`
`Journal of Controlled Release 49 (1997‘) 1857192
`
`
`
`iournal of
`controlled
`release
`
`
`
`Design of ocular/lacrimal and nasal systems through analysis of
`drug administration and absorption
`
`Yoshie Maitania, Tsuneji Nagaia, Kyle Kolliasb, Nikolaos A. Peppaswk
`“Department of Pharmaceutics, Hoslu' Univetst'or, Ebara 2—441, Shinagawa—kn, Tokyo 142, Japan
`hSchool of Chemical Engineering, Purdue University, West Iafayerte, IN 47907—1283, USA
`
`Received 30 October 1996', received in revised form 20 March 1997; accepted 14} April 1997
`
`
`
`A modeling analysis is presented of drug or peptide absorption and administration via the ocular, naso—lacrimal duct, and
`nasal routes. This method accounts for the fast absorption and retention of drug in the blood after administration. A lcey
`parameter in this process is the ratio of drug absorption rate in the conjunctival mucosa to the drug transfer rate from the
`naso—lacrimal duct to the nasal mucosa. This ratio depends on the polymer carrier of the formulation. Predicted values of
`drug concentration in the blood can be used to design new formulations (delivery systems) which will lead to long residence
`time or fast drug absorption. © 1997 Elsevier Science B.V.
`
`Keynmrds: Nasal delivery; Ocular delivery; Naso—lacrimal duct; Nasal enhancers
`
`
`1. Introduction
`
`Peptide and drug delivery through the ocular [1,2]
`and nasal [3,4] routes of administration has increased
`in recent years. Of particular interest is the delivery
`of peptides and proteins through these routes, as it is
`known that peptidase activity is reduced or mini
`mired in the nasal cavity or
`the eyes [5],
`thus
`allowing a significant absorption of the bioactive
`agent. Controlled release formulations for ocular or
`nasal delivery have been reported [6—81. Such sys-
`tems are based on hydrophilic or hydrophobic poly—
`meric carriers, usually in the form of micro- or
`uanoparticles.
`Nasal administration often requires the use of an
`that will promote increased absorption
`
`*Corresponding author.
`
`0l68-3659l97l$17.00 © 1997 Elsevier Science B.V. All rights reserved.
`PH SOl68-3659(97)00074-6
`
`[9,10]. Despite the significant interest in this field,
`little has been proposed in relation to the description
`of peptide and protein absorption simultaneously by
`the ocular/iacrimal and nasal routes [1,11,12].
`
`A predictive method was developed to estimate
`the systemic concentration of absorbed drug or
`peptide following ocular, naso—lacrimal duct, or
`nasal administration. This procedure requires the
`determination of seven kinetic constants. In a previ—
`ous contribution [13},
`these constants were calcu-
`lated based on a first order absorption process in
`order to fit experimental data with this model.
`Although several permeation studies of peptide
`transport using excised mucosa have been reported,
`and some studies of
`in vivo ocular and nasal
`administration are available, there is no report about
`
`the total drug absorption from the eye to the gastrofi
`intestinal
`(GI)
`tract.
`In this work, we evaluated
`
`AQUESTIVE EXHIBIT 1045 page 0001
`
`

