August 2015 Australian Public Assessment Report for Insulin glargine

IV. Clinical findings

Introduction

A summary of the clinical findings is presented in this section. Further details of these clinical findings can be found in Attachment 2.

Clinical rationale

The rationale for the drug development program is stated by the sponsor as:

‘The development plan for Abasria, informed by the scientific principles set forth in the Committee for Medicinal Products (CHMP) guidances on biosimilars and adopted by the TGA, reflects a stepwise approach to demonstrating the similarity of ABASR1A to the reference medicinal product (Lantus). The aim of the Abasria development program was to demonstrate that Abasria has a highly similar profile to Lantus in terms of quality, nonclinical, pharmacokinetics and pharmacodynamics, and clinical safety and efficacy aspects, allowing Abasria to adopt the data generated with Lantus and thus the Australian Product Information (AUPI) for Lantus.’

The drug development program is described in these terms:

‘The totality of data presented in this application, specifically the clinical data summarised in Module 2, support a sufficient demonstration of similarity of Abasria to Lantus:

The primary goal of the development program was achieved: Comparative PK and PD studies demonstrated highly similar PK and PD of Abasria to Lantus (Study ABEA) and of EU-approved Lantus to US-approved Lantus (Study ABEN) within predefined bioequivalence acceptance limits.

Study ABEN established a scientific bridge that justified presenting the analyses of clinical efficacy and safety with a comparator group comprising EU- and US approved Lantus in the multinational Phase III clinical studies (ABEB and ABEC). The scientific bridge was supported by subgroup analyses in the Phase III studies comparing the treatment effect of Abasria to either EU- or US-approved Lantus for selected efficacy and safety parameters, which showed no clinically meaningful differential treatment effects between Abasria and Lantus (irrespective of source).

Clinical data from Studies ABEB (T1DM) and ABEC (T2DM) provide evidence that Abasria and Lantus have equivalent efficacy by meeting the primary test of the noninferiority of ABASR1A to Lantus as well as the secondary, complementary test of the noninferiority of Lantus to Abasria with respect to change in HbA1c, and with no statistically significant difference between treatment groups for key secondary measures of efficacy.

Clinical safety data from the Phase III studies demonstrate a highly similar safety profile (including immunogenicity, allergic reactions, and hypoglycemia) of Abasria to Lantus. Importantly, the development of anti-insulin glargine antibodies (as measured by TEAR) was not associated with any detrimental effect on efficacy and safety outcomes in patients with 11DM or T2DM.’

Comment: As is evident from the submission, the drug development program was informed by discussions with the FDA and with the EMA.

Guidance

Adopted guidelines and statements apply to this submission.

EMEA/CHMP/BMWP/32775/2005 Annex To Guideline On Similar Biological Medicinal Products Containing Biotechnology-Derived Proteins As Active Substance: Non-Clinical And Clinical Issues Guidance On Similar Medicinal Products Containing Recombinant Human Soluble Insulin.

Guideline On Similar Biological Medicinal Products Containing

Biotechnology-Derived Proteins As Active Substance:

Non-Clinical And Clinical Issues

Guideline on the Investigation of Bioequivalence (CPMP/EWP/QWP/1401/98 Rev. 1/ 2010) (CHMP 2010 Guideline on Immunogenicity Assessment Of Biotechnology-Derived Therapeutic Proteins).

Guideline On Immunogenicity Assessment Of Biotechnology-Derived

Therapeutic Proteins

Evaluation of biosimilars Version 1.0, TGA July 2013 (now under review).

Contents of the clinical dossier

The submission contained the clinical information as outlined in the table above.

five clinical pharmacology studies, all of which generated pharmacokinetic data and also pharmacodynamic data.

no population pharmacokinetic analyses.

one pivotal efficacy/safety study (ABEC).

no dose-finding studies.

one other efficacy/safety study (ABEB).

Literature references.

Paediatric data

The submission did not include paediatric data.

