Safety and efficacy of GP40071 compared with originator insulin aspart (NovoRapid^® Penfill^®) in Type 1 diabetes mellitus

This was a Phase III, multicenter, openlabel, randomized, activecontrolled, parallelgroup, noninferiority 26-week study conducted in 19 study sites. Eligible study participants with T1D (n = 264) in accordance with the inclusion/exclusion criteria were randomized 1:1 to GP-Asp (n = 132) or NN-Asp (n = 132) treatment. The randomization was concealed using interactive web response system, allowing to manage the randomization.

The study period included screening procedure (up to 4 weeks), insulin dose titration (4 weeks) and treatment period with stable doses (22 weeks). The study participants administered insulin with pensinjectors. During insulin dose titration period, insulin doses of both bolus insulin aspart and basal insulin could be changed as much as needed to achieve optimal glycemic control. Insulin dose titration period took 4 weeks during which the participants measured capillary blood glucose level at least four-times a day or even more often due to their personal requirements each time they were supposed to have a hypoglycemic episode. Insulin dose adjustment was calculated depending on the blood glucose level during the day and optimal glycemic control. Optimal glycemic control was based on the individual HbA1c level according to standards of specialized diabetes care [1]. Measured values of glycemia, daily insulin doses and glycemia measurement frequency were registered by participants in the special diary. Basal insulin doses during the treatment period should have been stable. The study participants adjusted bolus insulin dose on their own according to metabolic requirements during all study periods. The investigators taught the study participants how to calculate bolus insulin dose to achieve target glycemic control.

To provide participants’ safety during the treatment period with stable doses (in case of hypoglycemic episodes associated with dietary or physical activity changes and presence of concominant diseases) basal insulin dose also could be adjusted within 15%, however, it was not recommended generally. To maintain stable blood glucose levels, the participants should have measured capillary blood glucose level at least four-times a day or even more often due to their personal requirements each time they were supposed to have a hypoglycemic episode.

The batch numbers of the used Novorapid Penfills are HJ30068 and HJ30086. Products of Novorapid were supplied by the official Novo Nordisk supplier OOO ‘Torgovyj Dom Zdorovye natsii’. The delivery of the drug to the study centers was provided by the transporting companies in compliance with required temperature conditions.

The study was registered under ClinicalTrials.gov number NCT04079413.

Eligible participants were 18–65 years old, with T1D diagnosed for at least 12 months before screening, with history of basal-bolus insulin therapy in stable doses for at least 30 days before screening, with HbA1c level 7.1–12.0% and BMI 18.5–35.0 kg/m² at screening.

The trial included only T1D patients since they are considered to be more sensitive developing insulin antibodies in response to exogenous insulin [4,5].

The key exclusion criteria were as follows:

Patients with several diabetes complications in advanced stage (proliferative diabetic retinopathy, severe peripheral diabetic neuropathy or autonomic neuropathy, diabetic nephropathy with estimated glomerular filtration rate <45 ml/min/1.73 m², diabetic foot syndrome) found at screening or earlier were also not enrolled to the clinical trial.

The full list of the exclusion criteria is provided at ClinicalTrials.gov.

Immunogenicity outcome measures were the primary safety end points in this study.

Anti-insulin antibodies (AIA) baseline concentration was assessed before the beginning of treatment at screening. Patients with AIA baseline concentration exceeding 10 IU/ml was estimated as AIA positive patients.

The primary end point was frequency of immune response at week 26. The immune response was assessed by the following criteria:

Secondary immunogenicity end points included:

AIA concentration was assessed using the validated enzyme immunoassay method in the central laboratory.

Clinically significant immune response included AIA concentration increase for more than 30% from baseline combined with the following criteria:

To detect neutralizing AIA, we used the method with iLiteTM insulin assay ready cells introducing the firefly luciferase reporter gene to their genome under the control of an insulin-dependent promoter [7]. Neutralizing activity was established via binding of insulin alfa-chain and CD220 receptor in samples.

In the present clinical trial, efficacy end points were considered as secondary in accordance with the regulatory guidelines [4,5] and included:

The study investigators established individual glycemic target for each participant in accordance with the national guidelines for diabetes treatment [1] at the randomization visit. They did not change this target during the study periods.

