`
`Dovepress
`
`REVIEW
`
`An evaluation of prefilled insulin pens: a focus
`on the Next Generation FlexPen®
`
`i J. /ol
`
`I Cr- 3-/if
`
`Estella M Davis
`Emily L Sexson
`Mikayla L Spangler
`Pamela A Foral
`
`Department of Pharmacy Practice,
`Creighton Unive rsity School o f
`Pharmacy and He a lth Professions,
`Omaha, N e br aska, USA
`
`Corresponde nce: Estella M Davis, PharmD
`Department of Pharmacy Practice,
`Creighton University Schoo l of Pharmacy
`and Health Professions, 2500 Califo rnia
`Plaza, Omaha. Nebraska 68178, USA
`Te l + 1-402-398-5646
`Fax + 1-402-398-5928
`Email edavi s@cre ighcon.edu
`
`Abstract: Insulin pen delive ry systems are preferred by pati ents over the tra ditional vial and
`syringe method fo r insuli n deli very because th ey are simpl e an d easy to use, improve co nfidence
`in dosing insulin, and have less interfe rence w ith activities and im proved di screti on w ith use .
`Insulin manufac turers have made num ero us improve ments to th eir fi rst marketed pen devi ces
`and are now introducing their nex t generati on of devices. Design modifi cati ons to th e newest
`generation ofprefilled insulin pen devi ces are intended to imp rove th e ease of use and safety and
`continue to positively impact adherence to insulin . Thi s review focuses on the Nex t Ge neration
`FlexPen® with regard to design considerati ons to reduce injecti on fo rce, im prove accuracy and
`ease ofuse, and evaluate the preference of pati ent and health-ca re provide r compa red with other
`disposable, prefilled insulin pen devices.
`Keywords: diabetes, dose accuracy, injection force, pati ent preference, insulin pen device
`
`Introduction
`Global estimates indicate the total number of indiv idual s with di abetes w ill increase
`from 171 million in 2000 to a proj ected 366 millio n people by 2030, likely due to the
`popul ation growth, ag ing, urbanizati on, and increas ing preva lence ofobesity and lack
`of physica l activity. 1 Estimates from 2007 indicate the prevale nce of undi ag nosed and
`di agnosed patients w ith di abetes in the United States alone to be 23.6 millio n people
`or 7 .8% of the populati on. 2
`Studi es s how that maintaining glycosylated hemoglobin (HbA 1) goals close to
`
`the range of nondiabeti c patients reduces the risk of microvascul a r compli cati ons.3- 8
`In order to achieve HbA 1, goals and maintain glycemic contro l, insulin remain s the
`
`cornerstone of therapy for patients w ith type I di abetes. 9 Furthermore, insulin admin(cid:173)
`istration is recomme nded as an addi tional method to intensify therapy when other
`antidiabetic agents and li fes tyle modi fi cation s are insu ffici ent to meet the HbA 1c goal s
`11
`for patients with type 2 di abetes.10
`A treatment algori thm , formul ated by a consensus panel of the American Di abetes
`Associ ation (ADA) and E uropea n Assoc iation fo r the Study of Di abetes (EASD ), to
`manage pati ents with type 2 diabetes recomme nds a n opti o n of additi o nal therapy with
`insulin after monotherapy w ith metfo rmin does not achieve the HbA1c goa ls. 10
`Th e treatment algorithm , formul ated by th e Ame rica n Assoc iation of C lini ca l
`E ndocrin ologists (AACE) and American College of E ndoc rin ology (ACE), stratifi es
`patients with type 2 di abetes based on their curre nt HbA 1, va lue w ith a goal of
`
`monitoring therapy every 2- 3 months and intensify in g therapy until the H bA 1c goa l
`
`•
`
`~u bn, it yo u r n1 ~nu~cr ip1 w wv, .-<, "+ ,1 · •
`Dovepn.: \ <,
`
`DO I: IC.2147 MDER.S l 17Hl
`
`41
`Medical Devices: Evide nce and Research 20 I 0:3 41-50
`© 2010 Davis et al, publisher and li ce nsee Dove Medica l Press Ltd.Thi s is an O pe n Access article which
`pe rmits un restricted no ncomme rcial use, provided the o rigi nal wo r k is pro perly cited.
