How to improve the quality of euglycemic glucose clamp tests in long-acting insulin studies.

Abstract

Background: The euglycemic clamp test stands as the best method for assessing the pharmacokinetic and pharmacodynamic properties of long-acting insulin. However, despite its widespread use, there remains a notable absence of an established gold standard for evaluating the test's quality. Existing recommendations from regulatory agencies lack specific threshold values, particularly concerning long-acting insulin. This study aimed to determine the evaluation criteria for assessing the quality of the long-acting insulin euglycemic glucose clamp test and to improve the overall quality of this testing method.

Methods: Fifty-three healthy volunteers were administered a single dose of insulin degludec (0.4 IU/kg) and underwent a 24-h euglycemic clamp test. Blood samples were collected to evaluate the pharmacokinetics and pharmacodynamics of insulin degludec. Volunteers were separated into group A (coefficient of variation in blood glucose [CVBG] ≤ 3.5%) and group B (CVBG > 3.5%). The quality difference of the clamp test between the groups was assessed using various quality control indices. Volunteers were also categorized into group C (C-peptide reduction rate < 50%) and group D (C-peptide reduction rate ≥ 50%). The clamp test quality, pharmacokinetics, and pharmacodynamics of groups C and D were compared.

Results: According to CVBG, group A had a mean CVBG of 2.95%, group B had a mean CVBG of 4.15%, and group A had a significantly lower CVBG than group B (p < 0.001). CVBG was positively correlated with other quality control indicators, such as the percentage of glucose excursion from the target range (GEFTR), duration of GEFTR, and area under the curve (AUC) of GEFTR. According to the reduction of C-peptide levels: group D had significantly higher C-peptide reduction than group C (p < 0.001). Groups C and D had CVBG < 3.5%. The quality of groups C and groups D was evaluated by the quality control indicators of the clamp test. Only the AUC of GEFTR was statistically different between Groups C and D (p = 0.043, < 0.05), and there was no statistical difference in other indicators between the two groups.

Conclusions: CVBG could be used as a standard for evaluating the quality of long-acting insulin euglycemic glucose clamp test, and the test quality was superior with a CVBG ≤ 3.5%. A C-peptide reduction ratio ≥ 50% indicated sufficient endogenous insulin inhibition; however, when the glucose fluctuation is small (CVBG is maintained at a low level) during the clamp test, even if the clamp test quality is slightly different, it is not sufficient to interfere with endogenous insulin secretion.