Clinical Therapeutics/Volume 29, Number 4, 2007
`
`Sanofi et al. v. Merck
`1 :16-cv-812-RGA
`
`PTX-0665
`
`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
`
`Thomas Haak, MD 1; Steven Edelman, MD2; Claudia Walter, CDE3; Brigitte Lecointre, CDE4;
`and Gerry Spollett, MSN 5
`
`1 Diabetes Centre Mergentheim, Bad Mergentheim, Germany; 2 University of California San Diego Veterans
`Affairs Medical Center, San Diego, California; 3Diabetes-Forum, Nurnberg, Germany; 4 Reseau Resdiab,
`Nice, France; and 5Yale University School of Medicine, New Haven, Connecticut
`
`ABSTRACT
`Background: Patients with diabetes have been found
`to have a preference for insulin pens over a vial and
`syringe since these devices offer improvements in com(cid:173)
`pliance, freedom, and flexibility.
`Objective: This study assessed the usability, specific
`pen features, and patient preference for 4 prefilled, dis(cid:173)
`posable, insulin pens: SoloStar®, Humulin®/Humalog®
`pen (Lilly pen), FlexPen®, and a fourth, prototype
`pen, Pen X, in patients with type 1 or 2 diabetes. In
`1-hour interviews, patients carried out simulated use
`(preparing the pens, setting a dose, and injecting into
`a receptacle, not the body) under observation, and
`answered qualitative and quantitative questions. Pa(cid:173)
`tients were supplied with the relevant user manual. The
`usability (ability and time taken to carry out handling
`tasks) and preference (based on 14 key pen features
`and overall preference) of each pen were assessed with(cid:173)
`out blinding for pen make/manufacturer. During the
`interviews, the patients prepared each pen and per(cid:173)
`formed injections into a receptacle. Comparisons were
`made between the pens at every step. Subgroup analyses
`of the usability exercises were carried out based on age
`(11-15 years; ;:::,:60 years), previous pen experience, and
`disability ( visual and dexterity).
`Results: In total, 510 diabetes patients (65% type 2
`diabetes; 51 % female; mean age, 43 years [range,
`11-82 years]) from 4 countries (United States, Germany,
`France, and Japan) completed the study. Overall, a
`greater proportion of patients correctly prepared the
`pen and performed an injection into a receptacle with
`SoloStar versus all comparator pens (P < 0.05). Simi(cid:173)
`lar findings were observed in the usability subgroup
`analyses based on age, previous pen experience, and
`visual/dexterity disabilities. A significantly (P < 0.05)
`
`higher proportion of patients expressed overall prefer(cid:173)
`ence for SoloStar (53%) versus FlexPen (31 %) or Lilly
`pen (15%).
`Conclusion: Of the 4 pens compared, both the
`SoloStar pen and FlexPen were found to have high pa(cid:173)
`tient usability, and the new SoloStar pen was found to
`have high patient preference in these patients with dia(cid:173)
`betes. 1Clin Ther. 2007;29:650-660) Copyright© 2007
`Excerpta Medica, Inc.
`Key words: Insulin device, diabetes, SoloStar®,
`FlexPen®, Humulin®/Humalog® pen.
`
`INTRODUCTION
`It is well established that improvements in long-term
`glycemic control can reduce the incidence and delay
`the progression of diabetic complications. 1,2 Patients
`with type 1 diabetes require insulin from diagnosis.
`Patients with type 2 diabetes initially benefit from
`lifestyle intervention programs, for example diet and
`exercise, 3- 5 but oral antidiabetic agents (OADs) and
`insulin therapy are usually required over time. 6
`It is becoming increasingly apparent that patients
`with type 2 diabetes benefit from the addition of in(cid:173)
`sulin to their therapeutic regimen with OADs. 7- 10
`However, for many patients fear of injections, the in(cid:173)
`convenience of a vial and syringe, and social accepta(cid:173)
`bility can present barriers to the initiation of insulin. 11
`
`Accepted for public:ation March 1, 2007.
`
`Online publication April 23, 2007.
`doi:10.1016/j.clinthera.2007.04.003
`0149-2918/$32.00
`
`Printed in the USA Reproduction in whole or part is not permitted.
`Copyright© 2007 Excerpta Medica, Inc.