`
`Sanofi Exhibit 2101.001
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`Davis et al
`
`Dovepres~
`
`,
`
`has been reached. lt recommends that for patients with HbA 1
`values > 9% and on antidiabetic medications or if medication
`naive and symptomatic, insulin therapy should be considered.
`For patients with HbA 1, va lues < 9% and combinations of
`
`dual or triple a ntidiabetic medications fail to achieve the
`HbA 1, goa l of s:; 6.5%, insulin therapy should be considered
`
`as an additional method of intensification. 11
`Despite these recommendations , it is estim ated th at
`on ly 27% of the adu lt American population diagnosed with
`diabetes are on some type of insulin treatm ent, whereas 73%
`take e ither ora l medicatio n or no medication at all. 2 Further
`research is needed to assess the percentage of patients with
`type 2 diabetes who shou ld have augmentation with insulin
`therapy accord ing to these guidelines .
`Mu ltiple patient factors and attitudes regarding insulin
`contribute to the overal l reluctance to initiate therapy.
`Certain patient attitudes presenting a barrier to insulin use
`include: fear of hypog lyce mic complications, increased
`compl ex ity of managing diabetes, li festyle restri c tion s,
`13 A
`soc ia l un acceptabi lity, a nd fear of self-injecting. 12·
`survey validation study confirmed a positive co rrel ation
`among three main pen product attributes that relate to the
`prefe rence for insulin pens compared with via ls and syringes
`including ease of use, less activity interference, and social
`acceptability. 14 Since the first introduction of insulin pens to
`the m arket, consideration of these three mai n attributes per(cid:173)
`meates throughout the des ign and evaluation of various pen
`dev ices in an effort to positive ly inAu ence pati ent preference
`and u ltimately adherence to insulin regimens.
`A lth ough the traditiona l via l an d syringe method is
`available for the delivery of insu lin , this method requ ires
`extensive training and the patient must have the approp ri ate
`visua l ac uity, manual dexterity, and coordination to properly
`prepare and administer an insu lin injection .15 Stud ies have
`shown patients with diabetes prefer insulin pens over v ials
`and syringes because of the improvements in the following
`features : ease of use, confidence in dosing, discretion with
`use, compliance, quality of life, and independence of admin(cid:173)
`24
`istration in patients with visua l or motor disabilities. 15
`-
`Furthermore, national health-care benefit studies revealed the
`transiti on from via ls and syringes to insulin pens improves
`medication adherence and reduces overa ll hea lth-care costs,
`emergency depar tment and physician visits, and the likeli(cid:173)
`27
`hood of experiencing a hypoglycemic event. 25
`-
`The purpose of this review is to prese nt an eva luation
`of the Next Generation FlexPen E (NGFP) (Novo Nordisk,
`Bagsvae rd, Denmark) compared with other dispo sa ble ,
`prefillecl insulin pen devi ces. Emphasis will be pl aced on
`
`eva luat ing the uti li ty of thi s device regarding the design
`considerations to improve accuracy, reduce injection force ,
`and eva lu ate the preference of patient and hea lth-care pro(cid:173)
`vider with NGFP compa red with other disposable, prefi ll ed
`insulin pen devices.
`A Pubmed search was conducted to identify studies
`publi shed from 1985 to February 20 10 using the search
`term sftexpen, 11 exl generationflexpe11 , prefillecl pe11 , insulin
`pen , and insulin delivery device. References of ide ntifi ed
`articles and pharmaceutica l websites were a lso reviewed for
`additional pertinent articles .
`
`The evolution of new-generation
`prefilled insulin pens
`lnsul in pen device delivery systems were created in 1985 with
`the intent to overcome barriers of the via l and syringe method.
