Immunogenicity, Safety and Efficacy Comparison of Wockhardt’s Biosimilar Insulin Glargine—Glaritus® with Reference Product— Lantus®: Study Protocol & Early Data Trends ()
1. Introduction
The diabetic population of India increased from 26 million in 1990 to 65 million in 2016 with increase of age-standardised disability adjusted life years rate for diabetes by 39.6% over the same period [1] . There were 1.2 million deaths (14.2% of all adult deaths) reported attributed to complications of diabetes from India during the year 2013 [2] . Recently updated WHO guidelines suggest that human insulin treatment shall be introduced in patients with type 2 diabetes mellitus (T2DM), who do not achieve glycemic control with metformin and/or sulfonylurea [2] . The DiabCare 2011 Study reported that 35.2% patients of T2DM in India are being managed with insulin with or without oral antidiabetics [2] . The “treat-to-target” (TTT) clinical trials established that the addition of basal insulin to existing oral glucose-lowering therapy achieves good glycaemic control in T2DM patients [3] [4] [5] .
Unfortunately, many patients with T2DM are unable to achieve the American Diabetes Association guideline-recommended glycosylated hemoglobin (HbA1c) level of <7% [6] [7] . Stringent glycemic control is typically associated with an increased risk of hypoglycaemia [8] in patients with T2DM [9] [10] . Basal insulins provide constant insulin activity over 24 h with only a single injection thereby improving fasting glycemic control, without an increased risk of hypoglycemia. The published literature supports early and timely basal insulin initiation as it improves FPG control and beta-cell function when compared to prolonged continuation of Oral Antidiabetic Drugs (OADs) [11] . Further, in T2DM, insulin glargine provides comparable glycemic control to that with neutral protamine Hagedorn (NPH) insulin with fewer severe hypoglycemic events (odds ratio 0.65, 95% CI 0.49 to 0.88) [3] .
Lantus® was the first long acting basal insulin analogue, insulin glargine manufactured by recombinant DNA technology. It was approved by the US FDA and European EMA in 2000 for use in type 1 and type 2 diabetes mellitus [12] [13] . Glaritus® is a biosimilar insulin glargine manufactured by Wockhardt with an identical primary amino acid sequence as Lantus®. Glaritus® is registered and marketed in India since August 2007, in Nepal, Peru, and Vietnam since August 2010, and in Mynamar since November 2011 for the same indications as that of Lantus® [14] .
As a part of physico-chemical-biological comparability exercise, Glaritus® was found comparable to Lantus® in in-vitro pharmacodynamics studies (receptor binding, receptor autophosphorylation and metabolic assays like glucose uptake and inhibition of lipolysis), in-vivo pharmacodynamics studies (hypoglycaemic potency studies in rodents and dogs) and sub chronic toxicity studies (in rodents including toxicokinetics and immunogenicity) [15] . In terms of human pharmacokinetics and pharmacodynamics, Glaritus® was found bioequivalent to Lantus® in 40 healthy volunteers [16] and 111 type 1 diabetics [17] under 24 hours euglycemic clamp settings glucose lowering effect and insulin glargine exposure.
The present study is the first Indian Phase IV Trial on biosimilar insulin glargine that is designed considering international guidelines such as from EMA [18] & USFDA [19] with “treat to target” strategy for study treatment.
2. Background and Rationale
Anti-insulin antibody (AIAs) can potentially alter insulin pharmacokinetics and pharmacodynamics in patients treated with modern, genetically engineered insulin analogues [20] . Such patients may present with insulin resistance, because AIAs not only bind to and sequester acutely administered insulin, but are also a source of long-acting bioavailable insulin, since insulin dissociates from these complexes in the fasting state [21] . Recently, the prevalence and levels of AIAs have remarkably decreased because of the purity of insulin preparations and the use of recombinant human insulin preparations [22] , but it has not been completely eliminated. Multiple sporadic case reports have been published reporting high titre serum AIAs while being treated with recombinant human insulin or human insulin analogues, especially in Asian populations [23] - [30] .