Good clinical practice (GCP)

As appended to the letter of application:

‘I certify that Eli Lilly Australia Pty. Limited is in possession of documentation to demonstrate that the clinical studies accompanying the letter of 01 October 2013 were carried out in accordance with the principles of the Declaration of Helsinki and, if conducted in Australia, in accordance with the NH&MRC ‘Statement on Human Experimentation’.

I further certify that such documentation will be provided to the Department of Human Services and Health within three months of any request.

I understand that the documentation referred to includes Ethics Review Committee approval letters, signed subject consent forms and the patient information sheet if there is one.’

The evaluator mentioned in the discussion of each study any relevant matters in regard to GCP, ethical certification and auditing. In brief, no major concerns were noted but some clarifications from the applicant might be needed.

Pharmacokinetics

Studies providing pharmacokinetic data

Table 2 shows the studies relating to each pharmacokinetic topic and the location of each study summary.

Table 2. Submitted pharmacokinetic studies.

PK topic	Subtopic	Study ID	*
PK in healthy adults	General PK- Single dose	ABEA ABEM ABEI	PK & PD PK & PD PK & PD
	Multi-dose
	Bioequivalence† - Single dose	ABEN ABEA ABEM ABEI	PK & PD of Lantus US v. Lantus EU PK comparisons PK comparisons PK comparisons
	- Multi-dose	Not submitted.
	Food effect	Not applicable.
PK in special populations	Target population - Single dose	ABEE	PK & PD

* Indicates the primary aim of the study. † Bioequivalence of different formulations. § Subjects with Type 1 Diabetes Mellitus (T1DM).

None of the pharmacokinetic studies in healthy volunteers had deficiencies that excluded their results from consideration. However, some studies are considered to be less relevant than others for the reasons briefly stated in Table 3.

Table 3 lists pharmacokinetic results that are considered to be less relevant due to study deficiencies.

Table 3. Pharmacokinetic results excluded from consideration.

Study ID	Subtopic(s)	PK results excluded
ABEE	Pharmacodynamics of Abasria Compared to Lantus^® in Subjects with Type 1 Diabetes Mellitus	All results – assay insensitivity led to incomplete characterisation of the PK attributes of Abasria in this study.

The design of ABEE is briefly described in the tabulation below. The study did generate useful pharmacodynamic data.

Table 4: Design of study ABEE

Evaluator’s conclusions on pharmacokinetics

An important assumption that this evaluator makes is that the analytical method will be found to be satisfactory by the chemistry evaluator.

The Phase I studies marginally address the requirements of the adopted EU guideline EMEA/CHMP/BMWP/32775/2005 annex to guideline on similar biological medicinal products containing biotechnology-derived proteins as active substance:

Non-clinical and clinical issues - Guidance on similar medicinal products containing recombinant human soluble insulin,

‘The relative pharmacokinetic properties of the similar biological medicinal product and the reference medicinal product should be determined in a single dose crossover study using subcutaneous administration. Comprehensive comparative data should be provided on the time-concentration profile (AUC as the primary endpoint and Cmax, Tmax, and T1/2 as secondary endpoints). Studies should be performed preferably in patients with type1 diabetes. Factors contributing to PK variability e.g. insulin dose and site of injection / thickness of subcutaneous fat should be taken into account.’

The Phase I studies used consistent methods and appear to have been conducted diligently. The one study in subjects with Type 1 Diabetes Mellitus (T1DM) (Study ABEE) yielded uninterpretable PK results. Hence the useful data came from studies in healthy volunteers – the analytical method required C-peptide correction; the assay method was not specific to insulin glargine and its principal metabolite.

