Evaluation of the Dual-Chamber Pen Design for the Injection of Exenatide Once Weekly for the Treatment of Type 2 Diabetes

Abstract

Background:

Exenatide once weekly, an injectable glucagon-like peptide-1 receptor agonist, has been shown to reduce A1C, fasting glucose, and body weight in patients with type 2 diabetes. Exenatide 2.0 mg is dispersed in poly-(D,L-lactide-co-glycolide) polymer microspheres, which require resuspension in aqueous diluent before subcutaneous injection. A single-use, dual-chamber pen was developed to improve the convenience of exenatide once weekly delivery and tested following Food and Drug Administration (FDA) guidance.

Methods:

Design development goals were established, and validation tests (dose accuracy, torque/force requirements, usability, and ease-of-use) were performed. Dose accuracy was tested under a variety of conditions. After 10 exploratory studies in 329 patients, the final design’s usability and ease-of-use were tested in untrained health care practitioners (HCPs; n = 16) and untrained/trained patients (n = 30/17). Usability testing evaluated completion of multiple setup, dose preparation, and injection steps. Ease-of-use impression was assessed using a scale of 1−7 (1 = very difficult, 7 = very easy).

Results:

The dual-chamber pen successfully met development goals and delivered target volume (650 µL ± 10%) under tested conditions (mean 644.7–649.3 µL), with torque and force requirements below prespecified maximum values. In the final user study, most participants (≥87%) correctly completed pen setup, dose preparation, and injection steps. Mean ease-of-use scores were 5.8, 6.3, and 6.5 out of 7 in untrained HCPs, untrained patients, and trained patients, respectively.

Conclusion:

With self-education or minimal training, participants accurately and precisely suspended, mixed, and delivered exenatide-containing microspheres using the dual-chamber pen with high ease-of-use scores. The dual-chamber pen was FDA-approved in February 2014.

Exenatide is the first-in-class glucagon-like peptide-1 receptor agonist (GLP-1RA) originally approved in 2005 by the US Food and Drug Administration (FDA) as a twice daily premeal subcutaneous injection administered by pen for the treatment of hyperglycemia in patients with type 2 diabetes mellitus (T2DM).¹ GLP-1RAs increase glucose-dependent insulin secretion, suppress glucagon secretion, and slow gastric emptying.² The exenatide molecule shares approximately 53% structural similarity to human GLP-1 but is resistant to enzymatic degradation by dipeptidyl peptidase-4 resulting in a prolonged physiological response and an increased half-life of 2.4 hours compared with 2 minutes for native GLP-1.^1-3

In 2012, the US FDA approved a long-acting formulation of exenatide for once-weekly administration that can be administered regardless of meal times;⁴ European Medicines Agency approval occurred in 2011. The long-acting, sustained release of exenatide is achieved by dispersion of the parent exenatide molecule in poly-(D,L-lactide-co-glycolide) polymer microspheres approximately 0.06 mm in diameter. Similar polymer material has been used in surgical sutures, bone plates, and orthopedic implants for decades.⁵ These exenatide-containing microspheres must be suspended in an aqueous diluent just prior to injection to preserve drug stability. After injection of microspheres into the subcutaneous space, gradual biodegradation of the polymer matrix into CO₂ and H₂O releases exenatide continuously over an extended duration. With regular once-weekly dosing, mean exenatide levels exceed the minimal effective concentration (50 pg/mL) in 2 weeks and achieve steady state in 6–7 weeks, resulting in consistent concentrations of exenatide.⁶ Treatment with exenatide once weekly has been shown to provide long-term (through 6 years) reductions in glycated hemoglobin (HbA1c), fasting plasma glucose (FPG), body weight, and other cardiovascular measures, with a low risk of hypoglycemia.^7-10 The most commonly reported adverse events with exenatide once-weekly therapy are transient mild-to-moderate gastrointestinal events and injection site reactions (nodules, pruritus).¹¹ Injection-site nodules are firm soft-tissue lumps approximately 0.5−0.75 mm in diameter that are an inflammatory response to injected microspheres.⁵ In clinical trials, the incidence of nodule-associated adverse events (prolonged pain, induration, bleeding, redness, or inflammation) was 0.7–6.7%.⁵