`
`650
`
`Volume 29 Number 4
`
`SANOF15_00000758
`PTX-0665.0001
`
`Sanofi Exhibit 2143.001
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Insulin pens have the potential to overcome many of
`the barriers to insulin initiation. Compared with a vial
`and syringe, insulin pens offer substantial improve(cid:173)
`ments in compliance, freedom, and flexibility for all
`insulin-using patients. 11- 13 Studies have reported a pref(cid:173)
`erence for pen devices versus a vial and syringe.14,15 In
`addition, pens may provide more accurate dosing, which
`could improve blood glucose control and long-term out(cid:173)
`comes, 12,13 and may also be associated with increased
`adherence and reduced therapy costs.16
`Prefilled, disposable pens have the advantage of sim(cid:173)
`plicity, with minimal training required, as patients are
`not required to install a new cartridge when the pen is
`empty. However, prefilled, disposable pens may be as(cid:173)
`sociated with ecological issues and cost implications.
`This paper presents the results of a series of qualita(cid:173)
`tive, quantitative, face-to-face interviews, during which
`patients with type l or 2 diabetes carried out simulated
`use exercises (injections were performed into a recepta(cid:173)
`cle, not the body) under observation and answered
`questions about the usability and preference for a new
`3.0-mL, prefilled, disposable insulin pen (SoloStar®*)
`compared with 2 currently available pens (FlexPen®t
`and Humulin®/Humalog®+ pen [hereafter referred
`to as Lilly Disposable pen]) and a fourth, prototype pen
`(Pen X§ ). Pen X was an alternative pen concept that
`was in development but was subsequently discontinued
`based on various technical and user feedback data, in(cid:173)
`cluding the results of this study. Hence, the discussion
`
`*Trademark of sanofi-aventis, Paris, France.
`tTrademark of Novo Nordisk A/5, Bagsvaerd, Denmark.
`+Trademark of Eli Lilly and Company, Indianapolis, Indiana.
`§ Produced by sanoti-avent1s.
`
`T. Haak et al.
`
`of the results presented in this paper focuses on
`SoloStar, the Lilly Disposable pen, and the FlexPen.
`This study was carried out as part of the develop(cid:173)
`ment program for SoloStar, which is used to deliver
`insulin glargine and insulin glulisine.
`The objectives of this study were to assess the us(cid:173)
`ability (based on simulated use), specific pen features,
`and patient preference of 4 disposable pcns-SoloStar,
`FlexPen, Lilly Disposable pen, and a prototype pen,
`Pen X. The key features of the 4 pens are summarized
`in Table I.
`
`PATIENTS AND METHODS
`Patients
`Patients with type 1 or 2 diabetes (duration, 22 years)
`were included in the study on a quota basis: insulin(cid:173)
`naive patients with type 2 diabetes receiving OADs
`(quota: 50% of participants); and insulin-experienced
`patients with type 1 or 2 diabetes receiving insulin
`(quota: 50% of participants) via reusable or disposable
`pen. For the participants in the US arm of the study,
`of those using insulin, 50% were to be pen users and
`the remaining 50% were to be vial and syringe users.
`Patients' age range was set at 11 to 85 years, with an
`equal distribution of males and females and concomi(cid:173)
`tant conditions. The study included cohorts of diabetes
`patients with dexterity problems and visual impair(cid:173)
`ments, with quotas of 25 patients per country with dex(cid:173)
`terity problems (typically caused by rheumatoid arthritis
`or neuropathy) and 25 patients per country with visual
`impairment (typically partial blindness due to cataracts,
`macular degeneration, or glaucoma). Patients with hear(cid:173)
`ing impairment and color blindness were also included.
`Impainnent was determined by direct questioning of re(cid:173)
`spondents during the screening phase.
`
`Table I. Comparison of the technical features of insulin pens.
`
`Feature
`
`SoloStar*
`
`FlexPent
`
`Lilly Disposable+
`
`Pen X§
`
`Dimensions (L X D), rnm
`Weight, g
`Maximum single dose, LJ
`
`163 X 15.5
`25.7
`80
`
`158 X 15.5
`24.1
`60
`
`158 X 17.5
`30.1
`60
`
`164 X 15.9
`29.5
`80
`
`L - length; D - diameter.
`*Trademark of sanofi-avent1s, Pans, France.
`tTrademark of Novo Nordisk A/5, Bagsvaerd, Denmark.
`+Trademark of Eli Lilly and Company, Indianapolis, Indiana.
`§Produced by sanofi-aventis.
