Drugs 23: 207-226 (1982) 0012-6667/82/0003-0207/$05.00/0 © ADIS Press Australasia Pty Ltd. All rights reserved.

Rate-controlled Drug Dosage John Urquhart ALZA Research, Palo Alto, California

Summary

Practical methods for rate-controlled drug dosing are bringing new capabilities to both therapeutics and experimental pharmacology, arising from the ability to maintain drug concentrations in plasma at desired levels for extended periods. Such dosing can prolong the duration of drug action and help to separate the beneficial effects of a drug from its undesirable actions. The new modes of drug release are distinguished from conventional sustained-release, depot, or slow-release products by the ability to specify precisely the release rate and duration in vivo on the basis of simple in vitro tests. Rate-controlledforms designed for research use now permit the 'isonemic' testing of drug effects - i.e. testing with drug concentrations in plasma maintained at constant levels. Thus, concentration-~ffect relationships become easier to discern than when testing is done with ordinary bolus dosageforms, which produce constantly.!1uctuating concentrations. Such testing appears to be basic in the effort to define the optimal regimen for many drugs. Published studies of various designs indicate that the usual regimens for a number offamiliar drugs are suboptimal, but can be improved with rate-controlled dosage. A variety o.ftechnologies are availablefor rate-controlled drug dosage. Infusion pumps and controllers are now widely used in hospital care, and demonstrate that rate-controlled delivery can assure the safe and effective use, for long durations, o.f drugs with narrow therapeutic indices and short half-lives, for example dopamine, nitroprusside, and dobutamine. More recently, rate-control is being incorporated into oral, transdermal, and various locally acting pharmaceuticals for use in routine outpatient therapy. 2 systemically acting, rate-controlled products are now available for clinical use: a matrix form o.f oral theophylline that has made that drug safer to use than previously; and a transdermal scopolamine product that prevents motion sickness without the objectionable parasympatholytic actions o.f this drug. Rate-controlled products available for local use include an ocular insert that delivers pilocarpine at a low steady rate for / week after each administration and a T-shaped intrauterine device which continually delivers progesterone. A decade of intensive technical development preceded the availability of rate-controlled products. For the ocular, trans dermal and intrauterine products presently available, the mechanism employed is drug diffusion through dense or microporous polymeric membranes. Membrane-controlled osmotic pumping of drug solution is the basis for other rate-controlled dosage forms for both laboratory and therapeutic use. There is a series of capsule-sized, /- or 2-week implants that provide rate-controlled drug administration in laboratory animals. For clinical research, the same osmotic technology is used in drug delivery modules administered orally or rectally. For therapeutic use, a tablet-sized oral dosage form is in fact a small osmotic

208

Rate-controlled Drug Dosage

pump. releasing drug at a constant rate over 12 to 18 hours. Clinical trials with a number of major drugs in this form are now underway. One important contribution of these and other developments is to broaden the basis for selecting new chemical entities as candidates for therapeutic use: short half-l(fe agents. as well as those with relatively narrow therapeutic indices. become practical for development as pharmaceuticals. Another contribution is to render safer or more effective the use of older drugs that have troublesome side effects or inconvenient regimens. Rate-controlled drug dosage is certainly no panacea; although the clinical application of this technique is slill in its infancy. it appears to offer a new approach for improving the selectivity and duration of action of many drugs to enhance their efficacy. safely. and convenience.

Advances in the technology of drug formulation have initiated a significant departure from the practice of specifying drug strength solely by quantity of drug in the dosage form. In the 1970s several locally acting pharmaceuticals entered the market bearing label specifications not only of drug quantity, but also of rate and duration of drug release. Subsequently, incorporating the principle of rate control into routinely used modes of both systemic and localised drug administration has become increasingly practical and economically feasible. Rate control brings 2 new dimensions to pharmacology and therapeutics: I) duration of drug action is substantially determined by duration of the specified rate of drug release from the dosage form (versus the familiar situation of duration determined by the drug molecule's kinetic properties); and 2) drug concentrations in plasma and biophases can be largely stabilised over time within the range that gives the best balance of therapeutic and untoward effects. The basis for these assertions and their implications for pharmacology and therapeutics are the focus of this review. A detailed description of technologies for achieving rate-controlled drug administration exceeds the scope of this article; however, section 2.2 summarises the principal technologies, which have been extensively reviewed elsewhere (see Chandrasekaran et al., 1978a; Eckenhoff et al., 1981; Heilmann, 1978; Jaminet, 1980; Kydonieus, 1980; Yates et al., 1~75; Zaffaroni, 1978).

I. Basic Considerations Paracelsus wrote that 'all things are poisons, for there is nothing without poisonous qualities. It is only the dose which makes a thing a poison' (Holmstedt and Liljestrand, 1963). This aphorism fmds practical expression in the dose-response relations that have been basic in translating pharmacological data to rational therapeutics. Adjusting the quantity of drug administered was the first - and for a long time the only - lever for extricating the drug's potential for efficacy from its toxicity. With growing sophistication in pharmacokinetics and analytical chemistry, there is a new and objective basis for understanding that the way in which a dosage is divided adds leverage to separating efficacy from toxicity. Identifying the optimum combination of how much/how often to realise a drug's fullest therapeutic potential is not always an obvious or straightforward undertaking. Consider as a reductio ad absurdum the 2 chronic regimens of (a) a 60pg dose taken every minute, and (b) a 31.5g dose taken every April I. These extremes add up to the same yearly total; the April Fool's regimen would possibly be lethal; the other, whatever its theoretical efficacy, is impractical. With conventional dosing methods, few selfadministered chronic regimens call for dosing more frequently than 6-hourly, and few acute regimens more frequently than 3-hourly. Although these frequency limits have never been formally proved (Haynes, 1981), they imply real limits on what one

Rate-controlled Drug Dosage

may expect patients to comply with in practical therapeutic settings. Yet attaining the full therapeutic potential of a number of drugs requires either that these frequency limits be breached (when conventional dosage forms are used) or that the drugs be administered continuously on a rate-controlled basis. It is instructive to review some of these drugs, noting how such evidence has been adduced.

1.1 Examples of Agents Requiring Impractically Frequent Administration of Finely Divided Doses for Optimum Therapeutic Action Ophthalmology provides several instances of a need for frequent drug administration; for example, the optimum use of idoxuridine eyedrops for herpetic keratitis calls for hourly administration (Laibson and Leopold, 1964). The regimen is necessary because of (a) the need for continuous presence of the agent, and (b) the rapidity of drug clearance from the tear film. The impracticality of once-hourly eyedrop use led to development of an ointment form, the drug release characteristics of which allowed 4-hourly dosing. An analogous example is the clinical impression (Lippas, 1964) that severe infectious ophthalmias are best treated in the fIrst 24 hours with instillation of chloramphenicol eyedrops every 5 minutes. A striking and well-researched example is the work of Jones et al. (I978) with topical diethylcarbamazine in ocular onchocerciasis: eyedrops given hourly to simulate continuous drug administration separated microf1laricidal activity from the inflammatory response previously believed to be its inevitable concomitant. Another instance is work showing that frequently administered microdrops of pilocarpine could drastically reduce the total daily dose needed to control intraocular pressure in glaucoma (Lerman and Reininger, I 97 J). Analogous examples exist in systemic therapy. Murphy et al. (J 96 I) divided the usual daily dose of hydrochlorothiazide and gave it on a 3-hourly basis; the drug's customary potassium-wasting effect disappeared whilst the diuretic effect remained. Similar

209

results with unusually frequent dosing of frusemide (furosemide) were noted (Urquhart, 1981) in the data of Wilson et al. (I975). Based on historical controls, frequent dosing of minocycline indicated that the incidence of vertigo is lower than with the usual twice-daily regimen (Greco et aI., 1979).

