Modeling and simulations to confirm a controlled hypoglycemic stress test in healthy subjects is not associated with clinically significant QT prolongations.

Abstract

A modified insulin tolerance test (ITT) can be used to simulate a physiological stress state through the induction of controlled hypoglycemia in healthy volunteers. This allows for evaluation of hypothalamic-pituitary-adrenocortical axis response to stress via a surge in cortical release. However, a consequence of severe, prolonged hypoglycemia is QT interval prolongation. The aim of this analysis was to confirm that blood glucose lowering to 60 mg/dL (previously identified as adequate for inducing stress) has low risk of inducing clinically significant QT prolongation. Continuous ECG monitoring was conducted as a planned sub study of an open-label, 2-period study involving 18 healthy male subjects. The QTcF response to hypoglycemia was measured over 2 identical periods, ~ 7 days apart. An indirect- response model adequately described the pharmacological relationship between blood glucose and QTcF intervals over the time-course of the ITT. The model correctly identified the steep glucose-QT relationship as an on-off response with a large Hill coefficient of 59 and the threshold glucose, EC50, as ~ 57 mg/dL with narrow between-subject variability of 10%. Simulated QTcF profiles over the course of an ITT did not demonstrate any QTcF interval changes of clinical concern, defined as QTcF observation > 500 ms, if hypoglycemia did not reach below 60 mg/dL. The statistical prediction that the chance of a mean QTcF observation > 500 ms was < 0.0001. Results support that an ITT maintained at or above 60 mg/dL is unlikely to cause QT prolongation in healthy volunteers and does not warrant continuous ECG monitoring in this group of subjects.