C-peptide correction is essential in studies involving healthy volunteers and it is accord with the need to have some form of baseline correction as articulated in the adopted guideline CPMP/EWP/QWP/1401/98 Rev. 1/ Corr ** Guideline On The Investigation Of Bioequivalence

Sampling times

‘A sufficient number of samples to adequately describe the plasma concentration-time profile should be collected. The sampling schedule should include frequent sampling around predicted tmax to provide a reliable estimate of peak exposure. In particular, the sampling schedule should be planned to avoid Cmax being the first point of a concentration time curve. The sampling schedule should also cover the plasma concentration time curve long enough to provide a reliable estimate of the extent of exposure which is achieved if AUC(0-t) covers at least 80% of AUC(0-∞). At least three to four samples are needed during the terminal log-linear phase in order to reliably estimate the terminal rate constant (which is needed for a reliable estimate of AUC(0-∞))…’

‘For endogenous substances, the sampling schedule should allow characterisation of the endogenous baseline profile for each subject in each period. Often, a baseline is determined from 2-3 samples taken before the drug products are administered…’

It is noted that the studies did not always achieve enough duration of sampling to achieve 80% of AUC from time 0 to infinity (AUC_(0-∞)). The extrapolation of AUC exceeded 20% in ABEA, ABEI, ABEM and ABEN.

‘Endogenous substances

If the substance being studied is endogenous, the calculation of pharmacokinetic parameters should be performed using baseline correction so that the calculated pharmacokinetic parameters refer to the additional concentrations provided by the treatment. Administration of supra-therapeutic doses can be considered in bioequivalence studies of endogenous drugs, provided that the dose is well tolerated, so that the additional concentrations over baseline provided by the treatment may be reliably determined.

If a separation in exposure following administration of different doses of a particular endogenous substance has not been previously established this should be demonstrated, either in a pilot study or as part of the pivotal bioequivalence study using different doses of the reference formulation, in order to ensure that the dose used for the bioequivalence comparison is sensitive to detect potential differences between formulations.

The exact method for baseline correction should be pre-specified and justified in the study protocol. In general, the standard subtractive baseline correction method, meaning either subtraction of the mean of individual endogenous pre-dose concentrations or subtraction of the individual endogenous predose AUC, is preferred. In rare cases where substantial increases over baseline endogenous levels are seen, baseline correction may not be needed.

In bioequivalence studies with endogenous substances, it cannot be directly assessed whether carryover has occurred, so extra care should be taken to ensure that the washout period is of an adequate duration.’

The use of C-peptide correction is in principle reasonable owing to the lack of a specific assay.

Pharmacodynamics

Studies providing pharmacodynamic data

Table 5 shows the studies relating to each pharmacodynamic topic.

The studies are the same Phase I studies that have been considered in regard to pharmacokinetics. The used the same euglycaemic clamp method with the same sampling times and data management, excepting that those studies in healthy volunteers ran for only 24 hours. The results across studies were as expected consistent and comparable.

Table 5: Submitted pharmacodynamic studies.

PD Topic	Subtopic	Study ID	*Aim of Study
Primary Pharmacology	Effect on glucodynamics in a euglycaemic clamp study	ABEA ABEM ABEI ABEE	PK and PD PK and PD PK and PD PK and PD
Secondary Pharmacology	Effect on C-peptide levels in healthy volunteers#	ABEA ABEM ABEI	PK PK PK
Gender other genetic and Age-Related Differences in PD Response	Effect of gender	Not done
	Effect of age	Not done
Comparison of Lantus EU versu Lantus US	‘Scientific Bridge’ to support Phase III studies use of both sources of Lantus	ABEN	PK and PD
Population PD and PK-PD analyses	Healthy subjects	Not done
	Target population	Not done

* Indicates the primary aim of the study.

§ Subjects who would be eligible to receive the drug if approved for the proposed indication.

# C-peptide levels are presented graphically as Figures only.

None of the pharmacodynamic studies had deficiencies that excluded their results from consideration.

Evaluator’s conclusions on pharmacodynamics

The Phase I studies were more successful as PD studies than as PK studies notwithstanding the duration of most studies (24 hours).