1.2 Agents Requiring Continuous Administration for Optimum Therapeutic Action Sikic et al. (1978) showed that, compared with 2 regimens of intermittent injections, continuous infusion of the same total dose of bleomycin in mice both enhanced tumoricidal activity and reduced pulmonary fibrosis, the agent's principal toxic effect. The infusions were done with an implanted, osmotically powered, capsule-sized pump (Theeuwes and Yum, 1976; see also section 2.2.2>. In man, continuous 5day infusion of cytarabine (ARA-C), repeated monthly, has emerged as a regimen superior to intermittent injections (Southwest Oncology Group, 1974). Methotrexate appears to be a safer agent when given by infusion than by intermittent doses (Bleyer, 1978), although, as Bleyer pointed out, defining the optimal regimen for methotrexate requires further research, notwithstanding 25 years of clinical experience with this agent. Pilocarpine, administered continuously by a ratecontrolled ocular insert (ALZA Corp., 1974; see also section 2.2.1), provides ocular hypotensive action without the marked myopia induced in prepresbyopic patients by the conventional 6-hourly pilocarpine eyedrop regimen (Brown et al., 1976; Place et al., 1975). The drug's miotic effect is also considerably less (Brown et al., 1976; Place et al., 1975). The total dose of pilocarpine needed with continuous delivery is only one-fifth to one-eighth that with eyedrops (ALZA Corp., 1974), thus confirming the frequently administered, fmely divided dosing study (Lerman and Reininger, 197 I), see section 1.1. In acetazolamide therapy for glaucoma, a rate-controlled, osmotically-powered oral form of aceta-

Rate-controlled Drug Dosage

zolamide provided ocular hypotensive action equal to that of ordinary tablets with significantly fewer occurrences of the drug's side effects (Theeuwes et al., 1978). Scopolamine given systemically on a continuous basis by a rate-controlled transdermal therapeutic system (see section 2.2. J) prevents motion sickness in 75 % of susceptible patients; partial and transient reduction of salivary secretion occurs, but other recognised parasympatholytic effects (Brand and Whittingham, 1970; Graybiel et al., 1976; Greenblatt and Shader, 1973) of the drug are infrequent or rare (Shaw and Urquhart, 1980; US FDA, 1980). In contrast, the conventional means of administering scopolamine usually produce tachycardia, profound reduction in salivary flow, drowsiness and, occasionally, CNS effects such as amnesia or hallucinations (Brand and Whittingham, 1970; Graybiel et al., 1976; Greenblatt and Shader, 1973). This new enhancement of the selectivity of scopolamine, accomplished by delivery system and regimen design, contrasts with the extensive but 'Iowyield' quest to find anticholinergic compounds that retain only certain specified anti-acetylcholine properties (Inch et al., 1973; Zwagemakers and Claassen, 198 J). Several studies have indicated therapeutic superiority of continuous intravenous infusions over intermittent injections of various drugs, for example: morphine for postoperative analgesia (Rutter et al., 1980); insulin for diabetic ketoacidosis (Lutterman et al., 1979); cis-platin (platinum diamminodichloride) for solid tumours (Slem et al., 1978); heparin for thromboembolism (Salzman et al., 1975). Buchwald et al. (1980) provided a striking extension of the findings of Salzman et al. (J 975) by showing the efficacy and safety of continuous heparin infusions lasting several months using a miniature, vapour-pressurepowered, implanted pump (Blackshear et ai., 1979) about the size of a cardiac pacemaker. Figur~ 1 illustrates their results. For inhospital management of postcoronary and postcardiac surgical patients, routine use is made of practicable, reliable, and (almost) 'fool-proof infusion

210

pumps and controllers (Kitrenos et aI., 1978), especially for nitroprusside, dopamine, and dobutamine. These potent, ultrashort half-life agents can be given in no way other than by rate-controlled intravenous infusion; in practice, the physician determines their 'duration of action' by specifying hours or days of infusion, depending on the therapeutic situation. The drugs' ultrashort half-lives allow rapid changes in their actions and rapid attainment of a new steadystate by adjusting infusion rate. Increasingly, infusion pumps or controllers are also being used by glyceryl trinitrate (nitroglycerin) to reduce afterload, noradrenaline (norepinephrine) - to regulate arterial pressure, aminophylline - to reverse bronchiolar constriction, and narcotic analgesics - to manage intractable pain.

1.3 Research Role ofTest Regimens Utilising Frequently Administered, Finely Divided Doses From the research perspective, administering a drug, either topically or systemically, by frequently administered, finely divided doses is a straightforward way to approximate continuous, rate-controlled administration of the drug. The frequency of dosing needed to approximate continuous administration is a pharmacokinetic calculation specific to each drug (De Jongh and Wijnans, 1950; Theeuwes and Bayne, 1977 ; Wagner, 1975). A frequently administered, finely dIVided dosing study allows one to substItute almost constant plasma concentrations of drug for peak and trough fluctuations, with the constant level being intermediate between peak and trough. With most conventional oral regimens, the peak concentration is at least 3 to 8 times the trough concentration. Some drugs, for example ticrynafen (Hwang et al., 1978) and minoxidil (Lowenthal et al., 1978), have peak to trough ratios of 20 or more to 1. In principle, it is also possible to design frequently administered, fmely divided dosing regimens in which the doses vary over time in order to induce, for research purposes, circadian or other rhythms in drug concentration.

Rate-controlled Drug Dosage

~:: ~I 0.2

0.6

I

211

Patient JF

~~

....

~--~~

. .......

o .I~~-'I___IL~_~I _-LI_----1.I_---'.I_--'I_ _LI_ _ JL-I ~ 11111 I

II I I

I

O·;~~

__"'·I~

0.6~ II

I

0.4

Patient MH

I

I

I

I

I

I

I

I

j~·~t~ c

o

.~0.6l~1 I I I II I I ai 8 u c

0.4

Patient MM

0.2

E O . '"'" »

.j ~.: ~ I

0.2

I

Patient AP

~I I !~.................0......

..

... .....

~.

J~~_--'I_--'lL-_LI_-L_~~I_--'I __--'I____LJ__-L-; 2

3

4

5

6

7

8

9

10

11

12

Duration of infusion (months)

Fig. ,. Plasma heparin concentrations from 5 patients during 1 year of continuous infusion. The arr.-s designate the times when the pump was refilled with heparin solution. The peaks that occurred at 9 and 10 months for patient MM resulted from spUlage of small amounts of heparin into the subcutaneous tissue during refill procedure [after Buchwald et al. (1980); reprinted with permissionL

Frequently administered, finely divided dosing experiments can usually be performed by simply dividing available formulations. Obvious logistical problems arise, however, in assuring the faithful compliance that is crucial for valid results. Until recently, such dosing studies have been seen as having so little practical therapeutic value that few have been carried out. At present, however, the situation is changed because of the growing practical capability for continuous drug administration from infrequently administered, rate-controlled dosage forms.

1.4 Research Role of Continuous, Rate-controlled Infusions of Drugs Infusion pumps bring an important new capability to short term care, especially in acute cardiovascular medicine and surgery. Their forerunners were long time 'creatures' of the physiology laboratory, but required too much close attention to be practicable in therapeutics, where critical issues are breaches in sterility, air embolism, infiltration, complex calibration and set-up procedures, and lack of failure mode

Rate-controlled Drug Dosage

warnings. A rapid succession of design improvements, however, has brought infusion pumps into widespread clinical use. In turn, 3 important new cardiovascular drugs (see section 1.2) are already in routine use that would not be, but for the advent of practical infusion equipment. Fortunately, these devices had largely - and safely - gone through the stages of rapid, iterative improvements in design before the advent of governmental regulation of devices. In contrast to the frequently administered, finely divided dosing method, which has had only sporadic application, pump-controlled infusions appear to be a research tool of growing use that should open up unprecedented opportunities for regimen optimisation studies. The pumps used include not only the relatively large pole-mounted, inhospital pumps, but also the portable (Pickup et al., 1979; Tamborlane et al., 1979), wearable (Leeper et al., 1977; Wright et al., 1977), and implantable (Blackshear et al., 1979) pumps for use in ambulatory humans. Studies using these small devices should indicate new therapeutic opportunities based on long duration intravenous or subcutaneous infusions in ambulatory patients, for example for cancer chemotherapy. Other studies should indicate optimal rates of drug administration to be provided non-invasively by novel oral, transdermal, suppository or other forms.

1.5 Pharmacokinetic Equivalence of Frequently Administered, Finely Divided Dosing and Various Modes of Continuous Administration The interchangeability of results from a frequently administered, fmely divided dosing regimen, and intravenous infusion, or a rate-controlled dosage form is based on the important pharmacokinetic principle of bioequivalence (DiSanto, 1980). Basically, the issue in judging bioequivalence is whether an observer - privy only to information as to the time course of drug concentration in plasma - can correctly distinguish one regimen of drug administration from another. Systemic regimens are considered bio-

212

equivalent if they produce indistinguishably different time courses of drug concentration in plasma. In general, the judgement as to what is 'indistinguishably different' is made informally by inspection. Time series analysis (Box and Jenkins, 1976) is used in other fields as a formal method of making such a judgement; its use in clinical pharmacokinetics is probably timely.

