Journal of Diabetes Science and Technology
`Volume 4, Issue 6, November 2010
`© Diabetes Technology Society
`
`ORIGINAL ARTICLES
`
`Dose Accuracy and Injection Force of Disposable Pens Delivering
`Pramlintide for the Treatment of Diabetes
`
`Douglas Merritt, B.S.M.E.,1 Barb Schreiner, Ph.D., R.N., C.D.E., B.C.-A.D.M.,1 Scott Harris, M.S.M.E.,1
`Mary Beth DeYoung, Ph.D.,1 Susan Strobel, Ph.D.,2 and Joseph Lauinger, M.T.1
`
`Abstract
`
`Background:
`The pen injection format, typically used for insulin administration, has been adapted for the injectable,
`noninsulin diabetes therapy pramlintide. Administered before major meals, pramlintide therapy requires two
`to four injections/day in addition to the patients' usual insulin injections. The dose accuracy and injection
`force was determined for the 60 and 120 µg pramlintide pens.
`
`Methods:
`Dose accuracy testing was conducted at two sites on multiple 60 µg (15, 30, and 60 µg doses) and 120 µg pens
`(60 and 120 µg doses) at prespecified temperatures (5-40 °C) and humidities (0-75%) using 29 G half-inch needles.
`All pens were stabilized under testing conditions for 4 h prior to testing. One site used a compression load cell
`(Zwick device) to test pens; one site performed tests manually.
`
`Injection-force testing was conducted at one site on multiple 60 and 120 µg pens at multiple temperatures
`(18-28 °C) and humidities (25-75%) using 29 and 31 G half-inch needles and an injection speed of 150 m/min.
`Injection-force testing was performed using a Zwick device.
`
`Results:
`Dose accuracy for all pens tested, regardless of location, reproducibly met/exceeded acceptance criteria.
`Mean percentage of dose accuracy was 96.04 to 102.45% [standard deviations (SDs) 0.3 to 1.4 µg] for the 60 µg pen
`and 98.16 to 101.83% (SDs 0.4 to 2.5 µg) for the 120 µg pen. The average injection force across both pens did not
`exceed 7 N regardless of needle size.
`
`Conclusions:
`High dose accuracy and low injection force were observed for the 60 and 120 µg pens under a variety of conditions.
`
`J Diabetes Sci Technol 2010;4(6):1438-1446
`
`Author Affiliations: 1Amylin Pharmaceuticals, Inc., San Diego, California; and 2SciComm LLC, San Diego, California
`
`Abbreviations: (ISO) International Organization for Standardization, (RH) relative humidity, (SD) standard deviation
`
`Keywords: accuracy, hyperglycemia, pramlintide, reproducibility, type 1 diabetes, type 2 diabetes
`
`Corresponding Author: Douglas Merritt, B.S.M.E., Amylin Pharmaceuticals, Inc., 9360 Towne Centre Dr., San Diego, CA 92121; email address
`douglas.merritt@amylin.com
`
`1438
`
`Sanofi Exhibit 2135.001
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`Introduction
`
`The first pen injection device was launched in 1989
`
`for the administration of insulin to patients with type 1
`or advanced type 2 diabetes. Since then, pens have
`continued to gain in popularity, primarily outside the
`United States; however, pen use in the United States
`now appears to be increasing.I
`
`Over the years, pens have been refined to improve
`accuracy and ease of use for self-injection. Pen properties
`associated with improved ease of use include portability and
`ease of reading, ease of adjustment, ergonomic design,
`sturdiness, and differentiation between multiple pens
`for patients using more than one injectable medication.I
`Some pens have additional safety features such as
`insufficient remaining dose stops that prevent users
`from receiving incomplete doses2 and audible or tactile
`feedback to ensure complete dose delivery.I Ease of use
`is an important device parameter for the patient who
`must inject medication multiple times per day indefinitely
`and may improve medication adherence.I One aspect
`of "ease of use" is how much force is required by the
`patient to inject medication. Low injection force pens
`allow patients to inject the full dose by applying a low
`and steady pressure without experiencing hand or grip
`fatigue.I By increasing dosing accuracy, the pen injectors
`developed for insulin have helped reduce the incidence
`of hypoglycemia and hyperglycemia.I
`
`Continuing refinement of the pen has contributed to