`

`Y. Mairam' et al. I Journal of Controlled Release 49 (I997) 185— 192
`
`possible methods of deciding the administration
`route and type of formulation by obtaining indepen-
`dent values of drug permeability coefficients. The
`pharmacokinetic model presented here takes into
`consideration several aspects of the physiological
`characteristics of the system and provides a better
`identification of the important design parameters of
`such systems.
`
`2. Theoretical analysis
`
`We consider the administration of a peptide or
`drug in a formulation that may or may not contain a
`polymeric carrier and is placed in the ocular cui—de-
`sac. A compartmental model can be defined as
`shown in Fig.
`1. The main compartments of this
`model are the eye with drug concentration CE,
`the
`nasal cavity with concentration cN, the GI tract with
`concentration cG, and the blood with drug con-
`centration CD.
`Drug absorption can be defined by seven kinetic
`constants describing various transport processes in
`the biological system. Direct drug transport from the
`eye to the circulation is described by k1. Similarly,
`transport from the nose and GI tract to the circulation
`
`is described by k2 and k3, respectively. A relatively
`high flow-rate is characteristic for transport of a drug
`from the eye to the nose;
`it
`is described by a
`translocation kinetic constant k4, An additional kinet-
`
`eye
`
`blood
`
`“fl“
`
`GI tract
`
`1. Schematic representation of a four-compartment phar-
`macokinetic model for drug administration by the ocular, naso—
`lacriinal duct and nasal routes.
`
`ic constant for transport from the nasal cavity to the
`GI
`tract
`is k5. Finally,
`two kinetic constants are
`defined, k6 and k7, to characterize the decomposition
`of peptides in the conjunctivai and nasal mucosae,
`respectively. A final decomposition kinetic constant,
`kg, for the circulation will be incorporated in the
`latter stages of this analysis.
`Ordinary differential equations may describe all
`the processes using first order kinetic expressions.
`Thus, the drug concentration in the eye is expressed
`as
`
`ch
`—E=klcE+k4cE+kficE
`
`The nasal concentration is expressed as
`
`ch
`dt
`
`= — k4cE + k5cN + 1(7ch + k7cN
`
`The GI tract concentration is
`
`doc
`— d: = —k,c,, +k3cG
`
`(l)
`
`(2)
`
`(3)
`
`Finally, the drug concentration in the circulation can
`be expressed as
`
`ch
`—‘“£F= mklcfi+kch *k3cG
`
`(4)
`
`Solution of the above four differential equations
`has been obtained for initial conditions representing
`the simultaneous drug administration in the ocular
`and nasal cavities, with initial concentrations CEO and
`CNo’ respectively. It is of course assumed that the
`initial drug concentration in compartment G is cG0 =
`0.
`
`is
`it
`To obtain the solutions of these equations,
`helpful to define two kinetic constants k10 and k” as:
`
`k”, = k, + k4 + k,
`
`and
`
`k,,=t,+k,+k,
`
`(5)
`
`(6)
`
`
`
`Then, the drug concentration in the eye is obtained
`by solution of Eq. (1) with the previous boundary
`conditions as:
`
`CE : CaneXM—kioT)
`
`
`AQUESTIVE EXHIBIT 1045 page 0002
`
`
`(7)
`
`