`
`April 2007
`
`651
`
`SANOF15_00000759
`PTX-0665.0002
`
`Sanofi Exhibit 2143.002
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Clinical Therapeutics
`
`Written informed-consent forms for participating
`in the study were completed by all patients. All re(cid:173)
`spondents signed a confidentiality agreement before
`taking part in the study.
`
`Study Design
`The study was conducted in November 2004 at
`24 centers in the United States (10 centers), France (5),
`Germany (5), and Japan (4). The study consisted of
`qualitative, quantitative, face-to-face 1-hour interviews
`with diabetes patients, carried out by independent mod(cid:173)
`erators from a research agency. The research agency
`developed the questionnaires with input from the
`sponsor.
`The study was divided into 2 major sections, which
`are described below.
`
`Part 1: Usability
`Respondents were asked to prepare each pen
`(SoloStar, FlexPen, Lilly Disposable pen, and Pen X) to
`perform injections into a receptacle. To avoid "practice
`bias," the order of pen use was rotated using a Latin(cid:173)
`squares design (4 versions) to balance first placement
`of each pen and the number of times it followed each
`of the other pens. Participants were not blinded to the
`make/manufacturer of the pens, as pen users would
`most likely be able to recognize their pen according to
`shape and color. This was also necessary as patients
`were asked to compare and rank the 4 devices overall
`and according to specific features.
`ln an attempt to replicate use in clinical practice, a
`user's manual was present for each pen; however, re(cid:173)
`spondents were not required to use this. The modera(cid:173)
`tors recorded the extent to which the respondents
`correctly completed each individual task, plus the ex(cid:173)
`tent to which all tasks were completed correctly in se(cid:173)
`quence (without any assistance). Moderators did not
`provide assistance throughout the study but, at its
`conclusion, examined the pens to ensure the pens were
`working correctly. The independent moderators did not
`provide any training or guidance for the use of pens;
`participants were asked to rely on prior knowledge, in(cid:173)
`tuition, and the relevant user manual for each pen.
`The usability section consisted of the following
`sequential tasks:
`
`• Getting started and removing the cap;
`• Attaching a needle;
`• Setting (including activation of the dose knob with
`
`the Lilly Disposable pen) and delivering a safety
`dose; and
`• Dialing a 40-U dose and delivering that dose
`
`Time to completion was recorded for the following
`groups:
`
`• Total sample;
`• Subgroup by age (11-15 years and ~60 years);
`• Subgroup by current therapy (insulin users ver(cid:173)
`sus OAD users [insulin naive]);
`• Subgroup by previous pen use; and
`• Subgroup by disability status
`
`The second section of the study was introduced
`once the usability section had been completed, and is
`described below.
`
`Part 2: Preference
`Part 2 consisted of the preference assessment. Re(cid:173)
`spondents evaluated 14 key features for each pen on a
`5-point scale (1 = poor to 5 = excellent). Respondents
`were then asked to choose which pen was the best, sec(cid:173)
`ond best, third, and fourth for each of those features.
`Again, a Latin-squares design (4 versions) was used to
`balance first placement of each pen and the number of
`times it followed each other pen, to avoid practice bias.
`The respondents were requested to rank the injection
`force of each pen by injecting 40 U into a receptacle;
`this task was completed for each pen in a random
`order. So that the respondents did not have to recall the
`injection force from a prior exercise, pens were inject(cid:173)
`ed one after another for a true comparative assess(cid:173)
`ment. To conclude the interview, respondents were
`asked to take everything into consideration and rank
`the pens in order based on their overall pen preference.
`
`Statistical Analyses
`Comparisons were made between the pens at every
`step. Statistical tests (ie, 2-, 3- or 4-way x2 analysis)
`were performed as appropriate, with a significance
`level of a= 5% at a power of 80%. Analyses were con(cid:173)
`ducted for exploratory purposes and were not adjust(cid:173)
`ed for multiple comparisons.
`
`RESULTS
`Patients
`A total of 510 patients with diabetes were included
`in the study, 150 from the US and 120 each from
`
`652
`
`Volume 29 Number 4
`
`SANOF15_00000760
`PTX-0665.0003
`
`Sanofi Exhibit 2143.003
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`T. Haak et al.
`
`Germany, France, and Japan. Of these, 176 (35%)
`had type 1 diabetes and 334 (65%) had type 2 dia(cid:173)
`betes. There was an equal distribution of female (260
`l51 %j) and male (250 l49%J) patients, the mean age
`of whom was 43 years (range, 11-82 years). The mean
`duration of diabetes was 10 years (range, 2-54 years);
`159 (31%) patients had hypertension and 137 (27%)
`had high cholesterol levels.