`growing acceptance by patients. A survey of studies
`between 1980 and 2008 assessing patient-reported outcomes
`of pen versus vial and syringe showed substantial
`preferences for the pen.3 Of 29 studies assessing patient
`preference, >66% of patients in 28 studies preferred the
`pen; in 8 of 9 studies assessing pain, >50% of patients
`said the pen caused less pain; and in 10 of 12 studies
`assessing acceptability, >75% of patients rated the pen as
`more acceptable than vial and syringe.3 The pen apparatus
`may be disposable or refillable.4 Disposable pens are
`used until the drug cartridge is empty, and then the
`entire unit is disposed. Durable (refillable) pens are
`used repeatedly, and spent drug cartridges are replaced
`with new ones as needed. Luijf and DeVries surveyed
`the literature (1998 to 2009) relative to pen accuracy
`and preference and found that pen devices had greater
`accuracy than a vial and syringe device, especially at
`low doses (<5 IU).5
`
`Pramlintide is an injectable therapy for patients with
`type 1 and type 2 diabetes who use mealtime insulin
`and who are not achieving desired glycemic goals
`despite optimized
`insulin therapy.6 The usual dose
`for patients with type 1 and type 2 diabetes is 60 and
`120 µg, respectively, although smaller doses of 45, 30,
`or 15 µg (type 1 patients) and 60 µg (type 2 patients)
`may be used. Pramlintide is an analog of the naturally
`occurring hormone amylin, which is cosecreted from
`the pancreatic ~ cells with insulin in response to a meal.
`Patients with diabetes are deficient in both insulin and
`amylin. Pramlintide, like amylin, slows gastric emptying,
`suppresses inappropriate postprandial glucagon secretion,
`and increases satiety. These effects complement those of
`insulin to help circulating glucose concentrations stay
`within the normal range, thus avoiding hyperglycemia
`and, over the long-term, lowering hemoglobin Ale.
`
`In patients with type 1 or type 2 diabetes, pramlintide
`is injected just before a meal, which is defined as food
`intake containing 2250 kcal or 230 g of carbohydrate.6
`Unlike insulin, pramlintide dosing does not vary based
`on body weight or the caloric/carbohydrate content of a
`meal beyond the 2250 kcal/230 g carbohydrate threshold.
`Patients using pramlintide require two to four injections
`per day (based on eating habits) in addition to their
`injections of rapid-acting insulin. Thus, a drug delivery
`system (pen) that increases ease of use and accuracy may
`be beneficial to patients.
`
`This article describes the design features, dose accuracy,
`and dose injection force of the pramlintide pen delivery
`system.
`
`Methods
`
`Design Features
`Two sets of dosing options are available for the pramlintide
`pen. The 60 µg pen delivers fixed doses of 15, 30, 45, and
`60 µg; the 120 µg pen delivers fixed doses of 60 and
`120 µg. Adjustable dosing allows patients to use less
`pramlintide if they better tolerate lower than higher
`doses or require gradual titration to improve tolerability.
`The fixed dose design differs from that used in insulin
`pens, which are designed to deliver a continuous drug dose
`across a wide range. Both pramlintide pens allow use of
`29, 30, or 31 G type A needles, and the pens may be used
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1439
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.002
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`for up to 30 days, depending on dose size and injection
`frequency. The filled pen is stable between 2 and 42 °C,
`and patients are instructed to store the pens at 2-8 °C before
`first use, after which the pens may be stored from 2-8 °C
`up to 30 °C. Pens have a maximum diameter of 18.9 mm
`(~0.75 in.) and, when capped, a length of 158 mm (6.2 in.;
`Figure 1; data on file, Amylin Pharmaceuticals, Inc.).
`Three key design features help ensure a high level of
`dosing accuracy. A stroke sleeve moves the toothed rod
`forward by a precise distance when the dosage knob
`is depressed, and a guiding sleeve with interlocking
`teeth prevents the toothed rod from traveling backward
`(data on file, Amylin Pharmaceuticals, Inc.).