Exenatide once weekly was first supplied as a single-dose injection tray that includes all the necessary components needed for administration (microsphere vial, prefilled syringe containing aqueous diluent, vial connector, and custom needle).⁴ While the process becomes routine over time, the multiple steps of using the single-dose injection tray include the transfer of diluent from the syringe into the vial of microspheres via the connector, mixing of diluent and microspheres, drawing the mixed suspension back into the syringe, connecting the needle, and injecting. Injectable therapies may be viewed by some patients with T2DM as inconvenient or beyond their skill to manage,^12,13 and pen injector systems that simplify medication delivery are often well regarded and preferred over traditional needle and syringe delivery.^14,15

Thus, a device enhancement from the single-dose tray for administration of exenatide once weekly was designed and developed to minimize assembly and simplify reconstitution of the exenatide-containing microspheres in accordance with the US FDA’s Draft Guidance for Industry and Food and Drug Administration Staff—Applying Human Factors and Usability Engineering to Medical Devices to Optimize Medical Device Design (version: June 2011). This FDA guidance includes core recommendations for the development of medical devices to help reduce use errors, injuries, and product recalls.¹⁶ The core recommendations include a comprehensive risk analysis and human factors validation testing. Risk analysis takes into account the intended device use, determines possible anticipated and unanticipated use-related hazards, and assigns a priority for addressing these concerns (Figure 1). Human factors validation testing is recommended if the analysis indicates that there may be a moderate to high risk of use error. Human factors testing should incorporate realism and completeness so that the final production version is tested under the expected clinical use environment with the intended user. Design improvements as a result of the risk analysis and testing should aim to remove hazards, improve error tolerance, alert the user, and, if needed, include written instructions to improve safe use.

Figure 1.

Risk analysis for medical devices. HCP, health care practitioner; ICU, intensive care unit. Adapted from the 2011 Draft Guidance for Industry and Food and Drug Administration Staff—Applying Human Factors and Usability Engineering to Optimize Medical Design.¹⁶

Risk Analysis and Development Goals

The target patient profile for the single-use dual-chamber pen included patients with T2DM receiving therapy for multiple years who may be aged with visual or motor skill impairments, may be injection naïve, may have failed oral glucose-lowering therapy, or may be migrating from other injection pens and devices. A secondary use population included health care practitioners (HCPs) who may train patients on the use of the device. The device was intended to be used in any private location (including travel destinations) with minimal or controllable noise levels. After identifying the intended user and use environment, a review of known device-user system problems of other similarly marketed pen injectors (human growth hormone, follicle-stimulating hormone, and parathyroid hormone) was conducted to help identify and prioritize risks and mitigate use errors. From this risk analysis, multiple development goals were conceived for the dual-chamber pen: ease-of-use and handling, maintenance of drug and diluent in sterile compartments prior to mixing, safe needle attachment, controlled transfer of diluent into the microsphere chamber, efficient mixing of diluent and microspheres with a low torque requirement, purging of headspace prior to injection, reliable and precise dose delivery, and visual, auditory, and tactile indicators to demonstrate the successful completion of each step in the medication delivery process. A dual-chamber pen that is similar to presently marketed pens for the delivery of a variety of drugs was designed and manufactured for the delivery of exenatide once weekly in patients with T2DM. The single-use, dual-chamber pen consists of 3 major parts: a prefilled dual-chamber cartridge, pen injector, and needle (Figure 2). The aseptically prefilled dual-chamber cartridge contains wet and dry sides housed within the pen injector, which maintains separation of the diluent and exenatide microspheres until mixing is initiated by the patient. The pen injector provides the mechanism for adding the diluent into the microsphere chamber to facilitate mixing and formation of a suspension before dose delivery. A specially designed hub at the top of the injection pen allows for the attachment of the provided custom 7-mm, 23-gauge extra-thin wall custom needle designed to accommodate the microspheres and reduce the risk of clogging, which might occur with commercially available needles. The device met the maximum length and diameter specifications of 225 mm × 28 mm, and was designed to be easy to hold during the operational sequence yet portable.

Figure 2.

Schematic of the dual-chamber injection pen for exenatide once weekly.

Herein, we further describe the design features and operational sequence of the dual-chamber pen, test results for dose accuracy, torque and force requirements, and human factors usability testing which demonstrate that the final design of the dual-chamber pen was successful.