`At baseline, patients were identified by question(cid:173)
`naire as having the following disabilities: visual im(cid:173)
`pairment (98 [19%] patients); manual dexterity im(cid:173)
`pairment (81 [16%]); hearing impairment (39 [8%]);
`and color blindness (26 [5%]).
`Patient characteristics according to treatment expe(cid:173)
`rience at baseline are summarized in Table II. Of the
`278 patients who used insulin, 232 (84%) were cur-
`
`rently using or had previously used insulin pens: 4 7
`(17%) were currently using or previously used the
`FlexPen and 83 (30%) were currently using or previ(cid:173)
`ously used the Lilly Disposable pen. A total of 209
`(75%) patients used pens other than those included
`in this study, of whom 57 (21 %) were currently using or
`previously used the OptiPen® (sanofi-aventis).
`
`Part 1: Usability Assessment
`The results for the usability assessment are report(cid:173)
`ed for the correct completion of all steps without the
`safety step or attach needle step (as these were deemed
`independent of the device). In the overall group, a
`greater proportion of patients correctly completed the
`steps (without the safety step or attach-needle step)
`with SoloStar versus all comparator pens (P < 0.05)
`
`Table II. Baseline demographic and clinical characteristics of the study patients.
`
`Characteristic
`
`Age, mean (range), y
`Sex, no. (%)
`Female
`Male
`
`Current diabetes regimen, no.(%)*
`OADs only
`Insulin only
`Insulin+ OADs
`
`Type of pen (ever used), no. (%)*t
`Disposable only
`Reusable only
`Both
`
`Total pen use, no.(%)*
`
`No pen used, no. (%)*
`
`Previous experience with study pens
`(ever used), no.(%)*
`FlexPen*
`Lilly Disposable§
`Other
`
`OADs Only
`(n - 232)
`
`55 (19-82)
`
`121 (52)
`111 (48)
`
`232 (100)
`0
`0
`
`0
`0
`0
`
`0
`
`232 (100)
`
`0
`0
`0
`
`Insulin Users
`(n - 278)
`
`33(11-80)
`
`139 (50)
`139 (50)
`
`0
`242 (87)
`36(13)
`
`75 (27)
`92 (33)
`50 (18)
`232 (83)
`
`51 (18)
`
`47 (17)
`83 (30)
`209 (75)
`
`OADs - oral antidiabetic drugs.
`*Percentage value corresponds to proportion within treatment group.
`tFour percent (n - 10) of respondents could not recall the correct name of the pen they had previously used;
`as such, their pen use could not be classified.
`*Trademark of Novo Nordisk A/5, Bagsvaerd, Den mark.
`§Trademark of Eli Lilly and Company, Indianapolis, Indiana.
`
`April 2007
`
`653
`
`SANOFI5_00000761
`PTX-0665.0004
`
`Sanofi Exhibit 2143.004
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Clinical Therapeutics
`
`and with FlexPen versus the Lilly Disposable pen and
`Pen X (P < 0.05) (Figure 1A). However, although sta(cid:173)
`tistically significant, the clinical relevance of a differ(cid:173)
`ence between SoloStar (482 l94%j) and FlexPen (457
`[90%]) remains to be established. In the overall group,
`results including the safety step or attach-needle step
`found the same trend for each pen but had lower suc(cid:173)
`cessful completion rates (Figure 1B), although this
`was expected from clinical experience, in which pa(cid:173)
`tients frequently omit safety tests.
`The results according to age, pen use, and dexterity
`subgroups are shown in Figure 2. SoloStar and FlexPen
`were comparable and both pens were more usable than
`the Lilly Disposable or Pen X in patient groups aged
`260 years or 11 to 15 years. In general, the younger
`age groups performed the steps more successfully than
`the older patients (Figure 2A), although there was a
`tendency for more patients in both age groups to suc(cid:173)
`cessfully complete the steps with SoloStar and FlexPen
`compared with the Lilly Disposable pen or Pen X.
`For pen-experienced patients, a significantly great(cid:173)
`er proportion of patients correctly completed the steps
`with SoloStar and FlexPen compared with the other
`2 pens (P < 0.05) (Figure2B). Significantly more insulin(cid:173)
`naive patients correctly completed all steps with SoloStar
`compared with FlexPen, Lilly Disposable pen, and Pen X
`
`(P < 0.05 for SoloStar vs the other 3 pens) (Figure 2C).