`
`Dose Accuracy Protocol
`Dose accuracy studies were performed at two different
`sites
`[Amylin Pharmaceuticals, Inc, San Diego, CA
`(site A) and Ypsomed AG, Burgdorf, Switzerland (site
`B)]. Both sites used
`International Organization for
`Standardization (ISO) guideline 11608-1, developed to
`test the "performance requirements regarding essential
`aspects" of pen injection devices? The ISO
`testing
`criteria provides guidance on preconditioning and
`testing conditions, including temperature and humidity
`as well as dose accuracy requirements and statistical
`considerations. The preconditioning/testing conditions
`used in these experiments were within the specifications
`noted in ISO 11608-1. Pens were prepared for testing by
`attaching a new needle and performing the pen setup
`described
`in patient instructions (needle attachment
`and depression of pen dose knob until the expulsion of
`fluid was seen from the needle). At site A, each pen was
`placed in a Zwick device (Zwick/Roell Z2.5 tester with
`compression load cell; Ulm, Germany), and doses were
`expelled until the pen was completely emptied. A Zwick
`device measures the compression or tension force on a
`sample. The mass of expelled samples was determined
`using a Mettler SAG285 scale (0.01 mg resolution;
`
`Toledo, OH). Testing was conducted at 25 to 26 °C, with
`a relative humidity (RH) of 36% to 39%, using 29 G
`half-inch needles of a single lot. Five 60 µg pens were
`tested using the 30 and 60 µg dose, with 42 and 21 doses
`expelled/pen,
`respectively;
`five 120 µg pens were
`tested using the 60 and 120 µg dose, with 42 and 21
`doses expelled/pen,
`respectively. Site B conducted
`dose accuracy testing manually under three different
`conditions: cold (5 ± 3 °C, no RH), standard (18 to
`25 °C, 25-75% RH), and hot (40 ± 2 °C, 50 ± 10% RH)
`using 29 G half-inch needles of a single lot. Fifteen
`60 µg pens were tested using the 15, 30, and 60 µg doses,
`with four doses expelled/pen; fifteen 120 µg pens were
`tested using the 60 and 120 µg doses, with four doses
`expelled/pen. At both sites, pens were stabilized under
`testing conditions for 4 h prior to testing. Data from the
`first (priming) dose of each pen were discarded from
`calculations. The conversion from mass to volume was
`vmeasured = Gmeasured/p, where vmeasured is the volumetric
`measurement value for a given dose; Gmeasured is the
`gravimetric measurements expressed in grams for a
`given dose; and p is density, expressed in grams per
`milliliter, 1.015 g/ml for pramlintide injection in cartridge.
`Acceptable dosing accuracy was defined as meeting or
`exceeding acceptance criteria described in section 9.2 of
`ISO 11608-1:2000(E).7
`
`Injection Force Protocol
`Injection force
`tests were conducted at site A only.
`Pens were prepared for
`testing by attaching a new
`needle and performing the pen setup described in
`patient instructions (needle attachment and depression
`of pen dose knob until the expulsion of fluid was seen
`from the needle). Testing was performed using 10 pens of
`both pen sizes using a Zwick device and was conducted
`at 18 to 28 °C and at 25 to 75% RH. Both 60 and 120 µg
`pens were tested using a 29 and 31 G half-inch needle.
`A standard injection speed of 150 m/min was used
`
`Pen cap
`
`Threads for
`needle attachment
`
`Pen label
`
`Dial
`
`Knob
`
`Cartridge
`
`Window
`
`Length with cap = 158 mm
`
`Figure 1. Pramlintide pen injection device.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1440
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.003
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`for all tests, as it mimicked a typical injection time of
`approximately 2 s and made determination of peak injection
`force easier. Data from the first (priming) dose of each pen
`were discarded from calculations. Data for each pen/
`needle configuration consisted of three sets of two force
`tests. Each set of two force tests was preceded by one
`unmeasured force test and separated from the other
`force tests by six manual expulsions of drug so that data
`were collected from the beginning, middle, and end of
`the pen capacity. There are no established criteria for
`force testing.
`
`Results
`
`Dose Accuracy Results
`Both 60 and 120 µg dose pens delivered accurate and
`reproducible doses at temperatures ranging from 5 to
`40 °C and RH ranging from O to 75%. Different sites,
`testing protocols, operators, and dose sizes produced
`consistent dose accuracy results. All pens met or exceeded
`the ISO 11608-1:2000(E) criteria.7
`
`At site A, the 60 µg pen delivered a 30.0 to 32.0 µg dose
`[overall mean percentage of dose accuracy 102.45%;
`standard deviation (SD) range 0.3 to 0.4] when the 30 µg
`dose was injected and delivered a 60.3 to 60.8 µg dose
`(overall mean percentage of dose accuracy 101.19%;
`SD range 0.4 to 0.8) when the 60 µg dose was injected.