1.6 Influence of Therapeutic Drug Monitoring on Regimen Design Rapid advances in analytical chemical technology have made the monitoring of drug concentrations in plasma increasingly possible, and simple. Initially such measurements were used only as a research tool, but more recently they have become a part of medical practice. All major anticonvulsive drugs, certain cardiac drugs, theophylline, some antibiotics, and some antidepressants are increasingly subject to therapeutic drug monitoring in routine treatment (Koch-Weser, 1972, 1981; Pippenger, 1980). The number of drugs for which therapeutic drug monitoring is indicated can be expected to grow. Much scepticism as to cost-effectiveness has ah tended the use of therapeutic drug monitoring (KochWeser, 1981; Richens and Warrington, 1979). Nevertheless, such monitoring has undoubtedly aided regimen design and has exposed the frequent need to individualise regimens for optimum benefit. Hoppener et al. (J 980) provided a telling vignette from a study aimed at optimising carbamazepine regimens in epilepsy: ' ... it was striking that many patients did not mention the side effects spontaneously, although in some cases they must have existed for years. A number of patients did not complain of sleepiness because they thought it normal that, as a result of their work, they were so tired during the afternoon that they had to go to bed. They believed that this fatigue was connected with their epilepsy and they accepted it, often the more so because in the past, no attention Was paid to their complaints. After adjusting the dose scheme of the medicament in such a way that

Rate-controlled Drug Dosage

the serum levels did not exceed 8mg/L, the side effects disappeared in all cases. None of the patients showed an increase of seizure frequency after this change.' One may be sceptical that maintaining a concentration of carbamazepine 8mg/L is an absolute panacea, but this study points up the extent to which many patients passively accept, and many doctors ignore, side effects that degrade the quality of patients' lives, although they may not produce measurable signs or biochemical alterations. Fortunately, clinical pharmacological studies have begun to focus more strongly than in the past on such ubiquitous but less readily measured drug effects. They include, for example, drowsiness (Carruthers et al., 1978), fatigue (Anonymous, 1980), subtle changes in mental state (Greenblatt and Shader, 1978; Prescott et ai., 1970), and a restricted range of activities of daily living (Weintraub et al., 1977). Relating these subjective effects to optimal drug regimen design is a largely unmet challenge to clinical pharmacology. For drugs whose actions correlate with their concentrations in plasma, the regimen has a clear objective: to maintain the accepted therapeutic concentration range for the drug. Obviously that objective is only as valid as the information on which the accepted therapeutic concentration range is based; the ultimate objective of any regimen has to be therapeutic benefit at an acceptable risk and an acceptable level of untoward actions. When an accepted therapeutic concentration range can be defined and used as a valid surrogate criterion for prescribing, it is in principle simpler - if currently unfamiliar - to prescribe the desired drug in a form specified by rate rather than in one specified by quantity. The crucial difference between th~ two bases for prescribing lies in their different pharmacokinetic consequences. Drug entering the bloodstream continuously at a specified rate can maintain a desired concentration. Drug entering the bloodstream discontinuously, in more or less bolus fashion, produces first a peak in concentration that subsequently declines to a trough by the time of the next dose. The latter situation prevails with most regimens used to-

213

day, and complicates the interpretation of measured drug concentrations, because the time since the last dose is such an important factor influencing the measured value.

1.7 Research Role of Rate-controlled Dosing in Improving the Definition of Accepted Therapeutic Concentration Ranges Correlating effects with concentrations during conventional dosing is inherently difficult, because concentrations fluctuate so widely, rapidly and often irreproducibly from one dosage cycle to the next. However, the many ways now available for achieving or simulating (e.g. by frequently administered, finely divided dosing) rate-controlled drug administration should help to improve the definition of accepted therapeutic concentration ranges for many drugs including some not currently perceived as having clear concentration-effect relationships. By maintaining constant plasma concentrations it should be possible to recognise correlations between drug concentration and those drug effects that require some hours to appear, after being triggered when concentration exceeds a certain threshold. In contrast, conventional dosing can be seen to be a relatively weak or even misleading tool in understanding concentration-effect relationships: when a briefly maintained peak concentration triggers an action that takes hours to develop, the clinical observance of the action could even coincide with the ensuing trough in concentration. Moreover, fluctuating interdose drug concentrations can confuse interpretations in basic pharmacological studies. An illustrative caution was provided in a recent study of anticholinergic drug actions: 'As time-action relations are not necessarily identical for various organs, a wrong choice of the time of measurement o,f the various anticholinergic effects might falsely indicate ... selectivity' (Zwagemakers and Claassen, 1981). For various reasons, many drug actions wax or wane slowly, while others do so in a manner directly

214

Rate-controlled Drug Dosage

related to rising and falling drug concentrations (Gibaldi et al., 1970. Rate-controlled dosing now enables researchers to maintain concentrations at equal levels over time, permitting 'isonemic' testing of drug actions (Urquhart, 1981). The significance of isonemic testing is that it permits correlations between drug concentrations and drug actions, whether the actions are slowly or rapidly developing ones. It also may help obviate the need for laborious studies of time-action relationships in comparing relative selectivities of compounds in a given series (Inch et ai., 1973). A well-documented accepted therapeutic concentration range and convenient therapeutic drug monitoring procedures can only improve precision of drug therapy and lower its risks. Rate-controlled dosing, by minimising dose-dependent peaks and troughs, stands to simplify both the timing of blood sampling and the interpretation of measured drug concentrations. It is expected that prescribing by rate, in contrast to prescribing by quantity, will simplify the achievement and maintenance of an accepted therapeutic concentration range provided, of course, that rate-specified forms are readily available. Even when drug concentrations are controlled at more or less constant values, certain factors can complicate the definition of concentration-effect relationships. For example, extensive protein binding of some drugs results in the active unbound moiety being but a small percentage of the total concentration. Thus, proportionally large variations in the active unbound moiety go unobserved when only the total concentration is measured; the effect is to mask concentrationeffect relationships. Another complicating factor may be the formation of pharmacologically active metabolites of the parent drug. Such metabolites may be seen as a form of 'pharmacodynamic shrapnel', arising in an only partly predictable manner from the parent drug, however chemically pure it may be when administered. The relative constancy of drug concentrations provided by rate-controlled dosing may reduce the confusing impact on accepted therapeutic concentration range estimations of variations in protein binding or in the formation of active metabolites.

2. Bases/or Designing Rate-controlled Pharmaceutical Preparations Rate-controlled pharmaceutical preparations integrate the active component into a dosage form at the time of manufacture. In contrast, rate-controlled infusion pumps and controllers are loaded with the chosen drug just prior to use, and so are classed as devices. Rate-controlled preparations pose more complex stability issues than do conventional, quantityspecified pharmaceutical preparations. With the latter, expiration dating is based upon how long drug purity and content stay within specifications with the product stored under stated conditions. Rate-controlled preparations should maintain not only static specifications of drug purity and content, but also kinetic specifications of drug release rate and duration (Michaels et al., 1976). Meeting these stability requirements calls for in vitro test methods that are 'bioanalogous' in that they reliably predict drug release in vivo. Considerations in defining bioanalogous in vitro test methods are discussed by Urquhart (1980). Three complementary bases exist for achieving kinetically precise rate-controlled drug administration: (a) to shield the process of drug dissolution from the kinetic vicissitudes of the biological environment; (b) to meter the flow of drug in solution by the use of well-understood, physicochemical mechanisms of mass transport, usually involving biologically inert membranes; and (c) to integrate those functions, as well as a suitable quantity of drug, into a conveniently compact dosage form. Drug in solution is generally the most readily absorbable form; drug in the solid state poses the fewest stability problems and occupies the least space.

2.1 Functional Elements of Rate-controlled Dosage Forms It is instructive to consider the intravenous infusion as a functional prototype of a rate-controlled dosage form. The familiar infusion assembly contains

Rate-controlled Drug Dosage

each functional element that has to be integrated into something as small as a tablet or an adhesive skin patch to realise rate-controlled dosing - the marked size discrepancy notwithstanding. Both rate-controlled preparations and infusion assemblies represent 5 functional elements plus I element of 'software', as outlined below:

2.1.1 Drug Reservoir For an intravenous infusion, the reservoir is a bottle, bag or burette containing the drug in solution. The quantity of drug (mg), divided by the intended rate of its infusion (mg /h), specifies duration of drug administration (h) before replenishment is needed. The infusion rate of the drug (mg/h) is the product of drug concentration in the reservoir (mg/ mO and the rate of solution flow (ml/h).

2.1.2 Energy Source For a gravity-flow intravenous infusion, the energy source is gravity acting on the infusate, creating hydrostatic pressure higher than venous pressure. For a pump-driven infusion, it is the mechanical force generated by the pump's motor.

2.1.3 Rate-controlling Element For an intravenous infusion, control of the rate of drug administration is attained by controlling flow of drug solution by a fluid flow regulator, e.g. a screw clamp, drop counter, or positive displacement pump.

2.1.4 Delivery Portal For the intravenous infusion, the delivery portal is the tip of the intravenous catheter or needle from which drug solution passes into the biological environment.

2.1.5 Platform The platform is the physical structure that holds the functional elements together and interfaces with the patient. With an intravenous infusion, it includes the bedside pole supporting the drug reservoir and rate controller, the administration set, the needle or catheter, and terminates at the portal.