`Overall, for the pen-experienced patients, more were
`able to complete the steps correctly with SoloStar and
`FlexPen than with the Lilly Disposable pen (Figure 2D).
`For the Lilly Disposable pen, patients who had expe(cid:173)
`rience with this pen were more likely to complete all
`of the steps than patients who had experience with ei(cid:173)
`ther FlcxPcn or OptiPcn.
`There were no differences between dexterity- and
`visually impaired patients with any pen; the propor(cid:173)
`tion of patients correctly completing the steps tended
`to be higher with SoloStar and FlexPen users compared
`with the Lilly Disposable pen and Pen X (Figure 2E).
`
`Part 2: Preference
`Pen-Feature Comparison
`The evaluation of pen-feature comparisons is shown
`in Table III. Of the 4 attributes relating to the pens'
`design and esthetics, the SoloStar pen was rated as
`best significantly more often versus all other pens for
`tactile feel (P < 0.05), while the Lilly Disposable pen
`performed better for the "how well the cap fits the
`pen" question (P < 0.05). Of the 10 attributes re(cid:173)
`lating tu the pens' usability, the SoloStar pen was rat(cid:173)
`ed as best more frequently for 7 attributes (all, P <
`0.05), and the Lilly Disposable pen was rated as best
`
`A
`
`100
`
`V,
`
`...-
`C
`QJ
`·;:;
`n:I
`0..
`4-
`0
`
`~
`
`80
`
`60
`
`40
`
`20
`
`0
`
`42
`
`41
`
`B
`
`VJ
`
`"' C
`QJ
`·;:;
`cu
`0..
`4-
`0
`
`~
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`SoloStar§ FlexPenl
`
`Lilly
`Disposable~
`
`Pen X#
`
`SoloStar
`
`FlexPen
`
`Lilly
`Disposable
`
`Pen X
`
`Pen Type
`
`Pen Type
`
`Figure 1. The proportion of patients completing: (A) steps not including the safety step or attach-needle step and
`( B) all steps including safety step or attach-needle step using SoloStar, FlexPen, Lilly Disposable pen, and
`Pen X in the total study population. * P < 0.05 versus FlexPen; tp < 0.05 versus Pen X; tp < 0.05 versus
`Lilly Disposable pen; §Trademark of sanofi-aventis, Paris, France; 11Trademark of Novo Nordisk A/5,
`Bagsvaerd, Denmark: IJTrademark of Eli Lilly and Company, Indianapolis, Indiana; "'Produced by sanofi(cid:173)
`aventis. (All P values were calculated without adjustment for multiple comparisons.)
`
`654
`
`Volume 29 Number 4
`
`SANOF15_00000762
`PTX-0665.0005
`
`Sanofi Exhibit 2143.005
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`T. Haak et al.
`
`■ 260 years (n = 144)
`□ 11-15 years ( n = 116) B
`
`Pen users (n = 232)
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`A
`
`100
`
`98
`
`80
`
`60
`
`40
`
`20
`
`0
`
`SoloStar* FlexPent
`
`Lilly
`Disposable+
`
`Pen X§
`
`SoloStar
`
`FlexPen
`
`Lilly
`Disposable
`
`Pen X
`
`Pen Type
`
`Pen Type
`
`C
`
`100
`
`Insulin-naive (n = 232) . D
`
`96 96 98
`
`96 95 98
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`80
`
`60
`
`40
`
`20
`
`0
`
`■ FlexPen (n = 47)
`□ Lilly Disposable (n = 83)
`□ OptiPen (N = 57)
`
`93**tt
`
`75
`
`70
`
`81
`
`73
`
`67
`
`SoloStar
`
`FlexPen
`
`Lilly
`Disposable
`
`Pen X
`
`SoloStar
`
`FlexPen
`
`Lilly
`Disposable
`
`Pen X
`
`Pen Type
`
`E
`
`100
`
`91
`
`94
`
`Pen Type
`
`■ Dexterity impaired (n = 81)
`□ Visually impaired (n = 98)
`
`80
`
`60
`
`40
`
`20
`
`0
`
`SoloStar
`
`FlexPen
`
`Pen X
`
`Lilly
`Disposable
`Pen Type
`Figure 2. The proportion of patients completing all of the steps (not including safety step or attach-needle step)
`using SoloStar, FlexPen, Lilly Disposable pen, and Pen X according to: (A) age; ( B) previous treatment
`in pen users; (C) previous treatment in insulin-naive users; (D) previous pen use; (E) disability.