`

`
`
`Y. Mancini e! at. I Journal of Controlled Release 49 {1997) 185—192
`
`From Eqs. (7) and (2), the nasal drug concentration
`may be calculated as
`
`B = —————r—-————k2k“
`(k1 _" k5 "kioxktl _ k5)
`
`k4CE‘U
`CN : wn—Wm_ k“)exp(rkwt)
`+ [0N0 v Manama]
`
`kacao
`
`C = a A - B
`
`and
`
`D =L
`(k2 _k10)
`
`(8)
`
`l3?
`
`(13)
`
`(14)
`
`(15)
`
`Obviously, when the initial nasal concentration is
`zero, Eq. (8) takes the classic form
`
`17¢,ch
`cs : miner—m r wan—k] 11)]
`
`(9)
`
`This completes the development of the model. The
`individual kinetic constants
`take various values
`
`depending on whether absorption of peptides or
`conventional drugs is considered.
`
`Setting now Eq. (8) into Eq. (3), we obtain the
`following solution for the drug concentration in the
`
`3. Model discussion
`
`I‘c4k5c‘E0
`( k r)
`: ——-—-w~——“_6X fi
`(kit W k10)(k3 _ J1‘10)
`10
`p
`k5
`[
`WE"
`m_.,_m_3x _
`_._u_
`(k3 _ kit)
`(k11_ kin)
`p
`EN”
`k5
`k4kscnt,
`+
`>< [ch -m] ]exp(k3r)
`
`( k n]
`H
`
`(kn ‘- k10)(k3 M km)
`
`0‘3 "” kill)
`
`k4cED
`
`(10)
`
`(7), (9), (10) the drug con~
`Finally, from Eqs.
`centration in the circulation, CD, can be calculated
`using Eq. (4). The final expression is as follows
`
`m kzka + 1510611 '" km)
`(kll _ k10)k1n
`
`Brim—km!)
`
`k2k4
`_(k11 W k5 "" kioxku _ k5)
`
`X (”PU—kn + [(5);] +
`
`162164 + k1(k“ — km)
`(1611 ”k1o)k:o
`
`lcflk,6
`_
`k4;
`(k11_ k5 _ kioxkn “‘15)
`0‘2 T km)
`
`=A exp(wk,,,t) + B exp[(——kE1 + k5)t} + C
`+1):
`
`(11)
`
`M kakd + kiUCu T km)
`(k11_k10)k10
`
`(12)
`
`3.1. Values of kinetic constants
`
`The compartmental model developed above can be
`used to investigate the influence of various processes
`on the absorption and distribution of drugs adminis-
`tered by the ocular/lacrimal and nasal routes. In this
`analysis, use was made of previous experimental
`studies in order to identify acceptable ranges of
`values of the kinetic constants.
`
`The kinetic constant of drug transport from the eye
`to circulation, k1, usually takes values of 20m50 h"1
`for fast releasing (hugs, although for slowly releasing
`drugs it may be as low as 1 h_I [14]. Transport from
`the nasal cavity to the circulation is significantly
`slower with 1:2 of about 0.2 to 0.5 h_1 [14}. Poorly
`absorbed drugs are characterized by very small
`values of the kinetic constant of transport from the
`GI tract to the circulation, of the order of k, 2 10’3
`to 10’4 n“.
`
`the
`is
`An important parameter of this model
`translocation rate constant, k,,, which characterizes
`transport from the eye to the nose [15}. Typical
`values of this constant are from 5 to 10 h_1. The
`kinetic constant for transport to the GI tract, k5,
`is
`known to be approximately zero for peptides due to
`peptidase action in the gastric area. However, values
`of 0.1—1 hil may be more appropriate for conven—
`tional drugs. Finally,
`the kinetic constants for drug
`degradation k6 and k7, can be calculated from the
`recent analysis of Lee et a1.
`[16]. Typical values
`range from 0.5 to 3 h”.
`
`AQUESTIVE EXHIBIT 1045 page 0003
`
`