Methods

Dose Accuracy

The ability of the dual-chamber injection pen to consistently deliver an accurate volume was tested 60 times each under different temperatures (cool [5°C], standard [25°C], and hot [28°C]) and 20 times each after a variety of free-fall drop test conditions (horizontal position, vertically with cap up, and vertically with cap down). Because the exenatide microspheres may agglomerate at temperatures >30°C, the temperature for the hot condition was reduced from the maximum value stated in the ISO 11608-1 standard using a risk-based assessment. The dual-chamber pen design was considered to have passed the dosing accuracy test if it delivered a volume of 650 µL ± 10% under these testing conditions. The volume and dose of the active ingredient in the exenatide once-weekly dual-chamber pen is the same as that of the single-dose injection tray.

Torque and Force Requirements

The torque required to attach the needle to the needle hub and to dial the pen from the starting position through the mixing and priming steps with the needle attached was measured 30 times with design goal values of ≤0.25 Nm and <0.35 Nm, respectively. The injection force required to sustain delivery of dose was measured 30 times, and the goal for the pen design to be considered successful was ≤25 N.

Usability

A series of 10 usability studies were conducted in 329 patients to guide development of the pen design. To evaluate the usability of the final design, the dual-chamber injection pen was tested in 3 groups of participants including 16 untrained HCPs, 30 untrained patients, and 17 trained patients. All patients had a diagnosis of T2DM and were >21 years of age. A small proportion of patients in the testing population had diabetes-related comorbidities. The rates of diabetic retinopathy, neuropathy, and color blindness were 4.3%, 23.4%, and 2.1%, respectively. HCPs were licensed, responsible for the care of patients with T2DM, and had ≥2 years of experience in training patients to use injectable devices.

Untrained HCPs and untrained patients received the needle, injection pen, and instructions for use, and were asked to perform an injection into an injection pad after reading the instructions on their own. In the event of questions, participants were asked to call a customer support number, where a study facilitator provided answers similar to how a call center would provide support during a customer inquiry. Trained patients received education from a certified diabetes educator, which included a demonstration of how to use the pen, an opportunity to try the pen themselves, and the ability to ask questions directly to the educator. A moderator assessed the ability of participants to complete each step for dose preparation and injection as acceptable, completed with difficulty, or user error. The percentage of participants who completed each step without a user error was calculated. After demonstrating the ability to use the pen, participants rated their experience with the pen using a scale from 1 to 7 (1 = very difficult, 4 = neither easy nor difficult, and 7 = very easy).

Results

Design Features and Operational Sequence

Features aimed to improve usability and reduce user error during operation of the device were implemented into the dual-chamber pen design. After removing the pen from storage in a refrigerator and allowing to warm up to room temperature (at least 15 minutes), operation of the dual-chamber pen begins with attaching the provided needle to the needle hub. Next, while holding the pen with the needle side straight up, turning the knob creates pressure which combines the diluent and microspheres into a single chamber for mixing prior to injection. Colored directional bands on the threaded sleeve cue the patient to the direction that twisting should occur. The green label disappears with continued twisting and an audible click confirms that the required twisting motion has been successfully completed. Mixing of diluent and exenatide microspheres occurs as the patient taps the pen against the palm of the hand which may take 80 taps or more until an appropriate mix is achieved. An inspection window on either side of the pen injector allows the patient to visually assess the quality of the mixture, and a sample image of a poorly mixed and well-mixed solution appears on the enclosed instructions for use allowing the patient to confirm appropriate mixing (Figure 3).

Figure 3.

Pen window before and after mixing.

Once a well-mixed solution has been achieved and an injection site has been prepared, the patient turns the knob at the bottom of the pen while holding the pen with the needle straight up to purge the remaining air from inside the chamber. The orange label disappears and an injection button pops out from the knob to indicate that all air has been successfully purged and that the pen is primed for dose delivery. After pressing the pen into the skin, which inserts the needle into the subcutaneous space, the patient delivers the dose by fully depressing and holding down the injection button for 10 seconds. An audible click and a hard stop, detected through the dose delivery button, signals to the patient that the dose delivery button has been fully depressed. A locking mechanism keeps the dose delivery button in the depressed position inside of the knob. After the dose delivery button has been pressed and the patient holds in place for 10 seconds, the injection pen can be disposed of with the needle attached in an approved sharps container.

Dose Accuracy

The results of the dose accuracy testing appear in Table 1. In all testing conditions, the dual-chamber pen successfully delivered a total mean volume within the goal of 650 µL ± 10%. The mean volume delivered was lowest for the cool temperature compared with standard and hot temperatures.