HbA1c (in %) was measured in blood samples at screening and weeks 12 and 26 in the central laboratory (automated HPLC method) (ZAO ‘Unimed Laboratories’, Moscow). The laboratory has been assessed and certified as meeting the requirements of ISO 9001:2015, providing validated methodology of testing.

FPG level was measured in venous blood samples at screening, week 12 and 26. The analysis of FPG level was also performed at the central laboratory (enzymatic UV test [hexokinase method] for the quantitative determination of glucose in human serum, plasma, urine, hemolysate and cerebrospinal fluid on Beckman Coulter analyzers).

SPGP was performed by the participants using glucometers Accu-Chek Active provided by the sponsor. The study subjects performed SPGP for 3 days, whereas two out of three measurement days should have been consecutive. SPGP was performed at the week preceding 4th, 6th and 9th visit (weeks 4, 12 and 26, respectively). They measured capillary blood glucose in fasting condition, 2 h after breakfast, before lunch, 2 h after lunch, before dinner, 2 h after dinner and at 3 am.

DTSQs and DTSQc questionnaires measured patients' satisfaction with the current diabetes treatment. Both questionnaires consisted of eight questions estimating absolute and relevant satisfaction values for DTSQs and DTSQc, respectively. Patients completed DTSQs at randomization and week 26, while DTSQc was completed only at week 26.

In the study, incidence and severity of adverse events (AEs) including adverse events of special interest such as hypoglycemic episodes, injection site reactions and hypersensitivity reactions, were assessed. In addition, the changes in vital signs, ECG and laboratory results compared with baseline were also assessed. All investigators reported AEs either observed at the study visits or spontaneously reported by study subjects. AEs were recorded in source documents and electronic case report forms. AE details were collected for all study participants from signing the informed consent form to their last visit.

Each AE was assessed by the study investigators according to the following criteria: seriousness, severity, relationship with the investigational or comparator insulin (WHO-UMC causality assessment). Therefore, any symptom, complaint or illness which occurrence might have been associated with GP-Asp or NN-Asp, was attributed to AE.

A serious AE (SAE) was any untoward medical occurrence at any drug dose that, in the opinion either of the investigator or sponsor, resulted in death, was lifethreatening, required hospitalization, led to persistent/significant disability/incapacity or was congenital anomaly/birth defect. If a SAE occurred, it was reported to the sponsor within 24 h. Every SAE was reviewed by the Sponsor’s central Pharmacovigilance and Medical teams based on their clinical judgment. All non-SAEs were periodically analyzed by the Sponsor's Medical team.

Hypoglycemic episodes, ketoacidosis, allergic and injection site reactions were evaluated separately from other AEs since they were considered as AEs of special interest (according to the study protocol). Hypoglycemic episodes were recorded by the participants in their personal self-care glycemic diaries. Only episodes with capillary blood glucose level less than 3.9 mmol/l were considered hypoglycemic episodes. In addition, symptoms of hypoglycemia, need in assistance from other people, loss of consciousness, starting time of the hypoglycemia event and nocturnal hypoglycemia were also considered to describe hypoglycemia episodes. Severe hypoglycemia was reported if a person needed any assistance from other people [1]. Local injection site reactions and hypersensitivity reactions were assessed during visits of each patient to the study site and were also recorded by the study participants in their diaries. The data were entered into electronic case report forms.

All data analyses were made using Windows software RStudio version 3.5.0 in accordance with the approved statistical analysis plan. The statistical analyses were performed in the Full Analysis Set population. Full Analysis Set population is the set of subjects that is as close as possible to the ideal implied by the Intention-to-treat principle. It is derived from the set of all randomized subjects by minimal and justified elimination of subjects (ICH E9).

Quantitative statistical analysis is described using:

ANCOVA was used to estimate changes in continuous variables from baseline. For each of these variables, ANCOVA included baseline variable value as a covariate and treatment group and study site as fixed factors. If the changes from baseline were not assessed, Student’s t-test or Mann–Whitney U-test were used to compare continuous variables depending on normality of distribution. In case of inequality of variances in groups (evaluated using Levene’s test), Welch’s unequal variances t-test was used instead of Student’s t-test.