215

2.1.6 Drug Programme The programme is the intended time course of drug administration. In the simplest case it is constancy, but for fastest onset of action, it is a high priming rate for a short duration, followed by a lower maintenance rate. Recent advances in understanding the pharmacodynamics of some agents, most notably gonadotrophin-releasing hormone (see section 3. I), indicate a need for complex programmes of timevarying rates.

2.1.7 Drug Regimen vs Drug Programme The drug regimen defines what the patient or health professional should do, at intervals, to re-medicate so as to attain the therapeutic objective. The drug programme defmes what the dosage form is supposed to do during the time between each remedication. It is a novel concept to expect a dosage form to 'do' something, other than to disappear as soon as possible. The aim in defming a functional mission for this new type of dosage form is to make the regimen as convenient, easy to remember, and simple as is technically and economically possible. A drug programme may be a single, constant rate, or it may be a specified time sequence of rates. Conventional dosage forms - including sustained-release forms - provide a complex, timevarying series of rates. These rates, however, are not specified nor in any sense designed; instead they are the outcome of largely serendipitous formulational procedures (Rogers and Kwan, 198 J). Such dosage forms cannot be kinetically specified because they leave drug dissolution subject to too many in vivo variations, for example, pH, intestinal motility, and surface area of yet-undissolved drug.

2.2 Physicochemical Mechanisms for Achieving Rate-controlled Drug Dosing Technologies exist for achieving rate-controlled drug dosing from objects small enough to be swallowed, to be affixed to a small area of skin surface, to be inserted under an eyelid or in the vagina,

216

Rate-controlled Drug Dosage

Backing membrane Drug reservoir Drug molecules \i. - - - - - / - - - . • Rate-controlling microporous membrane ~ O.20mm Skin contact adhesive - + ~:p'I:\.g::':P.i5li~ Skin surface

Epidermis

Sweat duct Blood capillary

Fig. 2. Schematic cross-section of the scopolamine transdermal therapeutic system in place on the skin surface. Drug molecules permeate skin by diffusion and enter dermal capillaries. through which they gain access to the systemic circulation.

uterine cavity or rectum, or to be implanted subcutaneously. Such technologies include the following:

2.2.1 Membrane-controlled Diffusion Using this technique, drug in solid state is enclosed by a membrane whose permeability permits the drug to diffuse across the membrane and into the surrounding tissues for absorption. For this mechanism to provide rate-controlled drug release in vivo, it is necessary that drug absorption is capable of occurring at high multiples of the rate at which the drug crosses the membrane. This condition ensures that drug does not accumulate outside the system; thus, the concentration of drug is nearly zero in the layer of tissue fluid between the membrane and the surrounding tissues. When this condition is met in vivo, the ratio is 1 : 1 between drug release in vivo and drug release in vitro in a well-stirred, isotonic solution at body temperature. These simple in vitro conditions then qualify as bioanalogous for testing purposes. Three diffusional systems have been developed and registered in a number of countries: 2 for localised therapy and 1 for systemic treatment. As discussed in section 2, bioanalogous in vitro test con-

ditions were the logical prerequisite for clinically relevant quality control tests of these systems and for label specifications of their drug release rate and duration.

Diffusion through dense membranes - the first rate-controlled preparations: If the membrane is dense, drug dissolves in it and then diffuses in the direction of the concentration gradient between drug within the reservoir and the very low, but pharmacologically significant, concentration external to the system. Modifications of the Fick diffusion equation govern the operation of such systems (Chandrasekaran et al., 1978a), which signifies that the energy source for drug release is the difference in the drug's chemical potential between the reservoir and the system's exterior. This is the principle of operation of the first rate-controlled pharmaceutical products: the I-week pilocarpine ocular therapeutic systems (ALZA Corp., 1974) and the progesterone uterine therapeutic system (ALZA Corp., 1976). In these locally acting systems, solid drug is enclosed by a dense membrane made of the copolymer of ethylene and vinyl acetate. The detailed physical chemistry of these systems is described elsewhere, along with the

Rate-controlled Drug Dosage

experimental demonstrations that rate control prevails in vivo in humans with each system (Chandrasekaran, I 978a). The pilocarpine ocular therapeutic systems were the first pharmaceuticals to carry rate and duration specifications of strength 20)lg/h or 40)lg/h pilocarpine, each for I week. The platform design is that of a thin, elliptically shaped, flexible fUm: both its surfaces serve as drug portals. These systems are registered in many countries. The next rate-controlled preparation to reach the market was the progesterone uterine therapeutic system. Its initial label specification of strength and duration was 65)lg/day progesterone for I year. Through a recent change in reservoir content from 38 to 52mg of progesterone, the duration specification is now being revised upwards to 2 years, although the rate of hormone release remains substantially the same. This product illustrates the principle of using controlled release technology to determine duration of drug action: these 1- and 2-year products are based on use of a hormone with a 45-minute half-life (Fylling, 1970). A thorough review of the clinical performance of this product is to be found in Mishell and Martinez-Manautou (I 978). Diffusion through porous membranes - the first transdermal rate-controlled preparation: In another type of diffusional system, a microporous membrane encloses the drug. In this case, drug diffuses through microscopic channels in the membrane. Usually solid drug is suspended in a liquid vehicle in the reservoir, with the vehicle also filling the membrane's channels. The scopolamine transdermal therapeutic system (fig. 2), for the prevention of motion-induced nausea, provides an example of this mechanism (Shaw and Urquhart, 1980). This system is also governed by a modification of Fick's diffusion equation. It acts systemically, and it was the first rate-controlled transdermal pharmaceutical preparation to pass regulatory review (thus far in the UK, the USA, and Canada). The platform is a small circular adhesive fUm. The adhesive inner face of the product constitutes the portal, from which released drug diffuses across intact

217

skin into the bloodstream. Rate control is provided by the microporous membrane interposed between the drug reservoir and the adhesive. The product carries a label specification of strength that reads: programmed delivery in vivo of 0.5mg scopolamine over 3 days. As this labelling implies, the product incorporates a rate programme: a priming quantity of 140)lg scopolamine released at a controlled, asymptotically declining rate over 6 hours, stabilising at the maintenance rate of 5)lg/h for the remainder of the product's 3-day functional lifetime. Virtual constancy of rate, after delivery of the priming dose, is assured as long as drug is present in excess in the reservoir (i.e. a

Drug capsule

System edge

@®@®@)~=>

'--1

Drug delivery

Fig. 3. Schematic cross-section of the osmotic film principle for rate-controlled drug delivery. The drug capsules. of microscopic size. are surrounded by a waterpermeable, drug-impermeable polymeric matrix. Water enters the matrix from the system edge. The rate of water entry is governed jointly by the permeability of the matrix to water and by the osmotic pressure difference between the drug and the surrounding aqueous biological medium. As water enters a capsule, it swells until the matrix parts to create a microscopic channel by which drug in solution escapes to the system edge. The process is repeated, with serial opening of ever more deeply situated drug capsules, until the system is exhausted of drlll!.

218

Rate-controlled Drug Dosage

Drug solution leaving - - - - via delivery portal .."",~.----- Flow modulator

--

Flexible. impermeable reservoir wall Saturated solution of osmotic agent

Water entering - - - rate-controlling membrane Reservoir

Fig. 4. Schematic cross-section of a generic osmotic pump. The reservoir !light stippling) is filled with a drug solution. The reservoir wall (cross-hatching) is impermeable both to water and solvent. Between the rate-controlling membrane (unstippled outermost layer) and the reservoir wall is situated a layer of osmotically active agent. the osmotic pressure of which causes water to permeate the rate-controlling membrane. That membrane is rigid and impermeable to the osmotically active agent. Hence. incoming water acts to compress the reservoir and to expel drug solution in an isovolumic fashion. The flow moderator provides the flow path to the portal. Delivery of drug solution continues at a constant rate until the drug reservoir is completely collapsed. Generic osmotic pumps range in size from that of an oral capsule to that of a human finger.

saturated solution in the vehicle). The priming dose is achieved by placing a quantity of drug in the adhesive layer, external to the membrane, which is thus available for prompt absorption. This product has been demonstrated to be bioequivalent to a 72-hour constant intravenous infusion of scopolamine (Shaw and Urquhart, 1980), which provided the basis of its strength labelling (Shaw and

Urquhart, 1980; US FDA, 1980). Its clinical pharmacological attributes are described in section 1.2. Compared with the osmotic systems described below (section 2.2.2), diffusional systems are restricted in their rates of drug delivery per unit area of membrane. They are nevertheless well-suited to both localised therapy and transdermal systemic therapy. The latter is necessarily limited by the skin's ability to absorb the drug. With present technology, transdermal systems of reasonable size are feasible for drugs active at daily parenteral doses of IOmg or less.