Table 1.

Dose Accuracy Testing Results.

Preconditioning	Passingdose volume (µL)	Measureddose volume (µL)	Datapoints(n)	Pass/Fail
Atmosphere
Standard (25°C)	650 ± 10%	648.8	60	Pass
Cool (5°C)	650 ± 10%	644.7	60	Pass
Hot (28°C)	650 ± 10%	649.3	60	Pass
Free fall
Horizontal	650 ± 10%	649.2	20	Pass
Vertical—cap up	650 ± 10%	647.1	20	Pass
Vertical—cap down	650 ± 10%	644.8	20	Pass

Torque and Force Requirements

The mean (standard deviation) torque required to attach the needle to the needle hub and to dial the pen from the starting position to the mixing and priming positions was 0.0035 (0.0009) Nm and 0.16 (0.03) Nm, respectively. These values were below the goals of ≤0.25 Nm and <0.35 Nm, respectively. The mean injection force required to sustain delivery of dose was 15.6 (1.1) N, which was below the goal value of ≤25 N.

Usability

Final usability testing results in untrained HCPs, untrained patients, and trained patients are presented in Table 2. For the setup of the dual-chamber pen, successful completion of the individual use steps was achieved by the majority of untrained and trained patients and untrained HCPs overall; the rate of successful completion of setup steps ranged from 87%–100%. For the dose preparation and mixing steps, 97%–100% of participants successfully transferred the diluent into the microsphere chamber, and 100% of participants successfully tapped the pen against the palm of their hands to mix the diluent and microspheres and looked inside the window to examine the mixture. A uniform suspension was successfully achieved by 87% of untrained patients and 94% of untrained HCPs and trained patients. Prior to injecting the suspension, ≥90% of participants successfully primed the pen by twisting the knob to purge the remaining air from the chamber, and 100% of untrained HCPs and trained patients and 97% of untrained patients successfully inserted the needle and injected the dose.

Table 2.

Frequency of Completion of Individual Use Steps.

Dual-chamber pen use steps	Untrained		Trained
	HCPs	Patients	Patients
	(n = 16)	(n = 30)	(n = 17)
Setup (%)
Take pen out of refrigerator for ≥15 min prior to use	94	97	100
Wash hands	94	87	94
Open tray and remove pen and needle from package	100	100	100
Look through the inspection window	94	100	100
Remove paper tab from needle and attach to pen	100	100	100
Dose preparation and mixing (%)
Hold pen upright; twist handle to transfer diluent into microsphere chamber	100	97	100
Hold pen by knob end	94	100	100
Tap pen firmly against palm of hand to mix the diluent and microspheres	100	100	100
Look inside the mixing window (to determine if the solution is well mixed)	100	100	100
Complete mix achieved	94	87	94
Injection process (%)
Select the injection site	100	100	100
Clean the injection site	94	87	94
Hold pen upright; twist handle to prime the pen	100	90	100
Remove outer needle cover	100	100	94
Insert needle and push the button to inject the dose (hold for 10 sec)	100	97	100
Discard pen with needle attached	100	100	100

Ease-of-use scores were high overall. The mean ease-of-use score for untrained HCPs, untrained patients, and trained patients, was 5.8, 6.3, and 6.5, respectively, out of a maximum of 7.

Discussion

To simplify the delivery of exenatide once weekly for patients with T2DM, the dual-chamber pen was developed and evaluated according to FDA guidance. The 2011 US FDA draft guidance provides recommendations to guide the development of a medical device that is safe and minimizes use errors. The risk analysis process of user and use identification and risk prioritization and mitigation were instrumental in guiding the development of the initial dual-chamber pen design. A critical look at the strengths and weaknesses of similar pen devices proved to be helpful in identifying potential safety concerns that made them less than optimal for patient use (product mix-up, incorrect needle attachment, problems purging air from the pen, not holding the pen at the injection site, and failure to perform ancillary tasks [hand washing, drug inspection, cleaning the injection site, safe disposal]). The final pen design and instructions for use mitigated these concerns.

FDA recommendations for human factors validation testing which incorporates realistic use environments provided input and insight at each step in the development process leading to further design improvements as well as revisions to instructions for use. A total of 10 usability tests were conducted before the final marketed product was achieved. The results of this testing showed that the dual-chamber injection pen for delivery of exenatide once weekly successfully achieved the goals of development (Table 3).

