Pharmacokinetics and Glucodynamics of Ultra Rapid Lispro (URLi) versus Humalog^® (Lispro) in Younger Adults and Elderly Patients with Type 1 Diabetes Mellitus: A Randomised Controlled Trial

This was a phase I, two-period, randomised, double-blind, two-centre (Profil, Neuss and Mainz, Germany), crossover glucose clamp study in younger adult patients and elderly patients with T1DM. The study was conducted in accordance with the Declaration of Helsinki, the International Conference on Harmonisation Guideline for Good Clinical Practice, and applicable laws and regulations. The protocol was approved by an independent ethics committee, and all patients provided written informed consent. The study is registered at www.​clinicaltrials.​gov (NCT03166124).

Eligible patients were male or female, aged 18–45 years (younger adults) or ≥65 years (elderly), with a body mass index of 18.5–30.0 kg/m², diagnosed with T1DM for ≥1 year, and receiving insulin as multiple daily injections or a continuous subcutaneous insulin infusion (CSII). Patients were to have had fasting C-peptide ≤0.30 nmol/L, haemoglobin A1c < 9.0%, and no episodes of severe hypoglycaemia within the past 6 months. Patients were excluded if they were receiving daily insulin >1.5 units/kg/body weight or if their insulin regimen had changed in the 3 months before screening. Other main exclusion criteria were significant lipohypertrophy in the target abdominal injection area, proliferative retinopathy or maculopathy, severe neuropathy, or a history of renal impairment demonstrated by a glomerular filtration rate < 60 mL/min/1.73 m² (estimated according to the Chronic Kidney Disease Epidemiology Collaboration creatinine equation) or serum creatinine ≥ 126 μmol/L (1.43 mg/dL) for male patients or ≥ 111 μmol/L (1.26 mg/dL) for female patients.

Patients were randomised to receive a single 15-unit subcutaneous dose of study drug (URLi or Humalog U100 formulations [Eli Lilly, Indianapolis, IN, USA]) in the first dosing period and the alternate study drug in the second dosing period (Fig. 1). Before each dosing period, patients discontinued their basal insulin for a predefined washout period (≥72 h for insulin degludec or insulin glargine U300, ≥ 48 h for insulin detemir or glargine, ≥ 24 h for neutral protamine Hagedorn or other intermediate-acting insulins, and ≥ 6 h for any bolus injection of short-acting insulin via CSII). Patients receiving CSII therapy were to switch to insulin glulisine (Apidra^®; Sanofi, Paris, France) ≥ 8 h before dosing and to discontinue basal insulin delivery ≥ 3 h before dosing. An outpatient period of 3–15 days occurred between dosing periods. Follow-up was conducted ≥ 14 days after the second dosing period.

At each dosing visit, subjects underwent a 10 h automated euglycaemic clamp procedure using the ClampArt^® device (Profil, Neuss, Germany) under fasted conditions. Prior to dosing, a target blood glucose of 100 mg/dL (5.5 mmol/L) ± 10% was achieved using a variable intravenous infusion over 1–6 h of either 20% dextrose solution or insulin glulisine (Apidra; Sanofi). Blood glucose was maintained at 100 mg/dL ± 10% (5.5 ± 0.55 mmol/L) without any glucose infusion for the last 30 min before dosing; insulin glulisine infusion was stopped at least 10 min before dosing. If stable blood glucose was not achieved, the run-in period was prolonged and dosing was postponed. Baseline was the mean of blood glucose concentrations at 6, 4, and 2 min before dosing, and the onset of insulin action was defined as when blood glucose dropped by 5 mg/dL (0.3 mmol/L) from baseline. After the onset of insulin action was reached, a variable intravenous glucose infusion was initiated to keep blood glucose at 100 mg/dL (5.5 mmol/L). The glucose infusion rate (GIR) needed to keep blood glucose at the target level was recorded every minute throughout the glucose clamp. Manual blood samples were collected and measured (SuperGL glucose analyser, Dr. Müller; Hitado, Möhnesee, Germany) for blood glucose at least every 30 min during the clamp procedure to validate clamp glucose sensor measurements. The clamp procedure was continued for 10 h postdose or until blood glucose concentrations increased to >200 mg/dL (11.1 mmol/L) without any glucose being administered for at least 30 min, whichever was earlier.

Blood samples for insulin lispro pharmacokinetic analysis were taken every 5 min during the first hour postdose, at 70, 90, 120, 150, and 180 min, then every hour thereafter up to 10 h. A validated sandwich enzyme-linked immunosorbent assay, specific to insulin lispro without cross-reactivity to endogenous insulin, was used to quantify free insulin lispro serum concentrations. The lower limit of quantification (LLOQ) was 8.6 pmol/L, and inter-assay accuracy (% relative error) and inter-assay precision (% relative standard deviation) were ≤16%. Quantification of insulin lispro was not affected by the presence of lipaemic serum, haemolysed serum, treprostinil (1 ng/mL), human insulin (10 ng/mL), insulin aspart (600 pmol/L), insulin glargine (150 pmol/L), or insulin glulisine (600 pmol/L).

Treprostinil sampling times were 15, 30, 60, and 120 min postdose. Plasma treprostinil was measured by liquid chromatography-mass spectrometry/mass spectrometry assay. The LLOQ was 0.0100 ng/mL, inter-assay precision and accuracy were ≤10%, and the assay was not affected by the presence of insulin lispro, lipaemic serum, or haemolysed serum.