2.2.2 Membrane-controlled Osmotic Pumping of Drug Solution This basic principle takes 3 forms: (a) osmotic film, (b) the 'generic' osmotic pump, and (c) the 'elementary' osmotic pump. In each instance the energy source for drug release is the pressure difference between the osmotic pressure of drug within the system and the osmotic pressure of the biological environment in which the dosage form is placed. Osmotic film: Osmotic film has been employed in a number of experimental systems. The principle is adaptable, depending on the platform configuration, to use as ocular, vaginal or rectal inserts, or as tablets. Its functional principle is illustrated in figure 3 and described by Gale et al. (1980). Generic osmotic pumps: Generic osmotic pumps are multipurpose experimental tools for animal or clinical studies. They are fabricated to be filled by the user with solutions or suspensions of any of a wide variety of agents - hence the term 'generic'. Their structure and rate-controlled functionality are illustrated in figure 4 and described by Theeuwes and Yum (! 976). This technology is the basis of the miniature osmotic pumps widely used in laboratory animals as 1- and 2-week duration, capsule-sized implants. They have all but replaced the infrequently used, cumbersome alternative of connecting a highly mobile small animal to a heavy infusion pump (Myers, 1977; Weeks, 1979). Moreover, they have greatly expanded the evaluation of bioactive agents in extended duration studies where drug concentrations in plasma

Rate-controlled Drug Dosage

219

and/ or tissues are stabilised over time (i.e. isonemic testing as described in section 1.7). There are now over 300 publications based on their use in experimental pharmacology, physiology, endocrinology and the neural sciences (e.g. Lynch et aI., 1980; Martucci and Fishman, 1979; Pettigrew and Kasamatsu, 1978; Sikic et al., 1978; Smits et aI., 1980; Wei and Loh, 1976). This technology, embodied in 8-, 12-, or 24-hour capsule- or suppository-sized delivery systems, has served for a number of unpublished clinical research studies of drug absorption by several pharmaceutical companies. One such study has been published - on rectal drug absorption of antipyrine (De Leede et al., 198 J). Eckenhoff et al. (J 98 J) have presented evidence of the constant functioning of such delivery modules in gastrointestinal fluids of various pHs.

Elementary osmotic pumps: These are rate-controlled dosage forms that enclose solid drug within a membrane and release drug in solution. They represent the simplest and most compact version of the osmotic pump (Theeuwes, 1975). Figure 5 illustrates the system, and the legend describes its principle of operation. Typically, 60 to 80 % of drug content is delivered at a constant rate. The duration of drug delivery is determined by one's choice of membrane permeability to water and drug content of the system. Rate-controlled oral forms resembling ordinary-sized tablets may achieve rates as high as 60mg/h. Duration of drug delivery is selected by design to be consistent with a predictable rate and extent of drug absorption, which of course is related to intestinal transit time and to drug absorption by the small and

Drug solution

Osmotic core containing drug

Fig. 5. Schematic cross-section of the elementary osmotic pump.

The drug acts as its own osmotic 'driving agent', causing water to permeate the outar membrane, which is impermeable to drug. A!s with the generic osmotic pump, the membrane is rigid, forCing drug solution to exit at the same volume flow rate as water enters. The departing solution is saturated with drug. Hence, the system delivers drug in solution at a constant rate until the last of the solid drug in the system's interior is dissolved, after which time the osmotic pressure within the systam commences to faH. At the same time, the concentration of drug solution leaving the system also falls, giving rise to a parabolic decline in drug release rate. TypicaHy, elementary osmotic pumps deliver 60 to 80% of their contained drug at a constant rate, folbwed by delivery of the remaining 20 to 40 % of drug at a parabolically declining rate. 8ementary osmotic pumps for pharmaceutical use are the size of ordinary tablets. The membrane is usually relatively much thinner than the diagram shows, and usually constitutes less than 10 % of the system's volume. For pharmaceutical applications as oral dosage forms, the delivery orifice -

which is about pinhole size -

is placed by

laser. The membrane gives the tablet a surface texture like that of an ordinary film-coated tablet. The membrane remains intact, however, and is excreted in the faeces.

220

Rate-controlled Drug Dosage

large bowel. As with the generic osmotic pumps, drug release from elementary osmotic pump dosage forms is rigorously independent of pH, stirring rate, and intestinal motility. The range of delivery times for reliable absorption depends on the drug, but with some drugs can extend beyond I 2 hours (Godbillon et al., 1981). Thus, osmotic technology offers one approach to rendering obsolete the requirement for more than twice-daily dosing. This technology was first described by Theeuwes in 1975. It has been extensively expanded since that time, to adapt the osmotic pump technique to drugs of widely varying water solubility (Eckenhoff et al., 1981; Theeuwes and Bayne, 1981; Theeuwes and Eckenhoff, 1979). A number of widely used, orally administered drugs are in the process of being incorporated in elementary osmotic pump form. Initial published reports show rates of drug absorption in man proceeding as quantitatively predicted from both the physicochemical principles of operation and the performance of the dosage form in vitro (Godbillon et al., 1981; Rogers et al., 1981).

2.2.3 Other Mechanisms By dispersing drug particles through a relatively hydrophobic matrix, it is possible to restrict access of water to the drug and so to retard its dissolution and release. This is the principle of almost every sustained-release, 'retard', depot, or slow-release dosage form (Robinson, 1978). However, with the notable exception of a recently developed oral theophylline product (Spangler et al., 1978), this approach has not yielded sufficiently predictable drug release to justify being called rate-controlled (Urquhart, 1981). This is not to say that such products are not beneficial improvements over simple, rapidly dissolving forms, but they do not release drug with the validated kinetic reproducibility in vivo and in vitro that would justify labelling them by rate. A related aspect of the various sustained-release dosage forms is a considerable degree of empiricism in the definition of in vitro quality control test methods (Cabana, 1979). Correlation is conspicuously absent between the kinetics of drug release in

vivo and in vitro. The basic problem is that dissolution, although familiar, is kinetically a very complex phenomenon, and is sensitive to pH, to stirring rate, and to various ionic and other constituents of the medium. Because these factors are subject to considerable fluctuations in vivo, it has not been possible to derme bioanalogous test conditions for sustainedrelease dosage forms. These considerations emphasise the significance of shielding the process of dissolution behind membranes which exclude the solutes that complicate drug release kinetics. There has been a great deal of work in the past decade on bioerodible polymers with erosion characteristics sufficiently reproducible to control the release rate of drug in solution from solid particles of drug dispersed through the polymer. These materials show promise as the basis for multi week , rate-controlled injectable preparations (Benagiano et al., 1979; Choi, 1980) and wound or burn coverings (Vistnes et al., 1976). However, they require further testing to determine if this promise will be realised in practical therapeutic products.

3. Future Developments Many new pharmaceutical products based upon the technologies described above are undergoing development and testing. Some have been the subject of preliminary reports, limited to bioavailability studies and other biopharmaceutical aspects, for example oral elementary osmotic pump forms of indomethacin (Rogers et al., 1981) and metoprolol (Godbillon et al., 198 J). Others, for example a transdermal therapeutic system form of glyceryl trinitrate, have been announced as being in development and undergoing clinical trials by their developers, although no data have yet been published. In the past several years, many of the major research-based firms have indicated their initiation of research work on novel drug delivery systems. In addition, newer firms have specialised in this area. A recent review (Johnson, 1980) covers about 600 patents in this field, issued in less than a decade.

Rate-controlled Drug Dosage

Assuming that some of these many activities provide new pharmaceutical products over the next several years, it is reasonable to expect that physicians will become increasingly familiar with prescribing drugs by rate. It also seems reasonable to expect growing recognition that advances in therapeutics may arise from novel methods of drug administration and new regimen design, as well as from novel drug molecules. The developers of new products based on rate-controlled dosing have several important responsibilities: (a) to provide rational, well-tested products (Prescott, 1981; Urquhart, 198 I); (b) to provide each product in a range of delivery rates that allows satisfactory individualisation of rate to achieve the best balance of therapeutic and side effects and/ or the accepted therapeutic concentration range; (c) to provide, in appropriate circumstances, priming dose capability for rapid onset of drug action; and (d) to provide lucid instructional materials, correlating prescribing and monitoring information. These materials should describe procedures for rate-adjustment when clinical and/ or therapeutic drug monitoring results are suboptimal with the rate initially prescribed. Obviously, the technology for rate-controlled dosing has had to precede the products. That technology is well in hand after a decade of work, although the products are as yet few. In many industries, new technology translates into new products in a 2- to 5year period; pharmaceutical product development proceeds much more slowly, in an environment of pervasive caution~ Today's caution differs radically from that which accompanied the transfer of insulin and penicillin from laboratory to clinic. Today the pace of such progress would be considered appalling haste, doubtless because of a shift in attitudinal focus away from benefits and toward risks. Medical devices, notably the various infusion pumps, have fortunately advanced with an alacrity long since unknown in pharmaceutical innovation. Yet even in a closely regulated industry, new technology moves ahead with its own imperative, often in advance of explicitly defined needs for it. Indeed, those needs may become apparent only after a new,

221

hitherto unrecognised capability is demonstrated. The pocket calculator exemplifies this phenomenon: no one in 1971 'needed' a pocket calculator, but in 1981 few are without them, and the slide rule has joined the abacus in the museum. At the present time, despite the developments described above, even the concept of rate-controlled drug administration is unfamiliar to most, not only to most practitioners (except those involved in acute cardiac care), but also to most pharmacologists. In contrast, pharmaceutical scientists and pharmacokineticists are keenly aware of this concept, of the new developments it has spawned, and of many of their implications. For practitioners, the lag is understandable: one can practise medicine only with today's tools, not tomorrow's. For pharmacologists, there is an immediate need to re-evaluate the conventional formating of pharmacological data and the protocols for gathering such data; at present, pharmacology is so thoroughly involved with conventional bolus modes of drug administration as to obscure beneficial uses of the new methods described here. For example, 'duration of action' has been part of the standard pharmacological profile of a drug. Now this concept needs re-examining, for it can be more an attribute of dosage form design than of pharmacology. The advantages of rate-controlled dosing methods in drug screening and evaluation are obvious from the preceding discussion of their properties.