`Insulin pen dev ices combine an insu lin rese rvoir cartridge
`and syringe into a single component in a n effort to overcome
`barriers to adherence with insulin self-admini stration and
`improve convenience and ease of use for patients. 28 Insulin
`pen devices are typically classified as being either durable
`(reusab le) or prefi ll ed (disposable). Durable insulin pen
`devices use repl aceable and disposable insulin cartridges that
`are loaded and removed from the in sulin de livery pen by the
`pat ient. Prefi ll ed insulin pen devices require no install ation
`of an insulin reservoir cartridge by the patient. The entire
`device including the body of the pen and prefilled insulin
`cartridge can be discarded once it is empty. Both types of
`devices contain 3 mL of insu lin ( I 00 U /mL) , for a total of
`300 U of insu lin and require attachment of an insulin pen
`need le to ad mini ster a dose. 29
`Dose preparation and insulin administration are simplified
`with prefilled in sulin pens compared with the via l and syringe
`method. Pen device preparation and insulin ad mini stration
`w ith new-generation µrefi ll ed pens share broad ly s imil ar
`techniques. Patients would fo ll ow the following basic steps:
`correct ly identifying the insulin analog for use, removing
`the pen cap, pl acing a n in sulin pen need le on the insu lin end
`of th e pen , and "dialing-up" or sett ing the insulin dose by
`twisting a dosage se lector. At this poi nt, patients can visualize
`their num erical insulin dose and concurrent ly hear audib le
`clicks for each incremental dose increase from zero. Patients
`typically perform a 2 U sa fety airshot of insulin to verify
`whether th e need le is worki ng. Once this is confi rmed and
`the patients have dia led up their insu lin dose, they in sert the
`pen at a 90° angle into subcutaneous tissue and depress the
`injection button on the end of the dosing knob of the pen.
`The dosing window returns to zero, resulting in de li very of
`
`42
`
`Dovep1 .. •<,,
`
`Medical Devices: Evidence and Research 20 I 0:3
`
`Sanofi Exhibit 2101.002
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`Doveprc,~
`
`Next generation of insulin pen devices
`
`insul in . Patients should be instructed to wait for a few seconds
`to allow the absorption of the appropriate amount of insulin
`and withdraw the insulin pen from the subcutaneous ti ssue.
`Due to the ease of administration , patients can correctly dial
`up and administer their insulin with minimal instructions
`using pen devices. 30- 33
`All three manufacturers of insulin dispensed in the United
`States. (Novo Nordisk; Eli Lilly and Company, Indianapo(cid:173)
`lis , Indiana , USA; sanofi -aventis , Bridgewater, New Jersey,
`USA) have disposable, µrefilled insuli n pens to facilitate the
`administration of their corresponding rapid- or long-acting
`insulin ana logs and premixed insu lin ana log preparations
`from the devices (Table l ). Insulin manufacturers have made
`improvements to their first marketed pen devices and are
`now introducing their next generation of devices by making
`design modifications that are intended to improve the ease
`of use and safety and conti nue to positively impact adher(cid:173)
`ence to insulin .
`
`New-generation pen devices:
`product improvements
`Compared with the original FlexPen®(FP) (Novo Nordisk)
`design , the NGFP device has product modifications pro(cid:173)
`ducing a lower injection force , improved accuracy of dose
`delivery, and an easier pen needle interface requiring a
`single-luer lock type of twist to secure a Novo Twist® (Novo
`Nordisk) need le to the pen. These features were implemented
`to enhance convenience and ease of use. To improve patient
`safety, the NGFP imitated the color coding of the pen injec(cid:173)
`tion button found in the original FP, but the design has been
`modifi ed to continue the color coding throughout the entire
`pen body (Figure I) . The color coding assigned to labeling
`and packaging of insulin aspart (NovoRapid®; Novo Nordisk)
`is orange, insulin detemir (Levemi r®; Novo Nordisk) is green,
`
`and insulin aspart protamine/aspart 70/30 mix is blue with
`a clear cartridge.
`To enhance the ease of use , compared with the
`origina l durable OptiC lik;,· (OC) pen (sanofi -ave nti s) , th e
`SoloSTAR" (SS) (sanofi-avent is) pen ha s been modified to
`a µrefilled, di sposa ble pen device (Figure 2). The OC and
`SS are the only pens that all ow a maximum dose adm in(cid:173)
`istration of 80 U. During deve lopment of the SS pens, the
`manufacturers wanted to maintain th e ability to allow the
`maximum insulin dose, but retain a manageable " thumb
`reach" distance, defin ed as the di al extension distance from
`holding the pen in one hand to extending the thumb, and
`low injection force .34 Compared with older-generation pre(cid:173)
`filled pens marketed at the time, the SS pen had the lowest
`mean injection force 35 and was preferred by patients with
`diabetes. 36 These cha nges were implemented to enhance
`convenience and ease of use. If a patient wants to mini(cid:173)
`mize the number of injections required fo r high doses that
`exceed 60 U but are less than 80 U, SS pen may be the ideal
`disposab le pen device .
