127
veviews
Responsivepolymersystemsfor rolleddeliveryof therapeutics Joseph Kost and Robert Langer The ideal drugdelivery system should provide therapeutics in response to physiological requirements, having the capacity to ‘sense’ changes and alter thz drugrelease process accordingly. Such responsive controlled delivery systems are still at an experimental stage. This review focuses on two basic approaches: (1) externally regulated
systems (utilizing triggers
such as magnetism,
ultrasound,
temperature and electricity), and (2) self-regulated systems (utilizing pksensitive polymers, enzyme-substrate
reactions, competitive binding, and antibody inter-
actions). The basic prcmisc that drug conccmration-ctct relationships in man arc more-or-less invariant as a function of time has led to the dcvclopmcnt of coustant-natcdrug-dciivcry systcux’, zud thcrc has been incrcascd activity in this am ovrr the past few ycnrs’. Thcsc dcviccs (which arc already svailablc COIIIIIIC~cially) CIII maintain ths drug in the dcsircd rhcmpcutic range with just a single dose, and 10~31~7~ dclivcry cf the drug to a particular body compartulcnt, which IOWCP the systemic &ug I~vcl. thus reducing the need for follow-up cart, pmcrving tnsdications that arc rapidly destroyed by the body, and increasing pnticut coFfcrt and/or improving compliance. Ncvcrchclcss, thcrc arc a number ofchnical situations for which such an approach is not the best answer. These include the dclivcry of insulin for patients with diabctcs rncllitus, antiarrhythmics for paticnts with heart-rhythm disorders, gastric-acid iuhibitors for ulcer control, uitratrs for patients with anginn pcctoris. as well as for USCin sclcctivc P-blockade, birth control, general hormone replaccrncric, immunization and cancer chcmotherapy. Rcccnt studies in the field of chtonopharmacology indicate that the omcts of certain discasts exhibit strong circadian temporal dcpcndcncy. Thus, drug-dclivcry patterns can be fkther optimized by pulsatitc or self-rcgutatcd delivery, adjusted to the staging of biological rhythms”. In rcccnt years, s&ml rcscarch groups have been dcvcloping rcsponsivc systems that more-closely rcseniblc the non:*nl &....L-’ ._. yrSJJS~q$cai process in which tlic miouut of drug rcicascd accords with physiolo$cnl
0 1992. Elsetier Science Publishers Lid (UK)
needs’. The responsive polymeric delivery systems can bc classified as closed-loop or open-loop system. III ctoscd-loop control systcmn, the controlled variable is detcctcd and, as a result, the system output is adjust4 accordingly. Howcvcr, in npw-loop systcrns, it&xmatiou about thr controltcd variable is not ~~ccessarily used to acljust the system inputs. In the controlleddrug-dclivcq CcId, open-loop systems ;I~Ck~iirwr~ as ‘p&tile or ‘caccrnally rcgutatcd’. and the ctoscdloop systems a~ ‘srtt^-rcgulatcd’. The cstcrnalty controlled devices zppiy cstcmnl triggers, such as tnagnctic, ultrasonic. thcmmal, clcctric and in-adiatkm, fix pulsntilc dclivcry while in the self-regdntcd dcviccs. the rslcnsc rate is controlled by feedback information. without any external intervention. The self-rcgulatcd systems utilizr scvcral approaches as rate-control ~nccha~~isn~s, such as pH-sensitive polymers, cnzymcsubstrate reactions, pH-scnsitivc drug solubility, cnmpctitivc binding, antibody interactions, and metal concentration dcpcndcnt hydrolysis. This rcvicw briefly outlines both cstcrnally rcgulatcd and self-rcgulntcd sytcms for drug dclivcry, including some esamptcs ot cxh. Both systems arc still in the cxpcrinicntat stages of devclopmcnt. Externally regulated (pulsatile or open-loop) systems Maprticaliy
triggered systems
For magnetically triggcrcd systems. drug motccth and magnrtic beads arc uniformly distributed within a solid polymeric matrix. Upon erposurc to rnvironIncntai &ids, the drug is released by ditfusion. Howcvcr, the drug is relcascd at a much highhcr rate in the presence ofan csternai oscillating magnetic field (Fig. la). The mqnctic system has been cvaludtcd irl r~itv’. implants comprising cthylcnc vinyl acetate (WAC) TIETECH
APRIL 1992 IVOL 101
128
reviews
Magnetic
tivity in aqueous swelling. The first is based on polymer-rvz,:?s interactions, especially specific hydrophobic/hydrop.llhc1 balancing eficcts and the The secol:d is based configtiration of side SOUPS. on palymcr-polymer interactions in addition to polymer-water interactions. A p&tile drug-release pattern regulated by temperature changes has brcn recently demonsrrated by several group@- lo.