`*Trademark of sanofi-aventis, Paris, France; tTrademark of Novo Nordisk A/S, Bagsvaerd, Denmark;
`*Trademark of Eli Lilly and Company, Indianapolis, Indiana; §Produced by sanofi-aventis; 11 P < 0.05 ver(cid:173)
`sus Lilly Disposable pen; ~p < 0.05 versus Pen X; #p < 0.05 versus FlexPen; **P < 0.05 versus previous
`OptiPen use; ttp < 0.05 versus previous FlexPen use. (All P values were calculated without adjustment
`for multiple comparisons.)
`
`April 2007
`
`655
`
`SANOF15_00000763
`PTX-0665.0006
`
`Sanofi Exhibit 2143.006
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Clinical Therapeutics
`
`Table Ill. Pen feature comparison: Percentage of time that the pens were rated as "best" by patients using the
`SoloStar, FlexPen, Lilly Disposable, and Pen X (N = 510).a,b
`
`Evaluation of pen features: no. (%) of patients
`who rated each pen as "best"
`
`SoloStarc
`
`FlexPend
`
`Lilly Disposablee
`
`Design/ esthetics
`Exterior design and styling
`Size and portability
`How well cap fits onto pen
`Tactile feel
`Usability
`Easy/imuiLive Lo rigure ouL
`Easy to set dose
`Easy to read that you have set exact dose
`Easy to correct dose if overdialed
`Auditory feedback
`Requires low number of turns to set 40 U
`How far the dose button sticks out ( 40 U)
`The effort it takes to inject 40 U
`Easy to determine entire dose delivery
`Easy to determine amount left in cartridge
`
`182 (36)g,h
`216 ( 42)g,h
`175 (34)h
`216(42)g,h,j
`
`278 ( 55 )g,h,j
`260 ( 51 )g,h,j
`198 (39)g,h,j
`257 ( 50)g,h,j
`186 (36)g,h
`211(41)g,h,j
`171 (34)h,j
`319 (63)g,h,j
`255 ( 50 )g,h,j
`194 (38)h,j
`
`183 (36)g,h
`203 ( 40)g,h
`146 (29)h
`153 (30)g,h
`
`164 (32)g,h
`150 (29)g,h
`124(24)g
`170 (33 )g,h
`186 (36)g,h
`150 (29)h
`110(22)h
`98 (19)h
`148 (29)g,h
`131 (26)
`
`111(22)h
`112(22)h
`224 (44)h,i,j
`78 ( 15)
`
`64 (13)h
`54 ( 11)
`82 ( 16)
`82 ( 16)
`66 (13)
`142 (28)h
`234 ( 46)h,l,j
`81 (16)h
`120(24)h
`171 (34 )h,j
`
`Pen xr
`
`59 (12)
`61 (12)
`94 (18)
`89 ( 1 7)
`
`43 (8)
`82 (16)g
`133 (26)g
`115 (23)g
`90 (18)g
`63 (1 2)
`33 (6)
`13 (3)
`56 (11)
`11 0 (22)
`
`a Respondents ranked the pens from "best" to "worst" for each attribute.
`b Percentages may total >100% as some respondents could not select one pen as "best" but instead opted for "ties."
`c Trademark of sanofi-aventis, Paris, France.
`d Trademark of Novo Nordisk A/S, Bagsvaerd, Denmark.
`e Trademark of Eli Lilly and Company, Indianapolis, Indiana.
`f Produced by sanofi-aventis.
`g P < 0.05 versus Lilly Disposable pen.
`h P < 0.05 versus Pen X.
`; P < 0.05 versus SoloStar.
`j P < 0.05 versus FlexPen.
`
`most often for 1 (how far the button sticks out at 40 U)
`(P < 0.05).
`
`versus FlexPen (160 [31 % ]) and Lilly Disposable pen
`(78 [15%]) (both, P < 0.05) (Figure 4).
`
`Injection Performance
`As a first choice, the injection force of the SoloStar
`pen, FlexPen, Lilly Disposable pen, and Pen X was
`preferred by 351 (69%), 81 (16%), 67 (13%), and 9
`(2 % ) patients, respectively (P < 0.05 for SoloStar vs
`all other pens) (Figure 3). As an overall first or second
`choice, SoloStar pen, FlexPen, Lilly Disposable pen,
`and Pen X were chosen by 451 (88%), 315 (62%),
`203 (40%), and 45 (9%) patients, respectively (P <
`0 .05 for SoloStar vs all other pens) (Figure 3 ).