`In 2006, the lnst itute for Safe Medication Practices
`(JSMP) reported that the digital display for the insulin dose,
`which is near the dial used to set the dose on the OC pen for
`the injection of insulin glargine and insulin glulisine, had the
`potential for dosing errors and patient harm if the pen was
`oriented in the wrong direction . For example, ifa left-handed
`practitioner or patient held the pen upside down, w ith the
`needle to the right, away from the hand, a dose that is actua lly
`52 U may appear as 25 U. ISMP believed th at the design of the
`pen was potentially dangerous and could lead to a sign ificant
`overdose or a subtherapeutic dose of insulin, and thus ISMP
`did not recommend clinica l use of the device until safety
`issues were reso lved. 37 Therefore, the SS pen was designed
`wi thout the digital di splay. Additional improvements were
`
`Table I Prefill ed disposab le insulin pen devices avai lab le in the United States
`
`Manufacturer
`
`Pen
`devices
`
`Insulin
`as part
`
`Insulin aspart
`protamine/
`aspart 70/30 mix
`
`Insulin
`detemir
`
`Insulin
`glulisine
`
`Insulin
`glargine
`
`Insu lin
`lispro
`
`Insulin lispro
`protamine/lispro
`75/25 and
`50/50 mix
`
`Delivery
`range
`(units)
`
`✓
`
`✓
`
`✓
`
`✓
`
`✓
`
`✓
`
`N ovo Nordisk
`
`sanofi •aventis
`Eli Lilly and
`Company
`
`Flex Pen'
`Next
`Generation
`FlexPen
`SoloSTAR
`Humalog
`pen
`
`KwikPen
`
`✓
`
`✓
`
`✓
`
`✓
`
`✓
`
`✓
`
`1-60
`1-60
`
`1-80
`1- 60
`
`1-60
`
`~currently Novo Nordisk manufactures only che Next Generation FlexPen; however, it is possible that bo th the original FlexPen may still be available in some areas
`(depending on use).
`
`Medical Devices: Evidence and Research 20 I 0:3
`
`,ub,ni t yo11r lll~llU~<::rip l W-A'W t'(Vlfl! ('','.' ,:,,·,,
`
`43
`
`Dove pn.•<:,<,
`
`Sanofi Exhibit 2101.003
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`D avis et al
`
`Doveprcss
`
`j
`
`~ .Ii
`I
`I
`~~ s~si
`
`I
`I
`~ :t.
`
`I
`ll
`
`J
`I
`I
`'t 3' i
`
`Figure I View of FlexPen Levemir and FlexPen NovoRapid (left) and Next Generation FlexPen Levemir and Next Generation FlexPen NovoRapid (right).
`
`C:
`0
`~
`
`made utilizing a di ffe rent co loring scheme of pen labeling
`to help di stingui sh between rapid- and long-acting insulin
`ana logs . The rapid-acting analog, insulin glul is ine, is dark
`navy blue, and the long-acting analog, insulin glargi ne, is
`gray. These co lor schemes were va lidated in stud ies includ(cid:173)
`ing pat ients w ith poor visual acuity or co lor blindn ess. 34
`An add itional change to help di ffe renti ate between in sulin
`glarg ine and glul is ine is a raised ring on the dose button of
`the insu lin gluli sine pen to assist w ith tactil e di ffe rentiation
`of the two insulin analogs. These des ign changes to the SS
`pen were impl emented to improve patient safety.
`To enhance the ease of use, compared w ith the orig inal
`Humalog"'!Humulin® pen (HP) (Eli Lilly and Company), the
`Kw ikPen"' (KP) (E li L illy and Co mpany) dev ice was mod i(cid:173)
`fi ed to simpli fy di alin g doses (Fi gure 3). The HP required
`the use r to line up an arrow in the dosin g w indow and pull
`out the dose knob to perfo rm the prim ing step until a di a(cid:173)
`mond appeared . After the pen was properl y prim ed, the user
`lined up the arrow in the dosing wi ndow again and had to
`pull out the dose knob to set the in su lin dose. These steps
`were qui te cumbersome and often led to poor sati sfacti on
`in compari son w ith other insulin pen dev ices.-1 6 Simil ar to
`the other new-generati on insuli n pens, now the KP only
`requires dia ling the dose, whi ch improves the convenience
`and ease of use. The KP is the shortest new-ge ne ra ti on
`
`prefill ed pen. Hence, the HP and KP devices have the shortest
`" thum b reach" distance overa ll. 35
`18 Thi s dev ice may be an
`..