Qualitative statistical analysis is described using:

Qualitative (categorical) variables were compared with X²-test (continuity correction was used if any units contained value of 6–10) or exact Fisher’s test (if any units contained value of 0–5). The change in dichotomous variables from baseline with McNemar test was also analyzed.

p-value < 0.05 was accepted as statistically significant.

Immunogenicity of insulin is considered to be relatively low, though there was no need to provide high study power in terms of frequency of immune response [4,5]. In the comparative studies of insulin immunogenicity, sample size is estimated based on HbA1c values in accordance with the regulatory guidelines [4,5]. We used α level 0.025, β level 0.2, HbA1c mean difference 0, SD 1.1 and δ 0.4% in accordance with regular practice for antihyperglycemic drugs [8]. Based on the estimated values, the total number of required subjects was estimated as 238 (119 participants per group). Taking into account a possible 10% early withdrawal rate, 264 patients were randomized.

The trial was conducted in accordance with the Guidelines on Good Clinical Practice and Ethical Principles for Medical Research Involving Human Subjects laid down in the Declaration of Helsinki [9]. All study participants provided the written informed consent before participation in the study. The trial was approved by the Ministry of Health of the Russian Federation (Clinical trial approval no. 217 dated 26 April 2019) and by the Ethics Council (abstract No. 188 dated 12 March 2019).

We screened 309 T1D patients and randomized 264 of them (132 to GP-Asp, 132 to NN-Asp groups). Three participants discontinued GP-Asp therapy (two subjects decided to withdraw from the study, the third one was lost to follow-up), and one participant discontinued NN-Asp therapy (death of the subject). Figure 1 provides the flowchart of the study participants. Baseline characteristics of the study subjects are presented in Table 1. All baseline characteristics were comparable between the groups except smoking status: there were more smoking patients in GP-Asp group compared with NN-Asp group.

At baseline there were 10 AIA positive (122 AIA negative) patients in GP-Asp group and 12 AIA positive (120 AIA negative) patients in NN-Asp group.

However, variety of immune response trends was registered: three participants in GP-Asp group and two participants in NN-Asp group became AIA negative at week 26 compared with week 12, while eight participants in GP-Asp group and five participants in NN-Asp group developed immune response only after week 26.

The frequency of immune response at week 26 did not differ between the groups (Table 2).

Mean baseline AIA concentration in GP-Asp group and NN-Asp group were similar (p = 0.545). Mean AIA concentration at week 12 was 6.06 ± 14.17 units/ml in GP-Asp group and 3.95 ± 4.60 units/ml in NN-Asp group. By week 26, mean AIA concentration was 6.53 ± 14.72 units/ml in GP-Asp group and 4.14 ± 4.94 units/ml in NN-Asp group. The changes in AIA levels by weeks 12 and 26 are shown in Table 2. These changes were not statistically significant in both groups.

Due to the evaluated wide SD for AIA change (2.5x the means), outliers were excluded from the statistical analysis. Figure 2 shows box and wafer plots with outliers at baseline, 12 and 26 weeks. Statistical analysis of AIA change also revealed no significant difference between groups, however the SD for means became rather smaller (Table 3).

During the study, no subject developed a clinically significant immune response. Developing AIAs did not affect treatment efficacy (p = 0.093).

Blood samples of all study participants were also tested for neutralizing AIA at screening and week 26. Seventeen of 132 patients in GP-Asp group and 15 of 132 in NN-Asp group developed neutralizing AIAs by week 26 (p = 1.000) and none of them had any clinical signs indicating the reduced efficacy of insulin treatment and the development of neutralizing AIA did not affect treatment efficacy (p = 0.895).

HbA1c level significantly decreased from baseline at week 12 for 0.55 ± 1.10% in GP-Asp group (p < 0.001) and for 0.64 ± 1.09% in NN-Asp group (p < 0.001). HbA1c level also significantly decreased from baseline at week 26 for 0.57 ± 1.21% in GP-Asp group (p < 0.001) and for 0.56 ± 1.10% in NN-Asp group (p < 0.001). The following changes did not differ between the groups either at week 12 (p = 0.542) or week 26 (p = 0.955). HbA1c changes are shown in Figure 3.