Free serum insulin lispro pharmacokinetic parameters were calculated by non-compartmental methods using Phoenix^® version 7.0 and S-PLUS^® version 8.2. Early exposure endpoints included time to early half-maximal drug concentration (early 50% t_max); area under the concentration-time curve (AUC) from time 0 to 15 min (AUC_0–15min), to 30 min (AUC_0–30min), and to 1 h (AUC_0–1h); and onset of appearance, defined as the time that serum insulin lispro reached the LLOQ. Determination of onset of appearance used a linear interpolation between the time of dosing (zero serum insulin lispro concentration) and the time of the first quantifiable serum insulin lispro concentration. Late exposure endpoints included time to late half-maximal drug concentration (late 50% t_max); AUC from time 2 to 10 h (AUC_2–10h) and from 3 to 10 h (AUC_3–10h); and duration of exposure, defined as the time from dosing until serum insulin lispro reached the LLOQ. Overall exposure endpoints were AUC from time 0 to infinity (AUC_0–∞) and maximum observed drug concentration (C_max).

Glucodynamic assessments were determined from the glucose clamp procedure, where the GIR over time was used as a measure of insulin effect. A locally weighted scatterplot smoothing function with a span of 0.1 was applied to all individual GIR-versus-time profiles in each treatment group and/or period. Fitted data for each patient were used to calculate glucodynamic parameters, except time to onset of insulin action (T_onset), which was based on raw GIR data. Glucodynamic analyses were conducted using Phoenix version 6.4 or higher and S-PLUS version 8.2. Early insulin action endpoints were T_onset, defined as the time when blood glucose drops by 5 mg/dL (0.3 mmol/L) from baseline; time to half-maximal GIR before time to maximum GIR (early 50% tR_max); and total amount of glucose infused over the first 30 min (G_{tot,0–30min}), first hour (G_tot,0–1h), and first 2 h (G_tot,0–2h). Late insulin action endpoints were time to half-maximal GIR after time to maximum GIR (late 50% tR_max); total amount of glucose infused from 3 h to end of the clamp (G_tot,3h–End) and from 4 h to end of the clamp (G_tot,4h–End); and duration of insulin action, defined as the time from T_onset to end of the clamp. Total insulin action endpoints were maximum GIR (R_max) and total amount of glucose infused over the duration of the clamp (G_tot).

Safety assessments included adverse events, injection site assessments (immediately postdose, 1 and 4 h postdose, and at end of the clamp), clinical laboratory assessments (predose in period 1 and as deemed necessary), vital signs, and electrocardiograms (predose and at end of the clamp).

At least 34 completers in each age group provided approximately 95% power to demonstrate a 40% increase in insulin lispro AUC_0–30min between URLi and Humalog within each age group. Pharmacokinetic and glucodynamic analyses included all patients who completed at least one euglycaemic clamp procedure. Log-transformed pharmacokinetic and glucodynamic parameters were evaluated using linear mixed models with treatment, sequence, and period as fixed effects and patient as random effect; treatment ratios of geometric least squares (LS) means and corresponding 95% confidence intervals (CIs) were estimated from the model. Pharmacokinetic and glucodynamic time parameters and glucodynamic parameters with at least one patient with a value of 0 were analysed using the same model without log transformation; treatment differences in LS means and 95% CIs were calculated from the model, and treatment ratios of LS means and 95% CIs were estimated using Fieller’s theorem [17]. Statistical analyses were conducted using SAS^® version 9.4 (SAS Institute, Cary, NC, USA) at a 5% significance level.

Prespecified exploratory analyses comparing the treatment effect (URLi vs. Humalog) between younger adult patients and elderly patients were conducted for pharmacokinetic and glucodynamic endpoints. The model was similar to that for each patient population, with the addition of age group and interaction of age group-by-treatment as fixed effects.

Of 93 patients screened, 80 patients with T1DM were enrolled (electronic supplementary Fig. S1). Of these, 41 were younger adult patients (aged 22–45 years) and 39 were elderly patients (aged 65–77 years). All 41 of the younger adult patients received at least one dose of study drug and 38 completed the study. Three younger adults withdrew after the first dosing visit (URLi: two; Humalog: one), one owing to a serious adverse event (SAE) of hypoglycaemia (occurring 6 days after URLi administration and considered unrelated to study treatment) and two owing to consent withdrawal. Of the 39 elderly patients, 38 received at least one dose of study drug and 37 completed the study. One elderly patient was withdrawn before study treatment, and one was withdrawn after the first dosing visit owing to an adverse event of hypotension following Humalog dosing that was considered by the investigator to be related to study procedure but not study drug. Demographics, baseline characteristics, and previous insulin therapy are shown in Table 1.

Both URLi and Humalog were well tolerated and no unexpected safety signals were identified. The incidence of treatment-emergent adverse events (TEAEs) was 15% in younger adult patients following each treatment, whereas in elderly patients the incidence of TEAEs was higher following Humalog dosing (31.6%) than URLi dosing (18.9%) (Table 2). Common TEAEs were vomiting, nausea, and headache in both age groups, and nasopharyngitis in elderly patients (Table 2). All TEAEs were mild or moderate in both age groups. One younger adult experienced an SAE of hypoglycaemia 6 days after receiving URLi, which was considered unrelated to study drug and resolved after 5 min with intravenous glucose. Of 154 injections administered, local reactions were identified in two patients after URLi injection (mild erythema and injection site urticaria 4 h postdose, and erythema and oedema at end of the clamp in one younger adult; moderate itching immediately postdose in one elderly patient) and in one patient following Humalog injection (mild pain immediately postdose in an elderly patient). No clinically relevant changes in vital signs, electrocardiograms, or laboratory values were observed.