`The 120 µg pen delivered a 59.5 to 61.3 µg dose (overall
`mean percentage of dose accuracy 100.41%; SD range
`0.4 to 2.4) when the 60 µg dose was injected and delivered
`a 118.9 to 120.7 µg dose (overall percentage of dose
`accuracy 99.84%; SD range 0.4 to 1.5) when the 120 µg
`dose was injected (Figure 2). The variability of pen 4
`(120 µg pen delivering a 60 µg dose) was greater than
`any other pen tested; however, it still met all pen release
`criteria (data not shown) as well as the ISO-11608-1
`acceptance criteria.7
`
`At site B, the 60 µg pen delivered a 14.4 to 14.9 µg dose
`(overall percentage of dose accuracy across all temperatures
`98.34%; SD range 0.4 to 0.64) when the 15 µg dose
`was injected, a 28.9 to 29.6 µg dose (overall percentage
`of dose accuracy across all temperatures 97.96%; SD
`range 0.8 to 1.4) when the 30 µg dose was injected,
`and a 58.9 to 59.6 µg dose (overall percentage of dose
`accuracy across all
`temperatures 98.81%; SD
`range
`0.6 to 0.7) when the 60 µg dose was injected. The 120 µg
`pen delivered a 58.9 to 61.1 µg dose (overall percentage
`of dose accuracy 100.26% across all
`temperatures;
`SD range 0.8 to 2.5) when the 60 µg dose was injected
`and delivered a 118.3 to 120.6 µg dose (percentage of
`
`dose accuracy 99.75% across all temperatures; SD range
`1.1 to 2.3) when the 120 µg dose was injected (Figure 3).
`
`Injection Force Results
`The injection force values of all pens were similar
`regardless of the dose/needle combination used (Figure 4).
`The average injection force for the 60 µg pen (60 µg dose,
`29 G needle) ranged from 4.1 to 6.6 N (SD range 0.2 to 0.3),
`with a maximum force ranging from 6.3 to 10.1 N; the
`average injection force for the 60 µg pen (60 µg dose,
`31 G needle) ranged from 4.5 to 6.4 N (SD range 0.1 to
`0.7), with a maximum force ranging from 6.8 to 9.5 N.
`The average injection force for the 120 µg pen (120 µg dose,
`29 G needle) ranged from 3.2 to 6.5 N (SD range 0.2 to 0.6),
`with a maximum force ranging from 5.0 to 10.0 N; the
`average injection force for the 120 µg pen (120 µg dose,
`31 G needle) ranged from 2.8 to 6.5 N (SD range 0.1 to 0.7),
`with a maximum force ranging from 4.3 to 9.6 N.
`
`Discussion
`
`The pramlintide pen device delivers a fixed dose of
`pramlintide for use as indicated by patients with type 1
`or type 2 diabetes who use mealtime insulin. Its design
`is based on a general insulin pen template and incorporates
`several engineering elements to increase accuracy. These
`studies showed that patients who use the pen as directed
`will consistently receive the prescribed dose of pramlintide
`with accuracy.
`
`The dose accuracy of the pramlintide pen across multiple
`sites,
`temperatures, RH, and operators, at both the
`minimum and maximum doses, was highly reproducible.
`Pen design, specifically the guiding sleeve, stroke sleeve,
`and the toothed rod, was viewed as key in producing
`accurate results across a variety of conditions and operators.
`The stroke sleeve moves the toothed rod forward by a
`precise distance when the dosage knob is depressed, and a
`guiding sleeve with interlocking teeth prevents the toothed
`rod from traveling backward. The dose accuracy was
`well within the ISO specifications set for pen injectors.
`
`The average injection force for both configurations of the
`pramlintide pen was low, <7 N, and did not appear to
`be affected by needle size. The injection forces for the
`pramlintide pen were substantially lower than those
`observed for insulin pens that have reported injection
`forces ranging from approximately 10 to 25 N.1 In a study
`comparing usability and patient preference for different
`pen injectors, patients preferred the pen with the lowest
`injection force. 8 Given that patients with diabetes will
`routinely
`inject medication multiple
`times per day
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1441
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.004
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`indefinitely, one could safely assume that a device that
`accurately delivers medication with a low injection force
`would be preferred.
`
`The strengths of these testing protocols included testing
`the indicated doses under multiple conditions for the
`entire pen volume under highly controlled conditions.
`The primary weakness for injection-force testing was
`that only one injection speed was used. Greater dosing
`variability is expected with patient use based on the range
`of patient understanding regarding the injection process
`and the ability to accurately execute an injection.