`

`
`
`
`
`
`
`:=~=.m.'.'4m
`
`
`
`
`
`
`Fig. 2. Normalized drug concentration in the blood (curve 1), eye
`(curve 2), nasal cavity (curve 3) and GI tract (curve 4) as a
`function of time for k =02 h ‘ 1, =10“ h ’ ,k =5 11',
`k 0111k —.0111,,,l<"=-.‘05h',andk,=10h1 (10,1; =20
`hi“,,'(b)ork=50h (c.)
`
`AQUESTIVE EXHIBIT 1045 page 0004
`
`1:
`
`5ra
`
`Y. Mairarrr' er (1!. I Journal of Controlled Release 49 (1997) 185— 1'92
`
`O {0
`
`
`NormalizedConcentration.005.1059
`CLIC-E
`
`NormalizedConcentration.
`
`
`NormalizedConcentration,[/05
`
`1.0
`
`0.9 g
`0.8
`0.7
`
`0.6
`0.5
`0.4
`0.3
`
`0.2 I
`0.1 -
`0.0 15—5
`0.0
`2.0
`
`4.0
`
`6.0
`
`8.0
`
`10.0
`
`Time (h)
`
`Values of these constants were also calculated for
`the systems of interest, i..,e for insulin delivery. The
`value of the insulin degradation constant k6 was
`calculated as 0 146 h from the time of loss of 10%
`insulin activity in conjunctiva] homogenates of the
`albino rabbit, according to the data of Hayakawa et
`[17]. Then, from the value of the insulin pe1-
`meability coefficient1n1 the eye, the value of k1 was
`calculated as 0 132 h for insulin,2when l"E =4.6X
`106,the conjunctival area is 8 cm2 and theE volume
`
`The nasal constants were calculated from recent
`the insulin permeability in the nasal
`mucosa, PN=8.56><10'6 cm 8—:
`according to
`Maitani et al.
`[18} with a nasal mucosa area of 61
`cm2 and a volume of 6 in]. Then, the value of k2 was
`calculated as 1.88 h when the nasal mucosa area is
`61 cm2 and the volumeIs 1 ml [19]. The value of the
`drug degradation constant A: was determined as
`0. 832 h_ from the values for 50% of
`insulin
`remaining in homogenates of nasal mucosa [20].
`Finally, as reported [5} before, k6<k7 for insulin and
`most other peptides.
`
`3.2. Computer simulations
`
`A number of computer simulations were carried
`out in order to examine the importance of the various
`transport processes on the overall absorption of drug
`in the blood In a first set of experiments, values of
`k =20h1,,k =02h ,,k =10 h" andk4=5
`h’1 were selected, and k5, k6 and k7 were varied to
`examine the effect of drug decomposition on the
`overall absorption.
`All results are expressed as normalized concen-
`tration in each compartment with respect to an initial
`dose in the eye, CE". Initial values in the nose and GI
`tract were assumed zero. Fig. 2(a, b, and c) shows
`the change of peptide concentration in the blood as a
`function of time (curve 1). These are typical data for
`insulin. 'A relatively fast absorption is observed
`passing through a maximum at about 20 min when
`k,=20 h”1 (Fig. 2b). The ocular insulin concen-
`tration is depleted very fast (curve 2), whereas the
`nasal drug concentration passes through a maximum
`(curve 3). Finally,
`the GI
`tract concentration is
`virtually zero (curve 4). The values of k6 were varied
`
`