The Phase I studies adequately address the requirements of the adopted EU guideline EMEA/CHMP/BMWP/32775/2005 Annex to guideline on similar biological medicinal products containing biotechnology-derived proteins as active substance: Non-Clinical And Clinical Issues - Guidance On Similar Medicinal Products Containing Recombinant Human Soluble Insulin, noting that the short duration of studies in healthy volunteers could not capture the full duration of action of insulin glargine and that the study in patients with T1DM did not capture the full duration of action of insulin glargine in a minority of subjects in each and/or both periods:

‘The clinical activity of an insulin preparation is determined by its time-effect profile of hypoglycaemic response, which incorporates components of pharmacodynamics and pharmacokinetics. Pharmacodynamic data are of primary importance to demonstrate comparability of a similar rh-insulin. The double-blind, crossover hyperinsulinaemic euglycaemic clamp study is suitable for this characterisation. Data on comparability regarding glucose infusion rate and serum insulin concentrations should be made available. The choice of study population and study duration should be justified. Plasma glucose levels should be obtained as part of the PK study following subcutaneous administration.‘

Abasria has a PD profile equivalent to Lantus and this is shown directly by glucodynamic parameters and by the influence on C-peptide levels in healthy volunteers. The Lantus used was EU derived Lantus.

Dosage selection for the pivotal studies

Abasria is modelled as a biosimilar version of Lantus, and the nonclinical (preclinical) data supported comparability of Lantus and Abasria. Therefore, the Phase III Studies treated the enrolled patients according to the locally approved PI of Lantus and according to a reasonable treatment algorithm (Study ABEC) or reasonable principles (Study ABEB).

Efficacy

Studies providing efficacy data

Lantus is registered with the following composite indication, as represented in the PI:

‘Insulin glargine is an insulin analogue indicated for once-daily subcutaneous administration in the treatment of Type 1 diabetes mellitus in adults and children and Type 2 diabetes mellitus in adults who require insulin for the control of hyperglycaemia.’

The first indication is subdivided in to adults and children; the second is limited to adults.

There are two Phase III studies in this submission, Study ABEC and Study ABEB.

Table 6: Brief description of the Phase III studies. Phase III Efficacy and Safety studies supporting the use of LY2963016 in patients with T1DM and T2DM

Evaluator’s conclusions on efficacy

The first indication is subdivided in to adults and children; the second is limited to adults.

In regard to T1DM, Study ABEB satisfied the requirements of its a priori sample size calculations and analytical plan. It was conducted in the light of advice received from the FDA and from the EMA. The study groups were well matched. However, Study ABEB was of open label design and it is uncertain as to how much this might have affected the behaviour of investigators in order to achieve similar improvements in both groups against baseline glycated haemoglobin (HbA1c) readings. Within this major limitation, Study ABEB supports equi-efficaciousness of Abasria and Lantus as the basal insulin component in the treatment regimen of patients with T1DM. The supportive efficacy outcomes in Study ABEB are of doubtful clinical value.

In regard to Type 2 Diabetes Mellitus (T2DM), the evaluator accepts that comparable efficacy to Lantus EU and Lantus US was shown in Study ABEC, a study that exceeded the predetermined sample size and that had a modest dropout rate. The results of the study show comparable efficacy, including comparable improvement from baseline in both treatment groups that is comparable also in terms of subgroups of previous Lantus use or insulin-naïve patients. The patient population was reasonably representative of patients with T2DM who require basal insulin, matching Lantus’ registered indication. It is clear that a large majority of subjects was not tightly controlled at study entry.

Safety

Studies providing safety data

The following types of studies provided evaluable safety data:

Phase I Studies

Phase III Studies

The Phase I Studies were numerically dominated by the use of healthy volunteers, so most adverse events were procedure related (invasive procedures, inter-current illnesses and hypoglycaemia). No new safety concerns arose from these Phase I studies.

The erratic absorption of insulin, in some individuals in at least one period of the replicate studies, suggests limits to the reliability of insulin glargine (whether as Abasria or Lantus) as a slow release pharmaceutical.

The Phase III Studies are of uneven quality (ABEC was blinded; ABEB was open) but both were large enough to define common adverse events and both included a blinded centralised review of possible immunological adverse events as well as binding activity of insulin antibodies.