In phase IV study in 161 type 1 diabetics, Glaritus® was found comparable to Lantus® in terms of glycemic control (mean reduction in HbA1c from baseline 0.69% ± 1.81% vs. 0.53% ± 1.94%, respectively; P = 0.454), incidence of hypoglycaemia (27.2% vs. 28.6%, P = 0.8432) and immunogenicity (mean reduction in AIA from baseline 0.25 ± 0.97 vs. 0.02 ± 0.68, respectively; P = 0.306) over 12 weeks of treatment [31] .
The proposed protocol was developed in line with following international guidelines:
EMA guidance [20] states that immunogenicity study duration should be at least 12 months, including a comparative phase of at least 6 months, but there is no need to power the study to formally demonstrate non-inferiority regarding immunogenicity.
USFDA guidance [32] states that assessment of the immunogenic potential of insulins and insulin analogues should be performed over a period of at least 6 to 12 months. Further, it states that reduction in HbA1c should be considered for demonstration of efficacy with non-inferiority design and non-inferiority margin of 0.3 or 0.4 HbA1c percentage units.
3. Objectives
Primary objective of this study is to evaluate the change in immunogenic response (measured as percentage change in AIA using a quantitative ELISA test) to glargine in Glaritus® and Lantus® treatment arms from baseline to 6 months. The secondary objectives are to 1) evaluate the change in immunogenic response to glargine in Glaritus® treatment arm from baseline to 12 months; 2) assess and compare efficacy in both treatment arms at 6 months in terms of change in HbA1c from baseline; and 3) assess and compare subjects’ safety in both treatment arms in terms of adverse events (including hypoglycaemia events). Further, an exploratory endpoint of assessing efficacy in Glaritus® treatment arm at 12 months in terms of change in HbA1c from baseline has also been included to judge sustenance or further improvement of glycaemic control obtained at 6 months.
4. Material and Method
4.1. Study Design
This is a prospective, open-label, randomized, active-controlled, non-inferiority, parallel-group, multi-centre, Phase IV study conducted in 180 subjects who had inadequately controlled T2DM with use of oral hypoglycaemic agents (OHAs). The study design is shown in Figure 1.
The study included a screening phase (maximum 2 weeks), treatment phase (12 months for Glaritus® arm and 6 months for Lantus® arm) and a follow-up phase (1 month for both arms as consensual supply). After providing a written informed consent for participation in the study, subjects underwent screening phase, which included recording of subject’s demographic characteristics and medical history, assessment of vital signs, physical examination, laboratory tests, including urine pregnancy test, and 12-lead electrocardiogram (ECG). The treatment phase started at 10 units (or 0.2 units/kg) once daily on Day 1 (Visit 2) immediately following randomization of subjects to 2 treatment arms in a 1:1 ratio by an interactive web response system. After Visit 8, in the Lantus® arm 1-month supply of the same is being provided for compassionate use followed by
Figure 1. Study design. *Dose was subsequently titrated based on the subject’s FBG.
Visit 9, which will constitute the end-of-study. After Visit 14, 1-month supply of the same is being provided for compassionate use followed by Visit 15, which will constitute the end-of-study.
The laboratory tests were performed by the central laboratory (Metropolis Healthcare Ltd., Mumbai). The trial was registered at Clinical Trial Registry of India before enrolment of first subject with no. CTRI/2015/05/005808 on 25/05/2015. Study was initiated on each site only after approval of institutional ethics committee and approval of Drug Controller General (India). Written informed consent was obtained from each study subject before starting any study related activities with them.
4.2. Duration of Treatment
In line with the EMA and USFDA guidance, comparative phase consisting of parallel treatment arms were considered for 6 months followed by continuation of only Glaritus® arm further till 12 months. Considering 1 additional month of consensual supply, the total duration of the study was expected to be approximately 410 days for subjects in the Glaritus® arm and approximately 225 days for subjects in the Lantus® arm.