Beads
Time
Time
0
T,
h
Time
T,
Ultrasound
Source
t..::. 0
Drug Dispersed Polymer Matrix
*_‘.I. .. . . . . _..::.‘-..Y:.-. :_.:>::.,y.::.::. . ;. r . y.‘.:::‘::. . . . . ;:..::‘;.:~: .; ; I’;?.
Time
in
-
0
Time
T,
Time
T,
Figure1 Schematic representation of (a) magnetically and (b) ultrasonically triggered controlled release systems. Upon exposure to environmental ffuid at T,, release by diffusion occurs, when exposed to magnetic or ultrasonic external trigger at T, release take; place at a much higher rate.
Electrically co&o/led defivery system Transmembrane solute flux can be modulated by the action of an applied electric field on the membrane and/or dir&y on the solute”. The clccerophorctic migration of a charged macrosolute wiGn a hydra4 mcmbranc results from the combined rcsponsc to the clcctricai forces 01. the sojuts and its associated counterions in tl-ic adjacent clectrolytr solution. A diffcrcnt approach to elec~rochemicnliy controlled rcleasc is based on polymers which bind :md release bionctive compounds in response to ark e!ectrical signalQ. The polymer has two redos states, only one of which is suitnblc for ion binding. Drug ions arc bound in one rcdox state and released fi-om the o&r. The attached electrodes serve to switch the rcdos states and the amount of current passed can control the amoun6 of ions rcleascd. Electrically control!.-d membrane permeabiiity is also of current intC;est in the field of electrically controlled or enhanced transdcrmal drug d&very (c.g. iontophorcsis)‘-‘*‘-‘.
I
veviews
Responsivepolymersystemsfor rolleddeliveryof therapeutics Joseph Kost and Robert Langer The ideal drugdelivery system should provide therapeutics in response to physiological requirements, having the capacity to ‘sense’ changes and alter thz drugrelease process accordingly. Such responsive controlled delivery systems are still at an experimental stage. This review focuses on two basic approaches: (1) externally regulated
systems (utilizing triggers
such as magnetism,
ultrasound,
temperature and electricity), and (2) self-regulated systems (utilizing pksensitive polymers, enzyme-substrate
reactions, competitive binding, and antibody inter-
actions). The basic prcmisc that drug conccmration-ctct relationships in man arc more-or-less invariant as a function of time has led to the dcvclopmcnt of coustant-natcdrug-dciivcry systcux’, zud thcrc has been incrcascd activity in this am ovrr the past few ycnrs’. Thcsc dcviccs (which arc already svailablc COIIIIIIC~cially) CIII maintain ths drug in the dcsircd rhcmpcutic range with just a single dose, and 10~31~7~ dclivcry cf the drug to a particular body compartulcnt, which IOWCP the systemic &ug I~vcl. thus reducing the need for follow-up cart, pmcrving tnsdications that arc rapidly destroyed by the body, and increasing pnticut coFfcrt and/or improving compliance. Ncvcrchclcss, thcrc arc a number ofchnical situations for which such an approach is not the best answer. These include the dclivcry of insulin for patients with diabctcs rncllitus, antiarrhythmics for paticnts with heart-rhythm disorders, gastric-acid iuhibitors for ulcer control, uitratrs for patients with anginn pcctoris. as well as for USCin sclcctivc P-blockade, birth control, general hormone replaccrncric, immunization and cancer chcmotherapy. Rcccnt studies in the field of chtonopharmacology indicate that the omcts of certain discasts exhibit strong circadian temporal dcpcndcncy. Thus, drug-dclivcry patterns can be fkther optimized by pulsatitc or self-rcgutatcd delivery, adjusted to the staging of biological rhythms”. In rcccnt years, s&ml rcscarch groups have been dcvcloping rcsponsivc systems that more-closely rcseniblc the non:*nl &....L-’ ._. yrSJJS~q$cai process in which tlic miouut of drug rcicascd accords with physiolo$cnl
