of the Insulin Resistant subgroup at baseline showed the fastest progression of diabetes based on their HbA1c trajectories over time.”
“The most important finding was that insulin resistance has the worst prognosis for people newly diagnosed with type 2 diabetes. Therefore, aggressive diagnosis and treatment of insulin resistance and obesity must be achieved to help with diabetes management,” said Michael Gonzalez-Campoy, MD, PhD, medical director and chief executive officer of the Minnesota Center for Obesity, Metabolism, and Endocrinology, who was not associated with the study.
“This work is a novel approach to prognosis in people with T2D. The caveat in this work is that it used a narrow population, one used for a specific trial. This observation needs validation over time,” Dr. Gonzalez-Campoy told EndocrineWeb.
Although the study was done with a tightly controlled population, the implications may be widespread, according to Dr. Brorsson.
“It may be too early to change the guidelines for clinicians who have patients newly diagnosed with type 2 diabetes today; however, I hope that we have highlighted that type 2 diabetes consists of patients with a heterogeneous presentation of their diabetic phenotype at onset, and that belonging to different patient subgroups could have implications for their rate of disease progression,” Dr. Bronrsson said.
The study was undertaken to test the feasibility of whether the heterogeneous traits patients with a new diagnosis of type 2 diabetes are indicative of disease progression, which was born out for the patients who participated in the DIRECT study. Importantly, the findings offer clinicians a keen insight as to possible variations in the progression of diabetes for patients and the implications for clinical management.
“The ultimate goal of this study would be if treatment strategies were be targeted towards specific patient subgroups in a ‘stratified medicine’ approach,” said Dr. Brorsson, “It may be that the individuals in the insulin resistant subgroup would benefit from a more intensive treatment strategy to prevent their diabetes from progressing, whereas for the individuals in the Insulin Sensitive subgroup, lifestyle interventions such as changing diet and exercise may be sufficient to reach individual treatment goals.”
To gain a clearer understanding of the impact a particular treatment protocol may have will require further studies that focus on specifically defined interventions with each subgroup of patients.
1. Brorsson CA, Pederson KH, Gudmundsdottir MA, et al. Clustering on baseline clinical variables identifies subgroups of type 2 diabetes patients with different rate of progression over 18 months: a DIRECT study. Presented at: The 53rd annual meeting of the European Association for the Study of Diabetes; September 11-15, 2017; Lisbon, Portugal.
2. Koivula RW, Heggie A, Barnett A, et. al. Discovery of biomarkers for glycaemic deterioration before and after the onset of type 2 diabetes: rationale and design of the epidemiological studies within the IMI DIRECT Consortium. Diabetologia. 2014;57(6):1132-42.