`

`Y. Maimni er al.
`
`I Journal of Controlled Release 49 (1997) 185—192
`
`189
`
`from 0.1 to 0.45 h_1 (always k6<lc7 according to
`Yamamoto et a1. {SD indicating that there was only a
`influence of the insulin degradation on the
`overall insulin concentration in the blood. As Yama-
`
`moto et al. [5] have indicated, the rank of proteolytic
`activity among the mucosa] homogenates decreased
`from the nasal, k7, to the rectal, ilea, vaginal, to the
`conjunctival area, k6. Therefore, these results are in
`agreement with biological observations.
`The influence of R, was further seen by running
`simulations for various values of the kinetic constant,
`
`k, for drug transport from the eye to the blood. Fig.
`2(b and 0) show the drug absorption for typical
`values of the kinetic constants with changing time. It
`is noted that the values of chcEfJ and cD ICED depend
`on the value of km, as seen in Eqs. (9) and (11).
`Therefore, an increase of the value of kE (appearing
`in km) induces the exhibition of maxima in curves 1
`and 3 at early times. Also, the maximum value of CD
`depends on the third term, C, of Eq. (11). On the
`other side, the slope of the elimination curves after
`the peak depends on the fourth term, D, of Eq. (11).
`However, the translocation constant, k4, seems to
`play a very important
`role in drug and peptide
`transport. Fig. 3(a and b) summarize data of drug
`transport
`for values of k423 h‘1 and 10 h“,
`respectively. Clearly,
`in the first case the peptide
`concentration in the blood reaches a higher level
`faster, whereas when k4: 10 h_1 a significant por-
`tion of the peptide has passed into the nasal cavity
`leading to an increased nasal concentration (curve 3).
`Further investigation of the effect of the transloca-
`tion constant on the drug or peptide transport could
`be obtained by calculating the half—life, 71,2, of drug
`in the blood as the time at which the concentration
`was 50% of the initial concentration. Fig. 4 shows
`the half~life in the blood as a function of k4 for
`different values of k,. Clearly, fast absorption is
`achieved as k4 increases and k1 decreases, indicating
`that the nasal route has a great potential for fast
`transport.
`To further analyze this behavior, the half-life was
`determined as a function of k,, as shown in Fig. 5.
`Clearly, as the drug absorption constant, k,,
`in—
`the half-life increased. The translocation
`constant, k4, is important in this process. As shown
`in Fig. 5, an increase of k4 led to a significant
`decrease of the half-life. When k4 increases,
`the
`
`L0
`
`0.9
`
`0.8;
`0.7
`
`0.6
`
`0.5
`0.4
`
`0.3
`0.2
`o,
`
`' ‘5.—0.0
`'—
`u.0
`2.0
`
`
`
`NormalizedConcentration.C's/C's
`
`.-.
`
`4.0
`
`6.0
`
`8.0
`
`10.0
`
`Time (h)
`
`
`
`
`
`NormalizedConcetratlon.CLJCE
`
`
`
`'u.o
`
`1.0
`
`2.0
`
`3.0
`
`Time
`
`(h)
`
`Fig. 3. Normalized drug concentration in the blood (curve 1), eye
`(curve 2), nasal cavity (curve 3) and GI tract (curve 4) as a
`function of time for k, =20 11%, k2 m0.2 h”, k3=10fll hi', k, =5
`hfi‘,k,w().lh'1,k6=0.1h’i,k,=0.5 a“, and k,:3 h" (a) or
`10 ifl (b).
`
`half-life becomes
`smaller,
`i.e., nasal
`absorption
`appears to dominate the increase of CD. This is due to
`the fact that k”, is included in constants km and D of
`Eq. (11), whereas the latter has a negative value.
`Fig. 6 summarizes the importance of the translo-
`cation process
`(constant k,) and the nasal drug
`absorption on the overall drug transport. A maximum
`in the half-life was observed. This was particularly
`important in peptides that have low mucosai per-
`meability. A further understanding of the importance
`of k2 was obtained from Fig. 7 where the half-life
`was plotted as a function of k2 and the ratios of
`k,/k4.
`
`AQUESTIVE EXHIBIT 1045 page 0005
`
`

`

`
`
`Y. Mairani et a1. / Journal of Controlled Release 49 (1997) 185—192
`
`HalfLife,’92(h)
`
`5
`
`IO
`
`15
`
`20
`
`TranslacatlonConstant, k+ m“)
`Fig. 4. Half-life of drug transport in the blood as a function of
`translocation constant, k_,, for various values of 1:1 (in h”), when
`k2=0.2 h", 19:10” h", k5=0.1 a", 1:50.] h", k,:0.5
`
`M
`
`”a,“(n)
`
`HalfLife,
`
`o
`
`2
`
`4
`
`s
`k,_ “1“)
`
`s
`
`10
`
`12
`
`Fig. 6. Half~life of drug transport as a function of k2 for various
`values of k4 when k1 = 10 h‘1 and the other parameters are the
`same as in Fig. 5.
`
`These results indicate that when chug transfer from
`the nose to blood becomes
`fast,
`i.e., when lc2
`increases close to k], then the value of CD increases
`slowly. The associated half~life increases since k2 is
`(kl >> k2) and,
`therefore,
`the increase in k4
`appears to lead to increase of the loss of drug in the
`
`eye, However, beyond a certain value of k2, transfer
`of drug from the nose to the blood becomes faster.
`Thus, the drug penetrates through the nasal mucosa
`fast, even when the value of k4 is constant. Once CD
`has fast increased due to k2 increase,
`the half-life
`decreases as shown in Figs. 6 and 7.
`Finally, we consider the conditions of drug elimi-
`nation. As discussed earlier, for drug elimination the
`
` o
`
`2
`
`4
`
`a
`
`a
`
`10
`
`12
`
`4;
`
`20
`
`40
`
`so
`
`so
`
`100
`
`120
`
`kl
`
`(tr' )
`
`kZ m")
`
`Fig. 5. HaEf-life of drug transport in the Mood as a function of k,
`for various vatues of R4 (in h_1), when k2=0.2 h", @210"|
`h", kd=5 h“, k,:0.t ir', k6=0.t a", and k,=o.5 h”.
`
`'7. Half-life of drug transport as a function of k2 for various
`Fig.
`vatues of k,/k,,, when the other parameters are the same as in Fig.
`5.
`
`AQUESTIVE EXHIBIT 1045 page 0006
`
`