`
`Overall Preference
`A significantly higher proportion of patients ex(cid:173)
`pressed an overall preference for SoloStar (269 [53%])
`
`DISCUSSION
`In this study, which compared SoloStar, the FlexPen,
`Lilly Disposable pen, and a prototype pen (Pen X), the
`overall usability and overall preference were highest in
`patients using the SoloStar pen; the results obtained in
`the overall population were also observed in the patient
`subgroups and with individual pen feature attributes. In
`the total study population, the majority (94%) of pa(cid:173)
`tients were able to complete all steps correctly with
`SoloStar. Similarly, compared with the Lilly Disposable
`pen and Pen X, the FlexPen was also found to be user(cid:173)
`friendly. This finding is in agreement with previous
`studies, in which patients reported a strong overall
`preference for the FlexPen (7 5 % ) compared with the
`
`656
`
`Volume 29 Number 4
`
`SANOF15_00000764
`PTX-0665.0007
`
`Sanofi Exhibit 2143.007
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`T. Haak et al.
`
`□ Most preferred
`■ 1st or 2nd preferred
`
`88*H
`
`V,
`
`'"' C
`(l)
`·;::;
`ro
`CL
`'-'-
`0
`??
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`SoloStar§
`
`FlexPenll
`
`Lilly Disposable'I
`
`Pen Type
`
`Figure 3. The proportion of patients preferring (as first choice or first/second choice) the injection force of
`SoloStar, FlexPen, Lilly Disposable pen, or Pen X, as assessed by the comparative questionnaire. The pa(cid:173)
`tients were asked to set a dose with each pen and inject. They were then asked which pen required the
`best, second best, or worst effort to inject. *P < 0.05 versus FlexPen; tp < 0.05 versus Lilly Disposable
`pen; +P < 0.05 versus Pen X; §Trademark of sanofi-aventis, Paris, France; 11Trademark of Novo Nordisk
`A/S, Bagsvaerd, Denmark; IJTrademark of Eli Lilly and Company, Indianapolis, Indiana; #Produced by
`sanofi-aventis. (All P values were calculated without adjustment for multiple comparisons.)
`
`60
`
`50
`
`53•t
`
`40
`
`V1
`µ
`C
`9!
`µ
`n:I
`CL
`'-'-
`0
`~ 20
`
`30
`
`10
`
`0
`
`SoloStart
`
`FlexPen&
`
`Lilly Disposablell No preference
`
`None
`
`0
`
`Pen Type
`
`Figure 4. The proportion of patients with an overall preference (first choice) for SoloStar, FlexPen, or Lilly
`Disposable pen as assessed by the comparative questionnaire. *P < 0.05 versus FlexPen; tp < 0.05 ver(cid:173)
`sus Lilly Disposable pen; +Trademark of sanofi-aventis, Paris, France; §Trademark of Novo Nordisk A/S,
`Bagsvaerd, Denmark; IITrademark of Eli Lilly and Company, Indianapolis, Indiana. (All Pvalues were cal(cid:173)
`culated without adjustment for multiple comparisons.)
`
`Humalog (Lilly Disposable) pen (14%) 17 and in which
`the FlcxPcn was found to be easier to use compared
`with a conventional vial and syringe (74% vs 21 % ). 14
`The Lilly Disposable pen has previously been reported
`to be widely accepted and preferred by patients with
`previous experience of both reusable and prefilled de-
`
`vices. 18 On the other hand, Pen X was preferred by only
`2 % of the participants for its injection force, for exam(cid:173)
`ple. Therefore, the development of Pen X was consid(cid:173)
`ered unlikely to offer a clinical advantage to patients
`over the currently available pens. Based on these and
`other data, development of Pen X was discontinued.
`
`April 2007
`
`657
`
`SANOF15_00000765
`PTX-0665.0008
`
`Sanofi Exhibit 2143.008
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`Clinical Therapeutics
`
`This study was designed to explore the usability
`and preference for different pens. Given this ex(cid:173)
`ploratory purpose, no individual feature was isolated
`as dominant for the study design (ie, as necessary for
`determination of superiority) or for sample-size deter(cid:173)
`mination. In addition, no adjustment for multiple com(cid:173)
`parisons was planned for the analysis. However, raw
`data showed an overall preference for SoloStar (53°/.,)
`versus FlexPen (31 %) and the Lilly Disposable pen
`(15%).