The evaluator has referred to ABEB and ABEC as ‘Phase III Studies’ rather than ‘pivotal’ studies in this section because it was not considered that ABEB is a pivotal study.

Phase III efficacy studies

In the Phase III efficacy studies, the safety data were collected according to this tabulation:

Table 7: Study schedule. Protocol I14L-MC-ABEC

Phase III studies that assessed safety as a primary outcome

Neither Study ABEC nor Study ABEB had specific safety objectives as primary outcomes. It is noted that both had one secondary outcome that is arguably safety-related,

‘To compare LY2963016 relative to Lantus^® with regard to intrapatient blood-glucose (BG) variability; basal and prandial (separately and as total daily) insulin dose; and weight when used in combination with pre-meal insulin lispro.’ This outcome has been presented in the efficacy discussion of each study (See sections 21.7 and 22.4.1.3 of this report). Study ABEC had also this secondary outcome: ‘To compare the safety of LY2963016 relative to Lantus^® (eg, incidence of anti-insulin antibodies, hypoglycemia, adverse events [AEs]) when used in combination with OAMs’.

Dose-response and non-pivotal efficacy studies

Not applicable.

Other studies evaluable for safety only

Not applicable.

Phase three studies that assessed safety as a primary outcome

Not applicable.

Patient exposure

Phase III studies

As noted in the sponsor’s Safety summary, a total of 536 patients with T1DM and 759 patients with T2DM were randomly assigned to treatment in Studies ABEB and ABEC, respectively. Of these patients, a total of 535 patients with T1DM and 756 patients with T2DM received at least one dose of randomly assigned study drug, comprising the Full Analysis Sets (FAS) and serving as the populations of interest for analyses in the applicant’s safety analyses. The mean duration of exposure for patients in Study ABEB was 23.32 weeks and 23.66 weeks for the Abasria and Lantus^® groups, respectively; mean duration of exposure in Study ABEC was 22.38 and 22.13 weeks, respectively. The dose of insulin glargine was according to the same treatment paradigm in each study; the locally approved PI informed the use of insulin glargine.

The following comparative table is from sponsor’s Safety Summary; it matches the patient numbers in the study reports.

Table 8: Exposure (by duration) to Abasria and Lantus in Phase III clinical studies.

There is no experience beyond the cut-off point of Studies ABEB and ABEC, that is, 24 weeks on treatment.

Safety issues with the potential for major regulatory impact

Neoplasms

A few neoplasms were reported in both studies. The matter deserves long term targeted surveillance because:

Insulin glargine is agonistic at the Insulin-like Growth factor (IGF-1) receptor

The approved PI of Lantus states,
‘IGF-1 receptor binding: The affinity of insulin glargine for the human IGF-1 receptor is approximately 5 to 8-fold greater than that of human insulin (but approximately 70 to 80-fold lower than the one of IGF-1), whereas M1 and M2 bind the IGF-1 receptor with slightly lower affinity compared to human insulin. The total therapeutic insulin concentrations (insulin glargine and its metabolites) found in type 1 diabetic patients was markedly lower than what would be required for a half maximal occupation of the IGF-1 receptor and the subsequent activation of the mitogenic-proliferative pathway initiated by the IGF-1 receptor. Physiological concentrations of endogenous IGF-1 may activate the mitogenic proliferative pathway; however, the therapeutic concentrations found in insulin therapy, including in Lantus therapy, are considerably lower than the pharmacological concentrations required to activate the IGF-1 pathway.’

The above might not always apply for example, when early release of insulin glargine occurs from the site of injection, as happened in the Phase I studies. A non-selective assay was used but it is likely that insulin glargine, not M1 was released. Consequently, intermittent release of insulin glargine from the site of injection, in a setting of long term use, has not been excluded.