4.3. Dosing
The study utilizes treat-to-target (TTT) strategy for insulin dose titration endorsed by the FDA as a means of evaluating the different insulins’ therapeutic potential. The starting doses of Glaritus® and Lantus® were recommended to be 10 units (or 0.2 units/kg) once daily, which had to be subsequently titrated based on the subject’s FBG (target FBG 70 to 130 mg/dL). Dose was titrated based on the average 3 FBG levels recorded by the subject using a glucometer and history of nocturnal hypoglycaemia. The dose titration algorithm is shown in Figure 2.
4.4. Concomitant Antidiabetics
Investigators are permitted to use any medication concomitantly as per their discretion and as per standard of medical care. However, details of all such concomitant drugs were recorded in CRF from 15 days prior to first dose of study treatment till end of the same.
4.5. Patients
Male or female subjects, ≥18 and ≤55 years of age, with uncontrolled T2DM were enrolled in the study. Additionally, subjects were to have the body mass
index (BMI) ≥ 18.0 and ≤ 38.0 kg/m2 and HbA1c levels ≥ 8.0% and ≤ 10.0%. Subjects had to be insulin-naïve or had received insulin for short term (≤2 weeks) only and ≥6 months prior to enrolment. Subjects had to be inadequately controlled by 1 or more OHAs and according to investigator needed glargine treatment as standard-of-care. Subjects had to be able to use the self-glucose-monitoring device and to self-administer insulin and willing to record the daily FBG values and the insulin dose in a subject diary. Schematic representation of the patient flow is given in Figure 3.
4.6. Data Analysis and Statistical Considerations
Considering the estimated difference in mean changes in HbA1c for Glaritus® versus Lantus® as 0 and a non-inferiority margin fixed at 0.4, at level of significance alpha at 0.025 with power of 80%, 72 completers are required in each treatment arm. Considering drop-out rate of 10%, initial sample size was calculated at 160, which was subsequently revised based on high drop-out rate of 20% to 180.
This interim analysis was undertaken after 100 subjects have completed their visit 8 (6 months treatment) needed for assessment of primary objective across 10 centres. Further, second interim analysis was added once the required number of completers (144) completed their visit 8 to analyse the comparative phase data. Final analysis would be done once Glaritus® arm also completes their study treatment.
The statistical evaluations will be performed using SAS®, version 9.2 or later. For change in AIA and HbA1c at 6 months from baseline, two-sided 95% confidence interval (CI) and p-value for the difference between treatments will be calculated using an analysis of covariance (ANCOVA) with treatment as main factor and baseline values as covariate. The Cochrane-Mantel-Haenzel test stratified by (pooled) center will be used to test for differences in the overall incidence of hypoglycaemia.
For the interim analysis, the results for the immunogenicity, efficacy, and safety are presented by descriptive statistics.
5. Early Data Trends
From July 2015 to November 2017, 180 patients have been enrolled in the study. Last (100th) subject completed visit 8 in March 2017 and interim analysis of the data was completed in July 2017. Both treatment groups were comparable in
demographic characteristics and baseline parameters. Trend of similarity was observed between Glaritus® and Lantus® in the data analysed as follows:
• For mITT population, AIA decreased in the Glaritus® group by 10.52% ± 23.06%, whereas, it increased by 0.48% ± 63.95% in the Lantus® group
• For mITT population, HbA1c reduction was 1.09% ± 1.29% in the Glaritus® group as compared to 0.63% ± 1.19% in the Lantus® group
• Hypoglycaemic episodes were reported in 1 subject in the Glaritus® group and 2 subjects in the Lantus® group
• Adverse events were observed in 8/60 subjects (13.3%) in the Glaritus® group and 9/59 subjects (15.3%) in the Lantus® group
6. Conclusion
Glaritus® has shown comparable efficacy and safety to Lantus® in the pre-clinical as well as clinical studies conducted so far. The present study represents a randomised, controlled design in line with international guidelines on biosimilar insulin glargine development with assessment of immunogenicity, glycaemic control, hypoglycaemic events and other adverse events in a comparative phase for 6 months and in a single arm phase for another 6 months thereafter. Further, the early data trends from first interim analysis presents expected similarity of Glaritus® in immunogenicity, efficacy and safety to that of Lantus® in treatment of T2DM.