0 1992. Elsetier Science Publishers Lid (UK)
needs’. The responsive polymeric delivery systems can bc classified as closed-loop or open-loop system. III ctoscd-loop control systcmn, the controlled variable is detcctcd and, as a result, the system output is adjust4 accordingly. Howcvcr, in npw-loop systcrns, it&xmatiou about thr controltcd variable is not ~~ccessarily used to acljust the system inputs. In the controlleddrug-dclivcq CcId, open-loop systems ;I~Ck~iirwr~ as ‘p&tile or ‘caccrnally rcgutatcd’. and the ctoscdloop systems a~ ‘srtt^-rcgulatcd’. The cstcrnalty controlled devices zppiy cstcmnl triggers, such as tnagnctic, ultrasonic. thcmmal, clcctric and in-adiatkm, fix pulsntilc dclivcry while in the self-regdntcd dcviccs. the rslcnsc rate is controlled by feedback information. without any external intervention. The self-rcgulatcd systems utilizr scvcral approaches as rate-control ~nccha~~isn~s, such as pH-sensitive polymers, cnzymcsubstrate reactions, pH-scnsitivc drug solubility, cnmpctitivc binding, antibody interactions, and metal concentration dcpcndcnt hydrolysis. This rcvicw briefly outlines both cstcrnally rcgulatcd and self-rcgulntcd sytcms for drug dclivcry, including some esamptcs ot cxh. Both systems arc still in the cxpcrinicntat stages of devclopmcnt. Externally regulated (pulsatile or open-loop) systems Maprticaliy
triggered systems
For magnetically triggcrcd systems. drug motccth and magnrtic beads arc uniformly distributed within a solid polymeric matrix. Upon erposurc to rnvironIncntai &ids, the drug is released by ditfusion. Howcvcr, the drug is relcascd at a much highhcr rate in the presence ofan csternai oscillating magnetic field (Fig. la). The mqnctic system has been cvaludtcd irl r~itv’. implants comprising cthylcnc vinyl acetate (WAC) TIETECH
APRIL 1992 IVOL 101
128
reviews
Magnetic
tivity in aqueous swelling. The first is based on polymer-rvz,:?s interactions, especially specific hydrophobic/hydrop.llhc1 balancing eficcts and the The secol:d is based configtiration of side SOUPS. on palymcr-polymer interactions in addition to polymer-water interactions. A p&tile drug-release pattern regulated by temperature changes has brcn recently demonsrrated by several group@- lo.
Beads
Time
Time
0
T,
h
Time
T,
Ultrasound
Source
t..::. 0
Drug Dispersed Polymer Matrix
*_‘.I. .. . . . . _..::.‘-..Y:.-. :_.:>::.,y.::.::. . ;. r . y.‘.:::‘::. . . . . ;:..::‘;.:~: .; ; I’;?.
Time
in
-
0
Time
T,
Time
T,
Figure1 Schematic representation of (a) magnetically and (b) ultrasonically triggered controlled release systems. Upon exposure to environmental ffuid at T,, release by diffusion occurs, when exposed to magnetic or ultrasonic external trigger at T, release take; place at a much higher rate.
Electrically co&o/led defivery system Transmembrane solute flux can be modulated by the action of an applied electric field on the membrane and/or dir&y on the solute”. The clccerophorctic migration of a charged macrosolute wiGn a hydra4 mcmbranc results from the combined rcsponsc to the clcctricai forces 01. the sojuts and its associated counterions in tl-ic adjacent clectrolytr solution. A diffcrcnt approach to elec~rochemicnliy controlled rcleasc is based on polymers which bind :md release bionctive compounds in response to ark e!ectrical signalQ. The polymer has two redos states, only one of which is suitnblc for ion binding. Drug ions arc bound in one rcdox state and released fi-om the o&r. The attached electrodes serve to switch the rcdos states and the amount of current passed can control the amoun6 of ions rcleascd. Electrically control!.-d membrane permeabiiity is also of current intC;est in the field of electrically controlled or enhanced transdcrmal drug d&very (c.g. iontophorcsis)‘-‘*‘-‘.
I