The matter has been discussed in the literature since the publication of a retrospective cohort study of German health insurance fund records.⁸ A dose-dependent increase in cancer risk was found for treatment with insulin glargine compared with human insulin. The matter has been kept under review by the EMA. The most recent statement was published on 31 May 2013 (EMA/329790/2013 EMEA/H/C/000309). The statement commenced with,

‘On 30 May 2013, the European Medicines Agency completed a review of new data on the cancer risk with insulin glargine-containing medicines. The Agency’s Committee for Medicinal Products for Human Use (CHMP) concluded that the data do not show an increased risk of cancer and that the balance of the medicine’s benefits and risks remains unchanged.’

New information was derived from two further cohort studies and from a case-control study.

‘Based on the assessment of the population-based studies, the CHMP concluded that overall the data did not indicate an increased risk of cancer with insulin glargine, noting that there is no known mechanism by which the insulin glargine would cause cancer and that a cancer risk has not been seen in laboratory studies. As for all medicines, the Agency will continue to assess any new data that become available in this area, as part of the routine monitoring of the medicine.’

There is difference between ‘no known mechanism by which the insulin glargine would cause cancer’ (there is no suggestion that insulin glargine is for example genotoxic) and a potential to promote tumours by an agonistic effect at IGF-1 receptors. Consequently, postmarketing surveillance will possibly be contributory but the studies would need to be long term and be capable of dealing with confounders such as HMG CoA reductase inhibitors and low dose aspirin, both of which are commonly prescribed to diabetic patients.

Nonetheless, there is no basis for suggesting that Abasria presents a different degree of risk from Lantus, so registration of Abasria cannot be opposed on the grounds of potential neoplasia.

Postmarketing data

At the time the TGA assessed this application, marketing had not yet occurred in any country.

Evaluator’s conclusions on safety

The Phase III studies are of sufficient size and duration to establish in terms of common adverse events. They enrolled reasonably representative populations of Type 1 and Type 2 diabetics that were using treatment regimens relevant to recommended clinical practice in this country. The study in Type 1 diabetics has an ongoing extension phase that should be submitted as a postregistration commitment. Its open design admits the possibility of bias.

No new safety signals emerged and Abasria was not worse than Lantus in terms of the frequency of serious adverse events. Abasria appears to be registrable on clinical safety grounds. The 12 month data on Study ABEB should be submitted for evaluation when they become available.

First round benefit-risk assessment

First round assessment of benefits

The benefits of Abasria in the proposed usage are:

Not different from those of Lantus EU and Lantus US based on two way non-inferiority in two Phase III studies.

First round assessment of risks

The risks of Abasria in the proposed usage are:

Not different from those of Lantus EU and Lantus US based on the experience in two Phase III studies.

There is an unresolved potential problem about the safety in use of the cartridge in an unspecified pen injector device (possibly HumaPen). There is in addition an unresolved issue concerning clinical data to support the KwikPen device. No data are able to be located by the evaluator to address these concerns.

First round assessment of benefit-risk balance

The benefit-risk balance of Abasria cartridges is unfavourable given the proposed usage, but would become favourable if the uncertainty raised should be resolved.

First round recommendation regarding authorisation

Registration should not proceed at present, pending resolution of the matters raised.

Submission of the completed (52 weeks of data) Study ABEB should be a condition of registration.

Clinical questions

Pharmacokinetics

1. 
   In Study ABEN, the source of the US Lantus in unclear. The applicant should confirm and clarify how this can be correct.
   
2. 
   The applicant should clarify who was the investigator and whose signature appears on I4L-MC-ABEI CSR Appendix Signature. The applicant should also confirm that Lantus EU was used in Study ABEI.
   

Second round evaluation of clinical data submitted in response to questions

The sponsor responded to the issues raised about the KwikPen as follows:

‘There is no basis for the assumption that medication error is more likely to occur with Abasria compared to other insulins currently available, including the reference product in this application, without a dedicated pen and should therefore be removed from the evaluation report. The evaluator also states no information is available on the KwikPen device with respect to dose accuracy. This statement is incorrect as Module 3.2R.3 includes the required device testing information.’