`idea l cho ice fo r a pati ent w ith dexteri ty issues . The KP has
`bee n modi fied to have a lowe r injection fo rce and is co lor
`coded to di stingui sh between ra pid and long-acting analog
`mi xes. The rap id-acting in sulin li spro is burgundy, lispro
`pro tamine/ li spro 75/25 mi x is yel low, and li spro protamine/
`li spro 50/50 m ix is red. Patients who are pen naive prefer
`th e KP over vials and syringes and FP possibly due to these
`des ign modi fica ti ons. 39
`Notab ly, Novo No rdi sk and E li Lilly and Company no
`longer manu fac ture human insulin in their new generati on
`of di sposabl e pen dev ices. Th e regul ar or Neutral protamine
`hagedorn (NPH) hum an insulin alone or combined mixes
`were provided in di sposabl e insulin pen model s of the di s(cid:173)
`continued InnoLd ' (Novo Nordisk, or Princeton, New Jersey,
`USA) and Humulin pens. The AACE/ ACE gu idelines do not
`recommend th e use of short-acting regular human insulin or
`intermed iate-acting NPH , if poss ible, fo r pati ents with type
`2 diabetes. 11 Thi s recommendati on is due to human insulin
`preparat ions' unpred ictabl e tim e course, in ability to mimic
`
`._.,_,
`
`Fi gure 2 View of OptiClik (top) and SoloSTAR (bottom) pens.
`
`Fig ure 3 View of Humalog pen (top) and KwikPen (bottom).
`
`44
`
`,11bn1i 1 yo ur n1,1nusc ript
`
`,·.v, ,,
`
`vr ,.:~
`
`Medi ca l D evices: Evid ence and Research 20 I 0:3
`
`Sanofi Exhibit 2101.004
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`Dovepress
`
`Next generation of insulin pen devices
`
`the normal physiologic profile, and increased risk ofhypogly(cid:173)
`cemia. 11 Similarly, the ADA standards recommend the use of
`rapid- and long-acting insulin analogs for patients with type l
`diabetes since they are associated with less hypoglycemia and
`similar HbA 1c lowering compared with human insulin. 94041
`The ADA/EASD consensus statement and algorithm for
`patients with type 2 diabetes recognizes the use of insulin
`analogs results in lower risk of hypoglycemia. However, their
`recommendations include use of either intermediate- or long(cid:173)
`acting basal insulin and use of either short- or rapid-acting
`prandial insulin. Interestingly, the algorithm omits inclusion
`of short-acting human insulin for prandial coverage. Despite
`their recognition of insulin analogs in reducing the risk of
`hypoglycemia compared with human in sulin , they do not
`conclude the analogs lower the HbA 1, value more effectively
`
`than the human insulin. 10 Therefore, it can only be assumed
`that ceasing the production of human insulin preparations
`in prefilled pen devic es was done in response to consensus
`statements discouraging their use and the shift toward the
`use of insulin analogs.
`
`Dose accuracy
`The accuracy of an insulin delivery system is of utmost
`importance in avoiding diabetes-related complications due to
`either hyperglycemia or hypoglycemia. The new-generation
`insulin pens available today have been shown to be exceed(cid:173)
`ingly accurate.
`Dosing accuracy for insulin pens is based on the regula(cid:173)
`tions set by the International Organization for Standardization
`(ISO). To define positive accuracy for insulin pen-inj ectors
`for medical use, the ISO standard allows for a deviation
`within I U of insulin when administering 20 U or less and no
`greater than 5% deviation for doses greater than 20 U. 42
`Only three studies have evaluated the NGFP compared
`with the original FP or other new-generation pens. 43
`5 The
`-4
`first study aimed to compare NGFP with FP using a total of
`180 delivered doses .43 It was found that neither of the pens
`delivered any doses outside the predefined TSO limits when
`tested at I, 30, or 60 U. The NGFP was more accurate than
`FP at delivering 30 U (P < 0.05) and 60 U based on the mean
`absolute deviation from the set doses. In addition, NGFP was
`more precise than FP at delivering 30 and 60 U (P < 0.05).