`

`Y. Maimni er a]. I Journal of Controlled Release 49 (1997) 1'85— l92
`
`191
`
`.0
`
`P P
`
`Nonnalized-Concemration D
`
`Time(h)
`
`Fig. 8. Normalized ding concentration in the blood (curve 1), eye (curve 2), nasal cavity (curve 3) and GI tract (curve 4) as a function of
`time for k3:5 h"1 when the other parameters are the same as in Fig. 23.
`
`constant k8 can describe drug transport from the
`blood. In this case, Eqs. (1)43) are still valid, but
`Eq. (4) becomes:
`
`is not the best delivery method. Instead, use of a
`viscous delivery carrier (low k4 value) is a much
`more efficient method of nasal delivery.
`
`~77: *kICE-kch—kflGJrkscD
`
`(16)
`
`With this equation, the solution of Eqs. (8)—(IO) is
`not affected. Eq. (11) is modified to incorporate the
`term k8. For example, the insulin elimination half—
`life is 10 min with k8 24.2 1171. Thus, a simulation is
`shown in Fig. 8 for [(8 =5 11—1. The systemic insulin
`concentration, CD, is fast increased, followed by drug
`elimination.
`
`4. Conclusions
`
`A predictive method was developed for estimating
`the systemic concentration of diugs absorbed foilow—
`ing ocular, nasomlacrimal and nasal administration.
`This procedure required estimates of seven kinetic
`constants. These parameters were calculated based
`on fitting experimental data based on first order
`absorption processes. The results indicate that use of
`a nasal enhancer (increase of parameters 1kl and k2)
`
`Acknowledgements
`
`This work was supported in part by a grant from
`the Nagai Foundation, Tokyo to N.A. Peppas and Y.
`Maitani.
`
`References
`
`[l] A. Yamamoto, A.M. Luo, S. Doddakashi, Vii-LL. Lee, The
`ocular route for systemic insulin delivery in the albino rabbit,
`J. Pharmacol. Exp. Ther. 249 (1989) 249455.
`{2] MP. Saettone. B. Giannaccini, F. Barattini, N. Tellini, The
`validity of rabbits for investigation of ophthalmic vehicles: A
`comparison of four different vehicles containing tropicamide
`in humans ad rabbits, Pharm. Acta. Helv. 57 (i982) 4755.
`[3] 3.}. Aungst, NJ. Rogers, E. Shefter, Comparison of nasal,
`rectal, sublingual and intramuscular insulin efficacy and the
`effects of a bile salt absorption promoter, Exp. Ther. 244
`(1988) 23—27.
`[4] DJ. Pillion, J. Recchia, P. Wang, ill. Marciani, CR. Kensil,
`
`AQUESTIVE EXHIBIT 1045 page 0007
`
`