`For the first part of the study, patients performed
`the sequential steps once per pen. For insulin-naive
`patients, this could result in practice bias, as the par(cid:173)
`ticipants might have "learned" how to use the pens
`and, thus, found the last pen easiest to use. To over(cid:173)
`come this limitation, a Latin-squares design was se(cid:173)
`lected to balance the order in which pens were
`introduced. In the second part of the study, partici(cid:173)
`pants were able to handle the pen as necessary and
`were not given restrictions on the use of each pen.
`The interviews in this study were based on task(cid:173)
`completion to examine usability and a question-and(cid:173)
`answer session to determine patient preference. Al(cid:173)
`though this type of structure, used in other studies, 19,20
`does not allow evaluation of each device over an ex(cid:173)
`tended period of actual use with insulin, 14,18,21- 23 it does
`allow for a comparison of the characteristics of pens
`alone, not the combination of insulin and pen. This
`study prospectively examined patient pen preference
`overall and for each feature, as participants were al(cid:173)
`lowed to concurrently handle each pen and directly
`compare features, without needing to recollect previ(cid:173)
`ous therapy, as was necessary in some studies. 18,22,23
`The nature of the present study ( 1-hour interviews
`during which the patients were able to use and com(cid:173)
`pare the pens and express a preference for a specific
`device) may also mirror clinical practice for many pa(cid:173)
`tients, in which a diabetes educator or nurse would
`offer the patient the opportunity to test and choose
`from a selection of appropriate devices commensurate
`with the prescribed insulin regimen.
`Pen devices that are easy and efficient to operate
`correctly with minimal discomfort appear to increase
`patients' acceptance ot and adherence to, treatment
`rcgimcns.14,24 From a treatment compliance perspec(cid:173)
`tive, simplicity of use could prove particularly impor(cid:173)
`tant for older patients (aged 260 years) who may have
`difficulty handling injection devices.2° Compared with
`a vial-and-syringe regimen, insulin pens can make in-
`
`jections easier and simpler for the elderly, thus pro(cid:173)
`moting self-care in this population.25 Indeed, clinical
`studies have reported that insulin pens are well(cid:173)
`tolerated, effective, and highly accepted in elderly pa(cid:173)
`tients.26,27 In our study, a large proportion of pa(cid:173)
`tients aged 260 years completed the steps correctly
`with the SoloStar pen (90%) and FlexPen ( 83 % ), sug(cid:173)
`gesting the suitability of these pens for older patients
`with diabetes.
`Tight blood glucose control is notoriously difficult
`to achieve in children and young adults due to com(cid:173)
`pounding endocrine, behavioral, and social factors. 28
`In our study, we found that in younger patients (aged
`11-15 years), in whom convenience and social accept(cid:173)
`ability of insulin regimens would represent imponam
`factors, there was almost a 100% successful comple(cid:173)
`tion rate using the SoloStar pen and FlexPen. There
`was also a relatively successful completion rate with
`Lilly Disposable pen and Pen X, which could ret1ect a
`young person's ability to operate mechanical devices.
`In patients with type 2 diabetes, and for those phy(cid:173)
`sicians treating them, fear of hypoglycemia remains
`one of the key obstacles to both initiating and appro(cid:173)
`priately titrating the dose of insulin. 6,29 Furthermore,
`both elderly and young people appear to be at partic(cid:173)
`ular risk for hypoglycemia, 3o,3 t and several smdies
`have reported that large variations in insulin doses are
`observed using a vial and syringe in both these age
`groups. 32,33 The high usability (tested through simu(cid:173)
`lated use) of the SoloStar and FlexPen devices, in par(cid:173)
`ticular, observed in both of these high-risk patient
`groups in our study would be expected to reduce the
`risk for over- or underdosing insulin.
`Common complications of diabetes include visual
`impairment and problems with manual dexterity
`owing to peripheral neuropathy.34 These complica(cid:173)
`tions make diabetes self-management difficult unless
`treatment regimens can accommodate these patient
`needs. For patients with visual or dexterity problems,
`injection devices with easily readable dose scales and
`easy-to-handle dose selectors are preferable. Here we
`found that the majority of patients with visual or dex(cid:173)
`terity impairments were able to complete all steps cor(cid:173)
`rectly using both the SoloStar and FlexPen devices. In
`addition, attributes relating to the case of use, such as
`"easy to set dose," were preferred in the SoloStar pen
`in head-to-head comparisons with all of the other
`pens. The greater preference among elderly patients
`and in patients with impai