`
`The first insulin pen was launched in 1985, and currently,
`pens account for just over 50% of all insulin injectables
`worldwide.1 The popularity of the pen compared to
`vial and syringe has grown because of increased dose
`accuracy, ease of use, and decreased cost. Most, if not
`all, pen development was done with insulin where
`the critical issue was dose accuracy so as to avoid
`hypoglycemia and hyperglycemia. 2 Numerous insulin
`studies have shown pens to be more accurate than vial
`and syringe, especially at doses <5 U.9-12 Ease of use for
`pens encompasses multiple issues from the practical/
`ergonomic to the psychological, including ease of dose
`
`Pramlintide 60 µg pen
`
`A
`
`35
`
`30 µg dose
`
`-·
`
`.
`
`30 "l~DMOSMAWIAAAtmm
`
`B
`
`65
`
`60
`
`55
`
`ci 50
`..=,
`45
`Q)
`Ill
`0
`C
`
`40
`
`35
`
`30
`
`60 µg dose
`
`ij 8 s;syaaeY~gigvyggggg
`
`■
`
`mean ± SD (mean% dose accuracy)
`• Pen 1: 60.6 ± 0.7 (100.99%)
`■ Pen 2: 60.3 ± 0.8 (100.44%)
`"' Pen 3: 60.4 ± 0.5 (100.65%)
`0 Pen 4: 60.8 ± 0.6 (101 .27%)
`V Pen 5: 60. 7 ± 0.4 (101 .13%)
`
`1
`
`4
`
`16
`10
`13
`7
`Injection number
`
`19
`
`22
`
`25
`
`ci 20
`..=,
`15
`Q)
`Ill
`0
`C
`
`10
`
`5
`
`0
`
`1
`
`mean± SD (mean% dose accuracy)
`
`• Pen 1: 32.0 ± 0.4 (106.51%)
`■ Pen 2: 30.6 ± 0.4 (102.09%)
`"' Pen 3: 30.3 ± 0.3 (101.09%)
`0 Pen 4: 30.0 ± 0.3 (99.92%)
`V Pen 5: 30.8 ± 0.3 (102.47%)
`
`11
`
`31
`21
`Injection number
`
`41
`
`Pramlintide 120 µg pen
`
`120 µg dose
`
`C
`
`ci
`..=,
`Q)
`Ill
`0
`C
`
`65
`
`60
`
`55
`
`50
`
`45
`
`40
`
`35
`
`30
`
`60 µg dose
`
`o o O
`0 00 =o0
`~~~~~.\,~~
`
`0
`mean ± SD (mean % dose accuracy)
`
`• Pen 1: 59.5 ± 0.5 (99.25%)
`■ Pen 2: 60.1 ± 0.6 (100.23%)
`"' Pen 3: 60.6 ± 0.8 (100.99%)
`0 Pen 4: 61 .3 ± 2.4 (102.12%)
`V Pen 5: 59.7 ± 0.4 (99.46%)
`
`1
`
`11
`
`21
`
`31
`
`41
`
`Injection number
`
`D
`
`125
`
`120
`
`115
`ci
`..=, 110
`
`Q)
`Ill 105
`0
`C
`
`100
`
`95
`
`90
`
`! •
`
`•
`
`•
`
`!
