Preserved pharmacokinetics and pharmacodynamics of insulin degludec and liraglutide when administered as insulin degludec/liraglutide in a Chinese population

Abstract We report the findings of a single‐dose, randomized, three‐period cross‐over, clinical trial in healthy Chinese individuals (n = 24) comparing the pharmacokinetics of insulin degludec/liraglutide (IDegLira) with its individual components. Furthermore, we report a population pharmacokinetic analysis of a 26‐week, phase III, treat‐to‐target, randomized trial of 720 Chinese individuals with type 2 diabetes. Participants were randomized to IDegLira, degludec or liraglutide, all once daily with metformin. The pharmacokinetic profiles of IDegLira were similar to its individual components. Dose proportionality was indicated for both IDegLira components. Although there were no relevant covariate effects on degludec exposure, liraglutide exposure was inversely correlated with bodyweight. In conclusion, for the Chinese population, the pharmacokinetics of the fixed‐ratio combination IDegLira is similar to that of its individual components.

A pharmacokinetic (PK) trial in a healthy European population, and a population PK evaluation of individuals with type 2 diabetes, showed general preservation of PK properties for degludec and liraglutide with IDegLira 11 . There were no relevant deviations from dose proportionality for either IDegLira component, and the glycemic response to IDegLira was larger than either component alone 11 .
The present analyses assessed whether the PK and exposureresponse of degludec and liraglutide are similar when administered as IDegLira or separately in a Chinese population.

Ethical consideration
All procedures were in accordance with the ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration 12 , and the International Conference on Harmonization Good Clinical Practice 13 .

Single-dose pharmacokinetics
The China PK trial (NCT03292185) was a randomized, doubleblind, single-dose trial investigating the pharmacokinetics of degludec and liraglutide when administered as IDegLira, or as separate administrations, in healthy Chinese individuals (Methods S1). Participants received the following single-dose treatments subcutaneously in randomized order on separate visits: IDegLira 17 dose steps (17 U degludec/0.6 mg liraglutide); degludec 17 U; and liraglutide 0.6 mg.
Bioanalysis of degludec and liraglutide was carried out using validated enzyme-linked immunosorbent assays. The primary end-point was exposure for each component of IDegLira, and for degludec and liraglutide given alone (Methods S1).
The population PK analysis assessed dose proportionality of degludec and liraglutide exposures after doses of IDegLira and investigated the effects of pre-specified covariates on exposure to both components during 26 weeks of treatment. The exposure-response analysis was used to show the contribution from each component of IDegLira to the overall glycemic response across the exposure range (Methods S1).

Single-dose pharmacokinetics
General PK patterns were similar, but with a maximum concentration (C max ) appearing slightly lower for degludec and liraglutide when administered as IDegLira compared with their administration alone (Figure 1a,b). The estimated relative bioavailability of degludec (treatment ratio of the area under the curve from 0 to last quantifiable observation [AUC Degludec, 0-tz ]) when administered as IDegLira versus degludec alone was 1.00 (90% confidence interval [CI] 0.96;1.04), meeting the equivalence criterion (90% CI 0.80;1.25) 15 (Table 1). The treatment ratio of AUC Liraglutide, 0-tz when administered as IDegLira versus liraglutide alone was 0.88 (90% CI 0.83;0.94), meeting the equivalence criterion 15 (Table 1). The corresponding analysis for the estimated treatment ratio of AUC 0-∞ was in accordance with the AUC 0-tz for degludec and for liraglutide. The lower limit of the 90% CI for C max , degludec and C max , liraglutide was below the lower limit for equivalence ( Table 1). The median time to C max for degludec and liraglutide was comparable when administered as IDegLira (degludec: 12.0 h; liraglutide: 12.0 h) or separately (degludec: 11.0 h; liraglutide: 11.0 h).

Population pharmacokinetic analysis
Dose-proportionality expressed as AUC 0-24 h at steady state versus dose is shown in Figure S1. The 90% CIs for both the degludec and liraglutide models with dose effects were within the 80-125% CI of the reference dose-proportional model for each component indicating dose proportionality.  For DUAL I China, the covariate analysis for degludec showed no relevant effects for any of the investigated covariates ( Figure 2). The covariate with the most influence on exposure was bodyweight, as degludec exposure decreased with increasing bodyweight, although this effect was within the equivalence range. There was no difference in degludec exposure when dosed as IDegLira or degludec alone (estimated mean ratio 1.09, 90% CI 1.04;1.14). The covariate analysis for liraglutide showed a lower exposure with increasing bodyweight. There was no difference in liraglutide exposure when dosed as IDegLira or liraglutide alone (estimated mean ratio 1.02, 90% CI 0.97; 1.07).

Exposure-response analysis
The exposure-response relationships were relatively flat due to dose titrations to specified fasting glucose levels for IDegLira and degludec, and due to the pharmacodynamic properties of liraglutide, acting only when glucose levels are elevated above normal ( Figure S2). The effect of IDegLira was larger than the effect of degludec or liraglutide alone at all exposure levels, confirming that both components contributed to the glycemic effects of IDegLira.

DISCUSSION
The single-dose PK analysis showed similar PK for degludec and liraglutide when administered as IDegLira or separately in Chinese participants. This was supported by the similar AUC 0-∞ of degludec and liraglutide obtained when administered as IDegLira or separately. Treatment ratios showed that C max , degludec was 10% lower with IDegLira than with degludec alone; and C max , liraglutide was 17% lower with IDegLira than with liraglutide alone. As IDegLira is intended to be titrated based on individuals' plasma glucose levels, these differences in C max are not expected to affect clinical use.
The population PK analysis showed dose proportionality across the IDegLira dose range for degludec and liraglutide, as previously shown for the individual components 16,17 , which is consistent with the global PK evaluation 11 . There were no relevant covariate effects on degludec exposure; bodyweight showed the most effect, but within the equivalence criterion 15 for the range tested (54.0-99.1 kg). However, liraglutide exposure was inversely correlated with bodyweight for the investigated range (54.0-99.5 kg). There was no difference in degludec or liraglutide exposure when dosed as IDegLira or individually, conforming with the China PK trial. The AUC ratio of IDegLira/degludec was similar in the DUAL I China and China PK trials, consistent with the global PK data 11 . However, the AUC ratio of IDegLira/liraglutide was not in the same direction in the two trials, although the equivalence criterion 15   In conclusion, the general PK pattern of degludec and liraglutide are similar, when administered as IDegLira, or separately, in both healthy Chinese individuals and in those with type 2 diabetes. Overall, the present findings in a Chinese population align with those previously published for a global population.

ACKNOWLEDGMENTS
The authors thank the trial investigators, research coordinators and participants. Medical writing support, under the direction of the authors, was provided by Sejal Varsani, MRes, and Anna Campbell, PhD, of Ashfield MedComms, an Ashfield Health company, and funded by Novo Nordisk. The China PK and DUAL I China trials were supported by Novo Nordisk. Sponsorship for these analyses and article processing charges were funded by Novo Nordisk.

DATA AVAILABILITY STATEMENT
The datasets generated and/or analyzed for this article are available from the corresponding author on reasonable request.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Table S1 | Baseline characteristics of participants in the final dataset for pharmacokinetic analysis of degludec and liraglutide.