`Both NGFP and FP had similar accuracy in delivering I U
`of insulin. 43
`The second study compared NGFP with SS using a total
`of 66 delivered doses. 44 NGFP was outside the predefined I SO
`limits for 1 dose (0.2%) at 10 U and 1 dose (0.6%) at 30 U.
`The SS pen was outside the predefined TSO limits for 2 doses
`
`(0.4%) at 10 U and 3 doses (1.8%) at 30 U. The NGFP was
`more accurate than SS at delivering IOU, with an absolute
`deviation of 1.63% ± 0.84% and 2.1 I%± 0.92%, respectively
`(P < 0.00 I). This was also seen at a dose of3O U, with an
`absolute deviation of 1.23% ± 0.76% and 1.54% ± 0.84%,
`respectively (P < 0.05). 44
`The most comprehensive study to evaluate the accuracy
`or NGFP compa red with the newer generation of prefilled,
`disposable insulin pens was conducted by Krzywon et al. 4 5
`The accuracy of NGFP, FP, SS, and KP was evaluated at
`doses of I, I 0, 30, 40, and 60 U and SS alone at 80 U using
`a total of 1,260 delivered doses . All pens at every dose tested
`were within the predefined ISO limits, and absolute average
`deviation of all insulin pens ranged between 0.09 and 0.81 U.
`The authors concluded that the dosing accuracy was excellent
`for all pens studied and there was no significant difference
`from one pen device to the next. 45
`The aforementioned studies were conducted in controlled
`laboratory settings, by trained professionals. However, when
`patients with or without diabetes, not dependent on insulin
`therapy, and naive to pen device were instructed on FP and SS
`pen use, the results demonstrated that the participants were
`able to administer a 20 U dose accurately. 46 A small amount
`of dosing errors occurred in this study, with less than 2%
`of doses from each pen delivered below the predefined ISO
`limits. 46 Another study in patients with diabetes, with approxi(cid:173)
`mately 90% of pati ents reporting pen device experience,
`found that patients were able to accurately administer six
`different doses (range, 5- 80 U) with the SS pen, with no mea(cid:173)
`surements outside the predefi ned ISO limits. 47 An interesting
`study evaluated the accuracy of administering injections with
`the SS pen under varying temperature conditions from 5°C
`to 4O°C and found the SS pen dosed accurately according to
`ISO standards at I, 40, and 80 U. 35
`All new-generation pens have excellent accuracy in a
`controlled laboratory setting45 and only the SS can claim its
`pen to be accurate under varying temperatures. 35 No accuracy
`studies have been conducted using the NGFP or KP in
`patients with diabetes; however, studies show that patients
`can dose FP and SS accurately. Further studies are needed
`to determine if pati ent administration of insulin using other
`new-ge neration pens impacts their accuracy and/or clinical
`patient outcomes.
`
`Injection force
`I nsu I in pens have grown in favor amongst providers and
`patients for a number of reasons. One of the identifi ed
`qualities affecting patient preference is the amount of force
`
`Medical Devices: Evidence and Research 20 I 0:3
`
`subn1it you r 1n:1nuscrip1
`
`D ove prt:"-"
`
`45
`
`Sanofi Exhibit 2101.005
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`Davis et al
`
`Doveprcss
`
`The most comprehensive study to evaluate the injection
`force ofNGFP compared with the newer generation ofpre(cid:173)
`filled, disposable insulin pens was conducted by Asakura
`et al. 48 A head-to-head comparison of NGFP, SS, and KP
`was conducted at 3 constant injection speeds with 2 needle
`si zes of 31 G and 32G. The set injection doses for this study
`were 20 U instead of 60 U doses delivered in the Rissler
`et a] 49 study. Results demon strated similar findings showing
`superiority of NGFP over SS in requiring lower injection
`force by 12%-25 % at both needle sizes and at all injection
`speeds (Tables 2 and 3). NGFP was also superior to KP
`requiring lower injection force by 35%--41 % at both needle
`sizes and at all injection speeds (Tables 2 and 3). This study
`also confirmed that increasing the injection speed and gauge
`si ze of the needle significantly increased the injection force
`with all pens tested. 48
`Head-to-head comparisons of current insulin pen devices
`demonstrate clear superiority ofNGFP to either SS or KP in
`requiring lower injection force needed to deliver a set dose
`of insulin. 48
`9 Because injection force has been described in
`.4
`patient survey data as a factor affecting satisfaction, it may
`be inferred that improving this could result in improved
`patient preference. Further studies are needed to determine
`if improved injection force with the NGFP improves clinical
`patient outcomes.