`;e~~~~ g8 e8e8!! geeeee
`• •
`•
`mean± SD (mean% dose accuracy)
`•
`Pen 1: 120.2 ± 1.5 (100.18%)
`■ Pen 2: 120.1 ± 0.6 (100.05%)
`"' Pen 3: 120.7 ± 1.2 (100.61%)
`0 Pen 4: 118.9 ± 0.4 (99.06%)
`V Pen 5: 119.1 ± 0.5 (99.28%)
`
`1
`
`4
`
`7
`
`10
`
`13
`
`16
`
`19
`
`22
`
`Injection number
`
`Figure 2. Dose accuracy, site A. Dose accuracy (mean dose ± SD, percentage dose accuracy) of the 60 µg pen to dispense (A, B) a 30 and 60 µg
`dose and of the 120 µg pen to dispense (C, D) a 60 and 120 µg dose using multiple pens. Temperature (25 to 26 °C), RH (36% to 39%), needle size
`(29 G), lot number, and operator were identical for all pens.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1442
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.005
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`dialing/correction, large print dose displays, low injection
`force, portability, easy cartridge change, dose confirmation,
`and visual/tactile differentiation of different pens. The pen
`format also helps ease some of the stress encountered
`when the patient must adjust an established oral therapy
`to
`include an
`injectable
`therapy.I Medications and
`associated devices can also represent a substantial cost to
`
`the patient, especially when treating a chronic long-term
`disease. Cost analyses have shown that pen use reduced
`all-cause annual
`treatment costs by ~$1590/patient.I
`Pen use with insulin has been shown to reduce the
`incidence of hypoglycemia/hyperglycemia, which has not
`only substantial health ramifications, but also potentially
`significant associated costs.I A review of patient-reported
`
`Pramlintide 60 µg pen
`
`Pramlintide 120 µg pen
`
`A
`
`cl
`.=,
`Q) ,,,
`0
`C
`
`B
`
`cl
`.=,
`Q) ,,,
`0
`C
`
`C
`
`cl
`.=,
`Q) ,,,
`0
`C
`
`5°C
`
`70r
`•,,,_.,..,.,,.,,,__,_,JI'-,.---.--,, 59.6 ± 0.7 (99.38%)
`
`•
`
`60
`
`mean± SD
`(mean % dose accuracy)
`
`50
`35
`
`25
`
`~~~~l;,;l~~..,~v-¢w'<lillv 29.6 ± 1.4 (98.56%)
`
`V
`15~Qsxxtl:l-~-~~-b:l:tl:~~ 14.9 ± 0.6 (99.79%)
`
`5~-....-----.----.------.--..----r---
`10
`30
`40
`50
`60
`20
`Injection number
`
`65f
`
`Room temperature
`
`• •
`•
`•
`·-·-·--.-· ...... -·,,.. - .......... . , 59.4 ± 0.6 (98.97%)
`
`mean ±SD
`(mean % dose accuracy)
`
`60
`
`5°C
`
`1
`v mean ± SD
`~}
`v
`v
`v
`v
`(mean % dose accuracy)
`,flw?""v-~v./'wJivvv vv711..v.Pv0ii;,vv..,v,, .. Y-v=-
`(
`o,)
`'17
`·--v 120.2±2.3 100.18 ,o
`VVWT
`v--.,,v-
`
`120
`
`v
`
`110
`70
`
`D
`
`---~
`..._
`
`Q)
`
`:g
`C
`
`so~----.------~--~-
`
`10
`
`~ ~ 40
`50
`Injection number
`
`50
`
`E
`
`Room temperature
`
`55
`
`35~
`29 6 • 09 {98'9%)
`;:;:
`:
`
`~ 149±04(9919%)
`
`10
`
`20
`30
`50
`40
`Injection number
`
`60
`
`so~ - - - - . - - - - - -~ - -~ -
`10
`50
`50
`~ ~ 40
`Injection number
`
`40 °C
`
`F
`
`40 °C
`
`] . . ..
`. .
`55}._._, .. ._ ,,.,,,,,,,,_ ... _,,. ... ✓.----.-... --- 58.9 ± 0.7 (98.08%)
`
`mean± SD
`(mean % dose accuracy)
`
`~b~""~ 289±08 (9642%)
`
`1~~
`mean ±SD
`(mean % dose accuracy)
`cl 1
`~ v,v.,,;~~~v.,,pvv,;~~ 118.3±1.1 (9816%)
`.=,
`
`110
`70
`
`Q)
`Ill
`0
`C
`
`0
`15~
`~
`10
`ro
`
`I
`I
`I
`so
`~ ~ 40
`Injection number
`
`I
`
`144±06 (9604%)
`
`I
`50
`
`so~ - - - - . - - - - - -~ - -~ -
`60
`50
`10
`~ ~ 40
`Injection number
`
`Figure 3. Dose accuracy, site B. Dose accuracy (mean dose ± SD, percentage dose accuracy) of the 60 µg pen to dispense a 15, 30, and 60 µg dose
`at (A) 5 °C and 0 RH, (B) room temperature and 25-27% RH, and (C) 40 °C and 50 ± 10% RH Dose accuracy of the 120 µg pen (mean dose ± SD,
`percentage dose accuracy) to dispense a 60 and 120 µg dose at (D) 5 °C and 0 RH, (E) room temperature and 25-27% RH, and (F) 40 °C and
`50 ± 10% RH Testing personnel, needle size (29 G), and lot number were identical for all pens.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1443
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.006
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`outcomes for pen versus vial and syringe studies
`between 1980 and 2008 showed highly favorable results
`for the pen. Improvement in areas such as reduced pain
`with injections, treatment satisfaction, and treatment
`convenience for the pen versus vial and syringe ranged
`from 50 to 100%.3 Improved patient outcomes can help
`patient adherence, which is of paramount importance in
`
`achieving and maintaining optimal health within disease
`parameters, and helps decrease overall costs.U3
`
`Despite the positive attributes of the pen, use of the vial
`and syringe format for delivery of injectable diabetes
`medications is still predominant in the United States.