Intergroup mean difference of HbA1c level change (95% CI) at week 26 from baseline was 0.00 [-0.26, 0.25] %. The CI did not cross the established 0.4% noninferiority margin demonstrating that GP-Asp was noninferior to NN-Asp in terms of hypoglycemic efficacy.

Moreover, 26 patients (25%) in GP-Asp group and 19 (22%) patients in NN-Asp group could achieve HbA1c ≤7% at week 26 (p = 0.298). GP-Asp and NN-Asp groups also did not differ in individual glycemic target achieved at week 26 (p = 0.602).

FPG level significantly decreased from baseline at week 12 for 1.67 ± 5.19 mmol/l in GP-Asp group (p = 0.002) and for 1.09 ± 5.22 mmol/l in NN-Asp group (p = 0.006). FPG level also decreased significantly from baseline at week 26 for 1.25 ± 5.32 mmol/l in GP-Asp group (p = 0.015) and for 1.15 ± 4.70 mmol/l in NN-Asp group (p = 0.001). These changes did not differ between the groups either at week 12 (p = 0.232) or week 26 (p = 0.720). FPG changes are shown in Figure 4.

SPGP changes did not differ between the groups at week 26 (p = 0.579; p = 0.327; p = 0.559; p = 0.633; p = 0.867; p = 0.309; p = 0.671 in fasting condition, 2 h after breakfast, before lunch, 2 h after lunch, before dinner, 2 h after dinner and 3 am, respectively). SPGP data are shown in Figure 5.

Total insulin dose was significantly increased in both groups at week 26 compared with week 4 for 1.25 ± 6.43 units/day in GP-Asp group (p = 0.025) and 1.45 ± 4.67 units/day in NN-Asp group (p < 0.001), but these changes did not differ between the groups (p = 0.667). Insulin dose per kg was slightly increased by week 26 in both groups (0.02 ± 0.09 units/kg/day in GP-Asp; p = 0.016 and 0.01 ± 0.07 units/kg/day in NN-Asp; p = 0.001), but these changes did not differ between the groups (p = 0.975).

In GP-Asp group, bodyweight of the patients did not change (-0.04 ± 1.77 kg; p = 0.772), however bodyweight was increased in NN-Asp group at week 26 compared with baseline values for 0.38 ± 2.13 kg (p = 0.004). These changes did not differ between the groups (p = 0.113).

Treatment satisfaction was significantly increased in both groups at week 26. The increase in DTSQs total score was 4.23 ± 6.95 points in GP-Asp group (p < 0.001) and 6.06 ± 7.32 points in NN-Asp group (p < 0.001) and was similar between the groups (p = 0.801). DTSQc scores did not differ between the groups in any question.

The incidence of AEs, SAEs, discontinuations due to AEs and adverse drug reactions reported in GP-Asp group were similar to those in NN-Asp group (Table 4). Two of 56 AEs in GP-Asp group and seven of 88 AEs in NN-Asp groups were classified as adverse drug reactions, and their incidence did not differ between the groups (p = 0.480).

Most of these AEs were mild. There were ten SAEs: two in GP-Asp group and 8 in NN-Asp group. Only one SAE was considered by the investigator as possibly related to the investigational drug. It was a serious hypoglycemic episode in NN-Asp group, the investigator assessed the episode as a possibly life-threatening, however, the subject recovered without medical care and continued participation in the study. One death was reported in NN-group due to heart failure. The investigator assessed the death as not related to NN-Asp. One patient in GP-Asp group had an injection site reaction, and patients in NN-Asp group did not develop any injection site reactions. No hypersensitivity reactions were observed in either of the groups. There were no differences in any laboratory safety values or vital signs between the groups (data not shown).

Table 5 contains the summary of hypoglycemic episodes. The number of patients experiencing hypoglycemic episodes did not differ between the groups (p = 0.497). GP-Asp group had less hypoglycemic episodes for 1.4 events/patient-year (p = 0.052). During the study, there were two severe hypoglycemic episodes in GP-Asp group and one episode in NN-Asp group. Most patients in both groups experienced a mild hypoglycemia. Most hypoglycemic episodes were symptomatic in both groups, the patients with symptomatic or asymptomatic episodes did not differ between the groups (p = 0.840). Most hypoglycemic episodes occurred at daytime (06:00–23:59).