3.1 A Unique Challenge for Future Developments Endocrinology is currently providing the most striking challenge to drug delivery methods of the future, through the example of gonadotrophin-releasing hormone (GnRH), whose temporally-patterned secretion or replacement can fundamentally change the qualitative nature of this hormone's action. These new findings, based on the pioneering work of Knobil and his associates (Knobil, 1980) have now begun to be carried over into clinical use (Crowley et al., 1981; Leyendecker et al., 1980). They challenge some of the most basic principles of the dynamics of hormone ac-

222

Rate-controlled Drug Dosage

tion, and perhaps of drug actions as well. The basic work shows that the physiological role of GnRH is permissive. That is, it allows cyclically varying ovarian secretion of oestradiol to do 2 things: either (a) at low concentrations of oestradiol, to inhibit gonadotrophin secretion, or (b) at high concentrations, to stimulate the ovulating surge of gonadotrophin secretion. Yet to exert this permissive action, GnRH has to be secreted or infused in a narrowly bounded, repeating temporal sequence of approximately 5 minutes on, and 60 (monkeys) or 90 (humans) minutes off. This regimen has successfully achieved ovulation and pregnancy in women with hypothalamic amenorrhoea (Leyendecker et aI., 1980). In contrast, regimens of ordinary injections or constant infusions of GnRH trigger a single surge of gonadotrophin secretion, followed by profound inhibition for the duration of the regimen. Indeed, an injection or infusion regimen has been used to arrest precocious puberty by inhibiting the adult cyclic pattern of GnRH secretion (Crowley et aI., 1980). This striking pattern-dependent hormonal action may turn out to be an isolated evolutionary 'fluke'. It may also be an object lesson and a challenge to those interested in the dynamics of drug actions, as well as to those seeking to provide better ways to administer therapeutic agents. How many therapeutically useful agents are obscured because of simplistic concepts of pharmacodynamics and primitive methods of administration is anyone's guess. It will take much basic research with novel delivery systems to answer these questions. 3.2 Conclusions Achieving rate control means that dosage form design can complement a drug's molecular structure in determining both selectivity and duration of drug action. Consequently, one can expect to see advances in both therapeutics and pharmacology that depend on improved dosage forms as well as advances based on new drug molecules. In particular, rate-controlled dosage forms open the way to developing very short half-life, potent substances into practical pharma-

ceutical products. Many of the short half-life physiological peptides expected from recombinant DNA techniques should benefit from these new dosage forms. In many ways, the most important contribution of these developments is to broaden one's views as to requirements for a useful new agent. In the context of this review it is interesting to reinterpret some rather pessimistic views on the therapeutic utility of a powerful natural substance, from a report on a recent conference on prostacyclin: .... Such substances could have potential, both as prophylactics and treatment in thrombotic disease, hypertension, platelet consumption, and all forms of vascular disease. Such broad application of prostacyclins is a mouth-watering prospect for any drug manufacturer. Unfortunately, however, there are likely to be difficulties. The dose-response curve for the native substance is steep. The onset of effect and onset of side effects occur close together within one log unit of each other. Some clinical pharmacologists at the meeting thought that a drug with such a steep dose-response curve would probably be unmanageable in a clinical situation. In this situation plasma levels of such a substance would only rarely fall into the narrow range between toxic and sub-therapeutic concentrations' (Lewis, 1981). While rate-controlled drug delivery systems are no panacea, they provide a new means for approaching and sometimes solving just such problems, opening the way for powerful, potentially valuable, but otherwise intractable, agents to be developed into practical pharmaceutical preparations.

Acknowledgements The author is indebted to Constance Mitchell for editorial assistance and to Rose Wright for bibliographical assistance.

References ALZA Corporation: The OCUSER~ (Pilocarpine) Pito-20t Pilo-40 Ocular Therapeutic System: A New Approach to Treating Ocular Hypertension. A Monograph (ALZA Corporation, Palo Alto, California 1974).

Rate-controlled Drug Dosage

ALZA Corporation: The PROGESTASERT® Intrauterine Progesterone Contraceptive System: A New Contraceptive. A Monograph (ALZ~ Corporation, Palo Alto, California 1976). Anonymous: Fatigue as an unwanted effect of drugs. Lancet I: 1285-1286 (1980). Benagiano, G.; Schmitt, E.; Wise, D. and Goodman, M.: Sustained release hormonal preparations for the delivery of fertility-regulating agents; in Sedhicek et al. (Eds) Medicinal Polymers: Chemical Problems, pp.129-148 (Journal of Polymer Science: Polymer Symposia 66) [John Wiley and Sons, New York 1979]. Blackshear, PJ.; Rohde, T.D.; Prosl, F. and Buchwald, H.: The implantable infusion pump: A new concept in drug delivery. Medical Progress through Technology 6: .149-161 (1979). Bleyer, W.A.: The clinical pharmacology of methotrexate: New applications of an old drug. Cancer 41: 36-51 (1978). Box, G.E.P. and Jenkins, G.M.: Time Series Analysis: Forecasting and Control (Holden-Day Publishers, San Francisco, California 1976). Brand, J.1. and Whittingham, P.: Intramuscular hyoscine in control of motion sickness. Lancet 2: 232-234 (1970). Brown, H.S.; Meltzer, G.; Merrill, R.C.; Fisher, M.; Ferre, C. and Place, V.A.: Visual effects of pilocarpine in glaucoma: Comparative study of administration by eyedrops or by ocular therapeutic systems. Archives of Ophthalmology 94: 1716-1719 (1976). Buchwald, H.; Rohde, T.D.; Schneider, P.D.; Varco, R.L. and Blackshear, P.J.: Long-term, continuous intravenous heparin administration by an implantable infusion pump in ambulatory patients with recurrent venous thrombosis. Surgery 88: 507-516 (1980). Cabana, RE.: Guidelines for evaluating the bioavailability of controlled release dosage forms; presented at the American Pharmaceutical Association, Academy of Pharmaceutical Sciences, Midwest Regional Meeting, Chicago, Illinois, May 1979. Carruthers, S.G.; Shoeman, D.W.; Hignite, C.E. and Azarnoff, D.L.: Correlation between plasma diphenhydramine level and sedative and antihistamine effects. Clinical Pharmacology and Therapeutics 23: 375-382 (1978). Chandrasekaran, S.K; Benson, H. and Urquhart, J.: Methods to achieve controlled drug delivery: The biomedical engineering approach; in Robinson (Ed.) Sustained and Controlled Release Drug Delivery Systems, pp.557-593 (Marcel Dekker, New York 1978a). Chandrasekaran, S.K.; Capozza, R. and Wong, P.S.L.: Therapeutic systems and controlled drug delivery. Journal of Membrane Science 3: 271-286 (I 978b). Choi, N.S.: Drug delivery by bioerodible polymer system. Pollimo (Polymer) 4: 35- 39 (1980). Crowley, W.F., Jr; Comite, F.; Vale, W.; Rivier, J.; Loriaux, D.L. and Cutler, G.B., Jr: Therapeutic use of pituitary desensitization with a long-acting LHRH agonist: A potential new treatment for idiopathic precocious puberty. Journal of Clinical