`This is surprising given that the vial and syringe format
`
`Pramlintide 60 µg pen : 60 µg dose
`
`A
`
`29 G needle
`
`12
`
`9
`
`6
`
`3
`
`-
`
`-
`
`-
`
`~
`
`G)
`
`(,) .. 0 ....
`
`C
`0
`+l
`(,)
`
`G) :s
`
`B
`
`12
`
`31 G needle
`
`~
`
`G)
`
`(,) .. 0 ....
`
`C
`0
`+l
`(,)
`
`G) :s
`
`9
`
`6
`
`3
`
`-
`
`o_.__ ___ __. __ L...-__ ____ _____ ___ ___ _ __ . ____
`
`o_.__ ___ __. __ L...-__________________ _ _ . ____
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`Pen number
`
`Pen number
`
`Pramlintide 120 µg pen : 120 µg dose
`
`29 G needle
`
`D
`
`31 G needle
`
`C
`
`12
`
`12
`
`~ 9
`
`G)
`
`(,) .. 0 ....
`
`C
`0
`+l
`(,)
`
`G) :s
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`6
`
`3
`
`0
`
`r i
`
`~ 9
`
`G)
`
`(,) ..
`~ 6
`~ G) :s
`
`0
`
`3
`
`0
`
`-
`
`-
`
`-
`
`-
`
`-
`
`i
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`Pen number
`
`Pen number
`
`D Average force
`
`• Maximum force
`
`Figure 4. Injection force rneasurernents, site A. Injection force required for the 60 µg pen to dispense a 60 µg dose using (A) a 29 G needle and
`(B) a 31 G needle and the 120 µg pen to dispense a 120 µg dose using (C) a 29 G needle and (D) a 31 G needle. Injection velocity (150 rn/rnin),
`temperature (18 to 28 °C), RH (25% to 75%), and operator were identical for all pens.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1444
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.007
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`is associated with poor dose accuracy, fear of injections,14,15
`inconvenience,16
`lack of social acceptance,17 and poor
`accuracy when self-mixing
`insulin.18 These negative
`attributes often result in a profound psychological resistance
`in health care providers and patients to use injectables
`such as insulin.
`
`The Food and Drug Administration Human Factors
`Guidelines for Medical Devices recognizes that errors
`associated with the incorrect use of medical devices can
`pose a hazardous situation for the patient and potentially
`others.19 Medical devices are often complex and may be
`used under stressful conditions. These devices are also
`used by patients whose comprehension of the disease/
`device may vary widely.
`
`The design of the pramlintide pen addresses some
`important human factors issues by combining many
`features seen in insulin pens such as ease of use, ergonomic
`design, color coding, and convenience with fixed dosing
`and improved dose accuracy while maintaining a low
`injection force. While dosing approaches differ between
`pramlintide and insulin (no insulin-like titration is required
`for pramlintide), all pen devices are required to adhere
`to a standard of accuracy. Improved accuracy as well
`as a low injection force requirement may alleviate some of
`the anxiety associated with introducing a new therapy(cid:173)
`for health care providers and patients alike.
`
`Conclusion
`
`A high degree of dose accuracy and low injection force
`make the pramlintide pen an attractive alternative to
`vial and syringe when administering pramlintide.
`
`Funding:
`
`This study was funded in full by Amylin Pharmaceuticals, Inc.
`
`Disclosures:
`
`All Amylin-affiliated authors own stock in Amylin Pharmaceuticals, Inc.
`
`Acknowledgments:
`
`Ypsomed AG, Burgdorf, Switzerland, developed and supplied the pen
`device as well as contributed to the experimental data described in
`this article.
`
`References:
`
`1. Clarke A, Spollett G. Dose accuracy and injection force dynamics
`of a novel disposable insulin pen. Expert Opin Drug Deliv.
`2007;4(2): 165-7 4.