Table 3.

Summary of Goals of Development and Achievement.

Goal of development	Evidence of achievement
Ease-of-use and handling	• Relatively high mean ease-of-use scores were recorded in all testing groups: 5.8–6.5 out of 7
Maintenance of drug and diluent in sterile compartments prior to mixing	• A dual-chambered cartridge was used to maintain wet and dry compartments prior to mixing
Controlled transfer of diluent into the microsphere chamber	• By twisting the pen knob, an internal plunger mechanism advances moving diluent into the compartment containing exenatide microspheres by pressure changes
Efficient mixing of diluent and microspheres with a low torque requirement	• The measured torque required in testing was less than half of the goal value of <0.35 Nm
	• 87%–94% of participants achieved a complete mix in usability testing
Safe needle attachment	• 100% of all study participants successfully attached the needle to the pen injector in usability testing
Purging of headspace prior to injection	• ≥90% of participants successfully primed the needle in usability testing
Reliable and precise delivery of dose	• The pen injector passed all dose accuracy tests at a variety of temperatures and after a variety of free-fall drop conditions
	• ≥97% of participants successfully administered exenatide once weekly in usability testing
Visual, auditory, and tactile indicators	• Colored directional bands indicate the motion of twisting during mixing and priming steps and provide a visual cue that the twisting motion has been completed
	• An audible click signals that the required twisting motion for the mixing and priming steps has been completed
	• A large inspection window allows the patient to visually assess the quality of the mixture prior to injection
	• An injection button will pop out once air in the chamber holding the suspension has been purged
	• An audible click is heard and a hard stop is detected through the dose delivery button after the dosing button has been fully depressed

Central to the successful pharmacologic management of T2DM is the recognition that available treatments come with inherent barriers and safety concerns that may impede successful use in patients. Injectable medications are no exception and medical devices should mitigate these problems. Dosing errors with injectable agents such as insulin or exenatide once weekly can result in inadequate glycemic control or hypoglycemia in the event of overdose.^17,18 Pen injectors have helped greatly to minimize dosing errors compared with traditional vial and syringe delivery where studies of insulin administration have shown patient dosing inaccuracies to be as high as 19% depending on age.^19,20 However, multidose pen injectors could introduce dosing errors if the dose is incorrectly dialed or if the dose display is misread.²¹ The single-use dual-chamber pen allows for the delivery of only 1 dose with no titration and mitigates concerns of overdose.

Under a variety of testing conditions, the dual-chamber pen consistently delivered an accurate dose within the allowed 10% of the expected total volume (650 µL). However, patient failure to hold the dual-chamber pen in place after the dose button is fully depressed, as recommended in the approved instructions for use, could possibly result in suboptimal dosing.⁴ Torque requirements for twisting the pen knob and force requirements for depressing the dose delivery button were below the maximum allowed for acceptable use. This is important because some patients with T2DM could be elderly or have a preexisting impairment or comorbidity resulting in reduced motor skills.

In usability testing, completion rates of required setup, dose preparation and mixing, and injections steps were very high in all 3 participant groups (≥87% to 94%). The results are reflective of first-time use with the dual-chamber pen and indicate that the device contains the necessary design features to ensure that a majority of patients will receive their treatment as intended starting with the very first dose. It would be expected that successful use would increase over time as familiarity and experience with the dual-chamber pen increases. Nonetheless, as with any injectable medication, HCPs should periodically reevaluate patient injection technique and provide corrective education as needed to ensure continued proper use.²²

Ease-of-use scores for the dual-chamber pen were high in all testing groups (5.8–6.5 out of 7), indicating that the device was considered to be easy to use overall. The ease-of-use of the dual-chamber pen may potentially increase adherence to treatment regimens.²³ A common barrier to the initiation of injectable therapy in patients with T2DM is the perception that traditional needle and syringe delivery may be too difficult to negotiate or handle.¹² The final injection pen design automates dose preparation and delivery to make administration of exenatide once weekly easier for continued use.

Conclusions

The final dual-chamber pen design, which was developed through a process of comprehensive risk analysis and human factors validation testing in accordance with the latest FDA guidance, minimizes potential issues such as product mix-up, incorrect needle attachment, and problems purging air from the pen, and provides a more convenient option than the single-dose tray for patients with T2DM who administer exenatide once weekly containing microspheres to control their hyperglycemia.