223

Endocrinology and Metabolism 52: 370-372 (1981). De Jongh, S.E. and Wijnans, M.: The influence of divided doses of drugs on the duration of effect and integral of effect. Acta Physiologica et Pharmacologica I: 237-255 (1950). De Leede, L.GJ.; De Boer, A.G. and Breimer, D.O.: Rectal infusion of the model drug antipyrine with an osmotic delivery system. Biopharmaceutics and Drug Disposition 2: 131-136 (J 98 I). DiSanto, A.R.: Bioavailability and bioequivalency testing; in Osol (Ed.) Remington's Pharmaceutical Sciences, 16th ed., pp.1369-1377 (Mack Publishing Co., Easton, Pennsylvania 1980). Eckenhoff, B.; Theeuwes, F. and Urquhart, J.: Osmotically actuated dosage forms for rate-controlled drug delivery. Pharmaceutical Technology 5: 35-44 (J 98 I). Fylling, P. : Disappearance rate of progesterone following simultaneous removal of the corpus luteum and the foeto-placental unit in women. Acta .Endocrinologica 65: 284-292 (1970). Gale, R.; Chandrasekaran, S.K; Swanson, D. and Wright, J.: Use of osmotically active therapeutic agents in monolithic systems. Journal of Membrane Science 7: 319-331 (1980). Gibaldi, M.; Levy, G. and Weintraub, H.: Drug distribution and pharmacologic effects. Clinical Pharmacology and Therapeutics 12: 734-742 (1971). Godbillon, 1.; Richard, 1.; Gerardin, A.; Moppert, J.; Leroy, D. and Theeuwes, F.: Plasma concentration profiles of metoprolol achieved after administration of oral osmotic controlledrelease systems (OROS); in Abstracts of Papers Presented at the I st European Congress of Biopharmaceutics and Pharmacokinetics, Clermont-Ferrand, France, p.163, 1-3 April (198 I). Graybiel, A.; Knepton, J. and Shaw, J.E.: Prevention of experimental motion sickness by means of scopolamine absorbed through the skin. Aviation, Space, and Environmental Medicine 47: 1096-1100 (1976). Greco, T.P.; Bonadio, M.; Lee, R.V. and Andriole, V.T.: Minocycline toxicity: Experience with an altered dosage regimen. Current Therapeutic Research 25: 193-201 (1979). Greenblatt, OJ. and Shader, R.1.: Drug therapy: Anticholinergics. New England Journal of Medicine 288: 1215-1219 (1973). Greenblatt, OJ. and Shader, R.1.: Dependence, tolerance, and addiction to benzodiazepines: Clinical and pharmacokinetic considerations. Drug Metabolism Reviews 8: 13-28 (1978). Haynes, R.B.: The effect of the therapeutic regimen on patient compliance and the possible influence of controlled-release dosage forms; in Urquhart (Ed.) Controlled Release Pharmaceuticals, pp.121-139 (American Pharmaceutical Association, Academy of Pharmaceutical Sciences, Washington, D.C. 1981). Heilmann, K: Therapeutic Systems - Pattern-Specific Drug Delivery: Concept and Development (G. Thieme, Stuttgart 1978).

Rate-controlled Drug Dosage

Holmstedt, B. and Liljestrand, G. (Eds): Paracelsus, 1493-1541; in Readings in Pharmacology (Pergamon Press, New York 1963). Hoppener, RJ.; Kuyer, A.; Meijer, J.W.A. and Hulsman, J.: Correlation between daily fluctuations of carbamazepine serum levels and intermittent side effects. Epilepsia 21: 341-3 50 ((980). Hwang, B.; Konicki, G.; Dewey, R. and Miao, c.: GLC determination of ticrynafen and its metabolites in urine, serum, and plasma of humans and animals. Journal of Pharmaceutical Sciences 67: 1095-1098 (( 978). Inch, T.D.; Green, D.M. and Thompson, P.B.M.: The central and peripheral activities of anti-acetylcholine drugs. Some concepts of practical relevance. Journal of Pharmacy and Pharmacology 25: 259-270 ((973). Jaminet, F.: Pharmacie galenique moderne et formes pharmaceutiques nouvelles. Journal de Pharmacie de Belgique 35: 200-222 (( 980). Johnson, J.c. (Ed.): Sustained Release Medications (Noyes Data Corp., Park Ridge, New Jersey 1980). Jones, B.R.; Anderson, J. and Fuglsang, H.: Effects of various concentrations of diethylcarbamazine citrate applied as eye drops in ocular onchocerciasis, and the possibilities of improved therapy from continuous non-pUlsed delivery. British Journal of Ophthalmology 62: 428-439 (I 978). Kitrenos, J.G.; Jones, M. and Mcleod, D.C.: Comparison of selected intravenous infusion pumps and rate regulators. American Journal of Hospital Pharmacy 35: 304-310 (( 978). Koobil, E.: The neuroendocrine control of the menstrual cycle. Recent Progress in Hormone Research 36: 53-88 (( 980). Koch-Weser, J.: Serum drug concentrations as therapeutic guides. New England Journal of Medicine 287: 227-231 ((972). Koch-Weser, J.: Serum drug concentrations in clinical perspective. Therapeutic Drug Monitoring 3: 3-16 (I 98 Il. Kydonieus, A.F. (Ed.): Controlled Release Technologies: Methods Theory, and Applications, Vols I and 2 (CRC Press, Boca Raton, Florida 1980). Laibson, P.R. and Leopold, I.H.: An evaluation of double-blind IDU therapy in 100 cases of herpetic keratitis. Transactions of thl; American Academy of Ophthalmology 68: 22-34 (( 964). Leeper, H.M.; Buckies, R.G.; Guittard, G.V.; Lorberbaum, M.A.; Sevilla, E. R. and Yum, S.l.: Role of the elasticity of rubber in the controlled release administration of drugs. Rubber Chemistry and Technology 50: 969-980 (( 977). Lerman, S. and Reininger, B.: Simulated sustained release pilocarpine therapy and aqueous humor dynamics. Canadian Journal of Ophthalmology 6: 14-23 (( 97 t). Lewis, P J.: Clinical pharmacology of prostacyclin. Trends in Pharmacological Sciences 2: iv-v (( 98 D. Leyendecker, G.; Wildt, L. and HansmalU1, M.: Pregnancies following chronic intermittent (pulsatile) administration of Gn-RH by means of a portable pump ('Zykiomar) - A new approach to the treatment of infertility in hypothalamic

224

amenorrhea. Journal of Clinical Endocrinology and Metabolism 51: 1214-1216 ((980). Uppas, J.: Continuous irrigation in the treatment of external ocular diseases. American Journal of Ophthalmology 57: 298-305 (1964). Lowenthal, D.T.; Onesti, G.; Mutterperl, R.; Affrime, M.; Martinez, E.M.; Kim, K.E.; Busby, J.; Shirk, J. and Swartz, C.: Long-term clinical effects, bioavailability, and kinetics of minoxidil in relation to renal function. Journal of Clinical Pharmacology 18: 500-508 (( 978). Lutterman, J.A.; Adriaansen, A.AJ. and van't Larr, A.: Treatment of severe diabetic ketoacidosis: A comparative study of two methods. Diabetologia 17: 17-21 (( 979). Lynch, HJ.; Rivest, R.W. and Wurtrnan, R.J.: Artificial induction of melatonin rhythms by programmed microinfusion. Neuroendocrinology 31: 106-111 (( 980). Martucci, c.P. and Fishman, 1.: Impact of continuously administered catechol estrogens on uterine growth and luteinizing hormone secretion. Endocrinology 105: 1288-1292 ((979). Michaels, A.S.; Mader, WJ. and Manning, C.R.: New concepts and standards of quality control as applied to controlled drug delivery systems; in Deasy and Timoney (Eds) The Quality Control of Medicines, W.45-54 (Elsevier, Amsterdam 1976). Mishell, D.R., Jr and Martinez-Manautou, 1. (Eds): Proceedings of the Symposium on Clinical Experience with the Progesterone Uterine Therapeutic System, 15-16 October 1976, Acapulco (Excerpta Medica, Amsterdam 1978). Murphy, 1.; Casey, W. and Lasagna, L.: The effect of dosage regimen on the diuretic efficacy of chiorothiazide in human subjects. Journal of Pharmacology and Experimental Therapeutics 134: 286-290 (( 96 D. Myers, R.D. (Ed.}. Chronic methods: Intraventricular infusion, cerebrospinal fluid sampling, and push-pull perfusion; in Methods in Psychobiology: Advanced Laboratory Techniques in Neuropsychology and Neurobiology, Vo\. 3, p.315 (Academic Press, New York 1977). Pettigrew, J.D. and Kasamatsu, T.: Local perfusion of noradrenaline maintains visual cortical plasticity. Nature 271: 761-763 (\978). Pickup, J.c.; Keen, H.; Parsons, J.A.; Alberti, KG.M.M. and Rowe, A.S.: Continuous subcutaneous insulin infusion: Improved blood-glucose and intermediary-metabolite control in diabetics. Lancet I: 1255-1257 (( 979). Pippenger, C.E.: Rational and clinical application of therapeutic drug monitoring. Pediatric CliniCs of North America 27: 891-925 ((980). Place, V.A.; Fisher, M.; Herbst, S.; Gordon, L. and Merrill, R.C.: Comparative pharmacologic effects of pilocarpine administered to normal subjects by eyedrops or by ocular therapeutic systems. American Journal of Ophthalmology 80: 706-712 (1975). Prescott, L.F.: Clinical limitations of controlled-release dosage