`
`2. Klonoff DC. The pen is mightier than the needle (and syringe).
`Diabetes Technol Ther. 2001;3(4):631-3.
`
`3. Molife C, Lee LJ, Shi L, Sawhney M, Lenox SM. Assessment of
`patient-reported outcomes of insulin pen devices versus conventional
`vial and syringe. Diabetes Technol Ther. 2009;11(8):529-38.
`
`4. Diabetes Health. Product reference guide 2010. http:Uwww.diabeteshealth.
`comlmedia/pdfs/PRG2010/2-Insulin Pens Chart-Diabetes Health 2010.pdf.
`Accessed July 13, 2010.
`
`5. Luijf YM, DeVries JH. Dosing accuracy of insulin pens versus
`conventional
`syringes
`and vials. Diabetes Technol Ther.
`2010;12 Suppl l:S73-7.
`
`6. SYMLIN (pramlintide acetate) injection. Prescribing Information.
`Amylin Pharmaceuticals Inc., San Diego, CA. Revised July 2008.
`https:llwww .s yml in. com/pdf/S YM LIN-pi- combined .pdf. Accessed
`September 8, 2010.
`
`for Standardization. Pen-injectors
`7. International Organization
`for medical use-Part 1: pen-injectors-requirements and test
`methods. ISO 11608-1:2000(E). http:Uwww.iso.org/isolcatalogue detail.
`htm?csnumber=19545. Accessed September 8, 2010.
`
`8. Haak T, Edelman S, Walter C, Lecointre B, Spollett G. Comparison
`of usability and patient preference for the new disposable insulin
`device SoloStar versus FlexPen, Lilly disposable pen, and a
`prototype pen: an open-label study. Clin Ther. 2007;29(4):650-60.
`
`9. Gnanalingham MG, Newland P, Smith CP. Accuracy and reprodu(cid:173)
`cibility of low dose insulin administration using pen-injectors and
`syringes. Arch Dis Child. 1998;79(1):59-62.
`
`10. Lteif AN, Schwenk WF. Accuracy of pen injectors versus insulin
`syringes
`in children with
`type 1 diabetes. Diabetes Care.
`1999;22(1): 137-40.
`
`11. Jehle PM, Micheler C, Jehle DR, Breitig D, Boehm BO. Inadequate
`suspension of neutral protarnine Hagendorn (NPH) insulin in
`pens. Lancet. 1999(9190);354:1604-7.
`
`12. Pearson TL. Practical aspects of insulin pen devices. J Diabetes Sci
`Technol. 2010;4(3):522-31.
`
`13. Selam JL. Evolution of diabetes insulin delivery devices. J Diabetes
`Sci Technol. 2010;4(3):505-13.
`
`14. [Fear of the injection must not be an argument. Every second
`type 2 diabetic patient needs insulin.] MMW Fortschr Med.
`2002;144(49):60.
`
`15. Wagner J, Malchoff C, Abbott G. Invasiveness as a barrier to self(cid:173)
`monitoring of blood glucose in diabetes. Diabetes Technol Ther.
`2005;7(4):612-9.
`
`16. Keith K, Nicholson D, Rogers D. Accuracy and precision of low(cid:173)
`dose insulin administration using syringes, pen injectors, and a
`pump. Clin Pediatr (Phila). 2004;43(1):69-74.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1445
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.008
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Dose Accuracy and Injection Force of Disposable Pens Delivering Pramlintide for the Treatment of Diabetes
`
`Merritt
`
`17. Summers KH, Szeinbach SL, Lenox SM. Preference for insulin
`delivery systems among current insulin users and nonusers. Clin
`Ther. 2004;26(9):1498-505.
`
`18. Bell DS, Oements RS Jr, Perentesis G, Roddam R, Wagenknecht L.
`Dosage accuracy of self-mixed vs premixed insulin. Arch Intern
`Med. 1991;151(11):2265-9.
`
`19. U.S. Department of Health and Human Services, U.S. Food and
`Drug Administration. About human factors. http:Uwww.fda.gov/
`MedicalDevices!DeviceRegulationandGuidance!PostmarketRequirementsl
`HumanFactorslucm119185.htm. Accessed February 4, 2010.
`
`J Diabetes Sci Technol Vol 4, Issue 6, November 2010
`
`1446
`
`www. iournalofdst. orq
`
`Sanofi Exhibit 2135.009
`Mylan v. Sanofi
`IPR2018-01676
`
`