`

`Y. Maitani e! a]. 1 Journal of Controlled Release 49 (I997) [85n1'92
`
`DS~i, a modified quilaja saponin, enhances ocular and nasal
`absorption of insulin, J. Pharm. Sci. 84 (1995) {276—1279.
`[5] A. Yamamoto, E. Hayakawa,V.H.L. Lee, Insuiin and proin-
`sulin proteoiysis in mucosal homogenates of the aibino
`rabbit: implications in peptide delivery from normal routes,
`Life Sci. 47 (E990) 246572474.
`[6] MR Saettone, B. Giannaccini, S. Ravecca, F. Lamrca, G.
`Total, Poiymer effects on ocular bioavaiiability. The influence
`of different iiquid vehicles on the mydriatic response and
`tropicamide in humans and in rabbits, Int. J. Pharm. 20
`(1984) 187—202.
`
`{7] NM. Davies, 5.]. Fair, I. Hadgraft, LW. Kellaway, Evalua-
`tion of mucoadhesive polymers in ocular drug delivery. I.
`Viscous solutions, Pharm. Res. 8 (1991) 1039—1043.
`[8] NF. Farraji, B.R. Jottaitscn, SS. Davis, L. Illutn, Nasal
`administration of insulin using bioadhesive microsptteres as a
`delivery system, J. Control. Release 13 (1990) 253—261.
`[9] Y. Maitani, T. Yamamoto, K. Takayama, T. Nagai, The effect
`of soybean-derived sterol and its gincoside as an enhancer of
`nasai absorption of insulin in rabbits in vitro and in vivo, Int.
`J. Pharm. 117 (1995) 129—137.
`
`[10] N. Uchida, Y. Maitani, Y. Machida, M. Nakagaki, T. Nagai,
`Influence of bile salts on the permeability of insulin through
`the nasal mucosa of rabbits in compatison with dcxtran
`derivatives, Int. J. Pharm. 74 (199i) 95%03.
`[I I] T.F. Patton, LR. Robinson, Quantitative precomeat disposi-
`tion of topically applied pilocarpine nitrate in rabbit eyes, J.
`Pharm. Sci. 65 (1976) 1295—1301.
`{12] "RE. Hilditeh, C.S. Kwok, LA. Amanat, Lacrirnai scintig—
`raphy. I. Compartmentat analysis of Liata, J. Ophthalrnol. 67
`([983) 713—719.
`
`{13] Y. Maitani, T. Yamamoto, K. Takayama, NA. Peppas, T.
`Nagai, A modelling analysis of drug absorption and adminis-
`tration from the ocuiar, nasoilacrimal duct, and nasal routes
`in rabbits, Int. J. Pharm. 126 (1995) 89—94.
`{14] I. Gouda, E. Gipps, Modci of disposition of drugs adminis—
`tered into the human nasal cavity, Pharm. Res. 7 (1990)
`69—75.
`
`in: Y.W.
`[15] A.A. Hussain, R. Bawarski—Nassar, CH. Huang,
`Chien (Ed), Transnasa] Systemic Medications, Elsevier,
`Amsterdam, 1985, pp. 1219137.
`to peptide and protein
`[16} V.H.L. Lee, Enzymatic barriers
`absorption and the use of penetration enhancers to modify
`absorption,
`in: 5.5. Davis, L. IlIum, E. Tomiinson (Eds),
`Delivery Systems for Peptide Drugs, Plenum, New York,
`1986, pp. 87—104.
`[17] E. Hayakawa, D. Chien, K. Inagaki, A. Yamamoto,W_ Wang,
`V.H.L. Lee, Conjuctivai penetration of insulin and peptide
`drugs in the albino rabbit, Pharm. Res. 9 (i992) 769—775.
`[18} Y. Maitani, S. Asano, S. Takahashi, M. Nakagaki, T. Nagai,
`Pcrmeabiiity of insulin entrapped in liposomc through the
`nasal mucosa of rabbits, Chem. Pharm. Bull. 40 (1992)
`156971572.
`
`[19} S. Gizurarson, The relevance of nasal physiology to the
`design of drug absorption studies, Adv. Drug Deliv. Rev.
`[1
`(1993) 329.341
`
`[20] WJ. Irwin, A.K. Dwivcdi, RA. Holbrook, MJ. Dey, The
`effect of cyclodextrins on the stability of peptides in nasal
`enzyrnic systems, Pharm. Res. 11 (1994) E698—1703.
`
`
`
`AQUESTIVE EXHIBIT 1045 page 0008
`
`