Rate-controlled Drug Dosage

forms; in Urquhart (Ed.) Controlled-Release Pharmaceuticals. pp.51-59 (American Pharmaceutical Association. Academy of Pharmaceutical Sciences. Washington. D.C 1981). Prescott. L.F.; Steel. R.F. and Ferrier. W.R.: The effects of particle size on the absorption of phenacetin in man. Clinical Pharmacology and Therapeutics II: 496-504 (1970). Richens. A. and Warrington. S.: When should plasma drug levels be ,monitored? Drugs 17: 488-500 (1979). Robinson, J.R. (Ed.): Sustained and Controlled Release Drug Delivery Systems (Marcel Dekker, New York 1978). Rogers, J.D. and Kwan, KC: Pharmacokinetic requirements for controlled-release dosage forms; in Urquhart (Ed.) ControlledRelease Pharmaceuticals, pp.95-119 (American Pharmaceutical Association, Academy of Pharmaceutical Sciences. Washington, D.C 1981). Rogers, J.D.; Lee. R.B.; Webster, P.R.; Ferguson, R.K.; Davies. R.O.; Theeuwes. F. and Kwan, KC: Osmotically controlled indomethacin delivery in man; in Abstracts of Papers Presented at the I st European Congress of Biopharmaceutics and Pharmacokinetics, Clermont-Ferrand, France, p.168, 1-3 April (1981). Rutter, P.c.; Murphy, F. and Dudley, H.A.F.: Morphine: Controlled trial of different methods of administration for postoperative pain relief. British Medical Journal 280: 12-13 (1980). Salem, P.; Hall, S.W.; Benjamin, R.S.; Murphy, W.K; Wharton, J.T. and Bodey, G.P.: Clinical phase I-II study of cisdichlorodiammineplatinum (JI) given by continuous i.v. infusion. Cancer Treatment Reports 62: 1553-1555 (1978). Salzman, E.W.; Deykin, D.; Shapiro, R.M. and Rosenberg, R.: Management of heparin therapy. New England Journal of Medicine 292: 1046-1050 (1975). Shaw, J. and Urquhart, 1.: Programmed, systemic drug delivery by the transdermal route. Trends in Pharmacological Sciences I: 208-211 (1980). Sikic, B.I.; Colins, J.M.; Mimnaugh, E.C and Gram, T.E.: Improved therapeutic index of bleomycin when administered by continuous infusion in mice. Cancer Treatment Reports 62: 2011-2017 (1978). Smits, J.F.M.; Van Essen, J. and Struyker-Boudier, H.A.J.: Is the antihypertensive effect of propranolol caused by an action within the central nervous system? Journal of Pharmacology and Experimental Therapeutics 215: 172-175 (1980). Southwest Oncology Group: Cytarabine for acute leukemia in adults: Effects of schedule on therapeutic response. Archives of Internal Medicine 133: 251-259 (1974). Spangler, D.L; Kalof, D.D.; Bloom, F.L and Wittig, H.J.: Theophylline bioavailability following oral administration of six sustained-release preparations. Annals of Allergy 40: 6-11 (1978). Tamborlane, W.V.; Sherwin, R.S.; Genel, M. and Felig, P.: Restoration of normal lipid and amino acid metabolism in diabetic patients treated with a portable insulin-infusion pump.

225

Lancet I: 1258-1261 (1979). Theeuwes, F.: Elementary osmotic pump. Journal of Pharmaceutical Sciences 64: 1987-1991 (1975). Theeuwes. F. and Bayne, W.: Dosage form index: An objective criterion for evaluation of controlled-release drug delivery systems. Journal of Pharmaceutical Sciences 66: 1388-1392 (1977). Theeuwes, F. and Bayne, W.: Controlled-release dosage form design; in Urquhart (Ed.) Controlled Release Pharmaceuticals, pp.61-93 (American Pharmaceutical Association, Academy of Pharmaceutical Sciences, Washington, D.C 19811. Theeuwes, F. and Eckenhoff, B.: Applications of osmotic drug delivery; in Baker and Good (Chairmen) Proceedings of the 6th International Symposium on Controlled Release of Bioactive Materials, New Orleans, Louisiana, pp.I-80, 6-8 August (1979). Theeuwes, F. and Yum, S.I.: Principles of the design and operation of generic osmotic pumps for the delivery of semisolid or liquid drug formulations. Annals of Biomedical Engineering 4: 343-353 (1976). Theeuwes, F.; Bayne, W. and McGuire, J.: Gastrointestinal therapeutic system for acetazolamide: Efficacy and side effects. Archives of Ophthalmology 96: 2219-2221 (1978). Urquhart, J.: Development of the OCUSERTe pilocarpine ocular therapeutic systems - A case history in ophthalmic product development; in Robinson (Ed.) Ophthalmic Drug Delivery Systems, pp.1 05-118 (American Pharmaceutical Association, Washington, D.C. 1980). Urquhart, 1.: Performance requirements for controlled-release dosage forms: Therapeutic and pharmacological perspectives; in Urquhart (Ed.) Controlled-Release Pharmaceuticals, pp.I-48 (American Pharmaceutical Association, Academy of Pharmaceutical Sciences, Washington, D.C. 19811. US Food and Drug Administration: Summary for basis of approval, NDA 17-784, TRANSDERMe transdermal therapeutic system-scopolamine (1980). Vistnes, L.M.; Schmitt, E.E.; Ksander, G.A.; Rose, E.H.; Balkenhol, W.J. and Coleman, C. L.: Evaluation of a prototype therapeutic system for prolonged, continuous topical delivery of homosulfanilamide in the management of Pseudomonas bum wound sepsis. Surgery 79: 690-696 (1976). Wagner, J.G.: Fundamentals of Clinical Pharmacokinetics (Drug Intelligence Publications, Hamilton, Illinois 1975). Weeks, J .R.: A method for administration of prolonged intravenous infusion of prostacyclin (PGI,) to unanesthetized rats. Prostaglandins 17: 495-499 (1979). Wei, E. and Loll, H.: Physical dependence on opiate-like peptides. Science 193: 1262-1263 (1976). Weintraub, M.; Jacox, R.F.; Angerine, C.D. and Atwater, E.C.: Piroxicam (CP 161711 in rheumatoid arthritis: A controlled clinical trial with novel assessment techniques. Journal of Rheumatology 4: 393-404 (1977). Wilson, T.W.; Falk, K.J.; Labelle, J.L. and Nguyen, KB.: Effect

226

Rate-controlled Drug Dosage

of dosage regimen on natriuretic response to furosemide. Clinical Pharmacology and Therapeutics 18: 165-169 (( 975). Wright, J.e.; Buckles, R.G.; Dunn. J.T.; Leeper. H.M. and Yum. S.L An elastomeric micrometering system for the controlled administration of drugs. Rubber Chemistry and Technology 50: 959-968 (t 977). Yates, F.E.; Benson, H.; Buckles, R.; Urquhart, J. and Zaffaroni, A.: Engineering development of therapeutic systems: A new class of dosage forms for the controlled delivery of drugs; in Brown and Dickson (Eds) Advances in Biomedical Engineering, Vol. 5, pp.I-34 (Academic Press, New York 1975).

Zaffaroni, A.: Industrial development of controlled delivery drug systems; in International Symposium on Science, Invention and Social Change. pp.73-lIO (General Electric Co. Schenectady 1978). ZWagemakers, J.M.A. and Claassen. V.: Secoverine selectively antagonizes muscarinic effects in various in vivo preparations. European Journal of Pharmacology 71: 165-168 (t 98 I).

Author's address: Dr John Urquhart. ALZA Research, Palo Alto. California 94304 (USA).

Libraries everywhere have found the easy way to fill photocopy requests legally and instantly, without the need to seek permissions, from this and over 3000 other key publications in bUSiness, science, humanities, and social science. PartiCipation in the Copyright Clearance Center (CCC) assures you of legal photocopying at the moment of need. You can: Fill requests for multiple copies, interlibrary loan (beyond the CONTU guidelines), and reserve desk without fear of copyright infringement. Supply copies simply and easily from registered publications. The CCC's flexible reporting system accepts photocopying reports and returns an itemized invoice. You need not keep any records, our computer will do it for you. The Copyright Clearance Center is your one-stop place for on-the-spot clearance to photocopy lor internal use. You will never have to decline a photocopy request or wonder about compliance with the law for any publication registered with the CCC. For more information, just contact:

Copyright Clearance Center 21 Congress Street Salem. Massachusetts 01970 (617) 744-3350 a not-for-profit corporation