`
`Patient-focused perspectives
`Pati ent perceptions of injection force, ease of use, product
`identification, and handling can influence pen preference.
`Only 2 studies evaluating patient preference ofNGFP com(cid:173)
`pared with other pen devices have been completed. Patient
`perceptions of injection force were evaluated in the study
`by Pfutzner et al. 43 Fifty patients with type 2 diabetes who
`had insulin pen experience were asked to complete a survey
`after delivering I dose each of 20, 40, and 60 U with NGFP
`and FP devices into an injection pillow in a randomized
`
`Table 2 Comparative injection force (N) at various speeds using
`a 3 1 G pen needle and NG FP, SS, and KP insulin pens48
`
`Pen device
`
`Next Generation FlexPen'·'
`SoloSTAR'
`
`KwikPen'
`
`Speeds of injection (mm/s)
`8,3
`5
`10.7 ± 1.4
`13.3 ± 0.8
`16.9 ± 1.2
`
`3.3
`
`8.1 ± 0.7
`9.2 ± 0.5
`12.5 ± 1.6
`
`15.6 ± 0.9
`20.7 ± 2.4
`24.5 ± 2.6
`
`necessary to inject insulin through a pen device (injection
`force). 48 It has been established that pen devices require less
`injection force to deliver an equivalent dose, in general , than
`insulin syringes because of the wide bore associated with
`the pen needles. 49 Injection force, measured in Newtons (N),
`is determined in clinical tria ls by mounting pens ready to
`deliver a set dose on a testing machine that is programmed
`to deliver the dose and depress the push button at a set speed.
`Injections are made into cushions designed to mimic adipose
`tissue in climate-controlled laboratories. 48 Injection force has
`been frequently validated with older insulin pens, and design
`modifications have been made that enable pens to require
`less injection force while maintaining accuracy. After the
`emergence of the SS insulin pen, a study comparing it with
`FP found that the FP required higher injection forces than SS
`to deliver equal amounts of insulin for numerous doses and
`speeds. 35 This validated survey data describing complaints
`from patients with manual dexterity problems in depressing
`the push button to deliver insulin doses with the FP. 43
`50
`-
`Thus, the NGFP was developed with a goal of requir(cid:173)
`ing lower injection force for dose delivery to improve user
`satisfaction .48 Three studies have evaluated the injection
`force ofNGFP compared with FP and other new-generation
`pens. 43,48 ,49
`
`A study comparing the injection force ofNGFP with its
`predecessor, the FP, evaluated 20 pens using a standard flow
`rate of IO U/s with I standard needle size (30G). NGFP was
`superior to FP with a relative reduction in injection force
`by 29.8%. 43
`The injection force of NGFP was also compared with
`SS in delivering a dose of60 U at 3 constant injection speeds
`of 4, 6, and 8 mm/s. 49 Twenty-four pens were tested using
`the following combinations of pens and needle sizes: NGFP
`with a 32G needle, SS with a 32G needle, and SS with a 31 G
`needle . Various injection speeds were evaluated to mimic the
`possible range of injection speeds at which a patient may
`perform self-administration of insulin , and two needle sizes
`were used to examine any influence of injection force from
`needle bore size. The NGFP with a 32G needle had signifi(cid:173)
`cantly lower mean injection force compared with SS with
`either a 32G or 3 I G needle (P < 0.000 I). Over the range
`of injection speeds, the NGFP with 32G needle reduced the
`injection force by 18%- 28% compared with the SS with
`32G needle and by 36%--45 % compared with the SS with
`31 G needle. Therefore , using a smaller-gauge (32G) needle
`reduced the injection force to the greatest extent and that
`increasing the injection speed required higher injection force
`using either pens. 4
`'1
`
`Note: All values are given as mean ± SD.
`Abbreviations: N, Newton ; NGFP, Next Generation FlexPen; 55, 5olo5TAR;
`KP, KwrkPen.
`•P < 0.05 for all comparisons made between NGFP and KP; bp < 0.05 for all
`comparisons made between NGFP and 55 .
`
`46
`
`~ubmi l you.- m,u11ncript
`
`DoveprC'o'.-.
`
`Medica l Devices: Evidence and Research 20 I 0:3
`
`Sanofi Exhibit 2101.006
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`Dovepre~s
`
`Next generation of insulin pen devices
`
`T abl e 3 Comparative injection force (N) at various speeds using
`a 32G pen needle and NGFP, SS, and KP insulin pens••
`Pe n device
`S peeds of inje ction (mm/s)
`3.3
`8.3
`
`5
`
`Next Generation Fl exPe n''
`
`SoloSTAR'
`
`Kwik Pen'
`
`5.7 ± 0.4
`6.7 ± 0.3
`9.1 ± 1.3
`
`8.2 ± 0 .7
`10.4 ± 2.1
`
`13. 1 ± 0 .8
`
`12.7 ± 0.5
`16.3 ± I. I
`2 1.6 ± 2.0
`
`Note: All values are given as mean ± SD.
`Abbreviations: N, Newton; NGFP, Next Generation FlexPen; 55, 5olo5TAR; KP,
`KwikPen.
`•p < 0.05 for all comparisons made between NGFP and KP; ' P < 0.05 for all
`comparisons made between NGFP and SS.
`
`order. Significantly more patients rated the inj ection force as
`"good" or " very good" using the NGFP at all 3 tested doses
`of 20, 40, and 60 U compared with the FP (80%, 72%, and
`38% vs 48%, 32%, and 20%, respective ly, P < 0.0001). In
`addition , significantly more patients found the NGFP to be
`"simpler and more comfortab le to use" than the FP (76% vs
`24%, respectively, P = 0.0002). 43
`Sixty-four patients with type I or type 2 diabetes were
`enrolled in a survey study to eva luate the visua l appearance
`and perceptions ofNGFP compared with FP. 50 All patients
`were shown the range of FPs and NGFPs prefill ed with
`three types of analogs (insulin aspart, insuli n detemir, or
`insulin aspart protamine/aspart mix) along with their packag(cid:173)
`ing and asked to answer nine survey questions corresponding
`to their ability to identify the type of in sulin . Patients were
`also asked to attach a NovoFine® (Novo Nordisk) need le on
`the FP and NGFP, then attach a Novo Tw ist needle on NGFP
`(NovoTwist is not compatible with FP), and answer three
`survey questions about the ease of attaching the needle .
`Finally, the patients were asked to inject 1 dose of 20, 40,
`and 60 U of insulin detemir into an injection pillow and
`were randomized to inject either FP fo ll owed by NGFP or
`NGFP fo ll owed by FP. They the n answered 22 additiona l
`survey questions relating to the injection force and device
`handling. Si gnificantly more pati e nts found the insu lin ana(cid:173)
`log type with the NGFP was easier to identify with rega rd to
`labeli ng (P < 0.00 1 ), packaging (P < 0.001 ), and cartridge
`(P < 0.001 ) compared w it h the FP. Significant ly more
`patients rated that attaching the NovoTwist need le on NGFP
`was eas ier than attaching a NovoFine need le on either FP or
`NG FP (P < 0.001 ), and significantly more patients preferred
`using the Novo Twi st need le over the Novo Fine need le with
`th e NGFP (77% vs 6% respectively, P < 0.00 l ). The NGFP
`was eas ier to inject than FP at all 3 do ses of 20, 40, and 60
`U of insulin (P < 0.001 ). In addition , significantly more
`pati e nts beli eved it was "easy" or "very easy" to push down
`the inj ecti o n butto n on the NGFP compa red with the FP for
`
`the 20 U (9 1 % vs 67% respectively, p < 0.00 l ), 40 U (72%
`vs 22% respect ive ly, P < 0.00 I), and 60 U doses of insu lin
`(38 % vs 2% respective ly, P < 0.00 l ). More patients rated
`the NGFP as " very easy" for overa ll use (P < 0.00 I) , the
`most convenient pen (P < 0.00 I), a nd the simplest pen to
`use (P < 0.00 I) compared with th e FP. Patients were more
`confident th at th e fu ll do se of in sulin was delivered using
`th e NGFP than with the FP (P < 0.00 I) . Accordingly, 83%
`of pat