TOWARDS BETTER SAFETY OF DRUGS
`AND PHARMACEUTICAL PRODUCTS
`
`Proceedings of the 39th International Congress of
`Pharmaceutical Sciences of F.I.P., held in Brighton, U.K.,
`September 3-7, 1979.
`
`Editor
`D.D. BREIMER
`
`Secretary of the Board of Pharmaceutical Sciences
`F#d#ration Internationale Pharmaceutique (F.I.P.)
`
`Professor of Pharmacology, Department of Pl~armacology and Therapeutics,
`Subfaculty of Pharmacy, University of Leiden, The Netherlands
`
`1980
`
`ELSEVIER/NORTH-HOLLAND BIOMEDICAL PRESS
`AMSTERDAM NEW YORK . OXFORD
`
`Lupin Ex. 1041 (Page 1 of 18)
`
`

`

`© 1980 Elsevier/North-Holland Biomedical Press
`
`All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
`or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording
`or otherwise, without the prior permission of the copyright owner.
`
`ISBN: 0-444-80216-9
`
`Published by:
`Elsevier/North-Holland Biomedical Press
`335 Jan van Galenstraat, P.O. Box 211
`Amsterdam, The Netherlands
`
`Sole distributors for the U.S.A. and Canada:
`Elsevier NOrth-Holland Inc.
`52 Vanderbilt Avenue
`New York, N.Y. 10017
`
`Library of Congress Cataloging in Publication Data
`
`International Congress of Pharmaceutical Sciences, 39th,
`Brighton, Eng,, 1979,
`Towards better safety of drugs ~[d pharmaceutical
`products.
`
`Bibliography: p,
`Includes index.
`1. Drug trade--Quality control--Congresses.
`2. Drugs--Testing--Congresses. 3. Drugs--Safety mea-
`sures--Congresses. I. Breimer, Douwe D., 19~3-
`II. F~d~r&tion international pharmaceutique. III.
`Title. EDNLM: 1. Drug therapy--Adverse effects--Con-
`gresses, 2. Drug evaluation--Congresses. 3. Phar-
`macology--Congresses. W3 IN541 39th 1979t / QZ42 I61
`1979t3
`R8189.I53 1979
`ISBN 0-44~-80216-9
`
`80-11924
`
`363.1’94
`
`Printed in The Netherlands
`
`Lupin Ex. 1041 (Page 2 of 18)
`
`

`

`This m~tedal may be protected by Copyright law (1 t e 17 U,S. Code)
`
`@ 1980 Elsevier/North-Holland Biomedical Press
`Towards Better Safet)/ of Drugs ~nd Pharmaceutical Products
`
`D.D. Breimer, editor
`
`265
`
`THE PROBLEMS OF DRUG INTERACTIONS WITH EXCIPIENTS
`
`P. HEINRICH STAHL
`Pharmaceutical Development,
`CIBA-GEIGY Ltd., CH-4002 Basle (Switzerland)
`
`INTRODUCTION
`Drug substances are molecules with a variety of functional
`groups, polar and non-polar, exhibiting hydrophilic and hydropho-
`bic properties. The specific structure of a drug determines its
`actions and reactions within the biophase where the interactions
`with target receptors result in the desired therapeutic effects.
`Between a drug and its respective environment many interactions
`are possible. A series of interact±ons is summarized in Figure i.
`
`Bulk
`
`Processing
`
`Dosage Form Mixing of
`Parenterals
`
`Absorption D[stribu- Site of
`Action
`tion
`
`Elimination
`
`MANUFACTURE
`
`STORAGE
`
`PRE-ADMIN-
`ISTRATION
`
`BODY
`
`= desired interaction with drug specific receptors
`
`Fig. i. Possible interactions of a drug substance with various
`factors during manufacture, storage, and use.
`
`Here we also find the group of interactions we have to deal with
`now. It would be surprising indeed if drugs were not to interact
`with non-biological materials as well, including pharmacodynami-
`cally inert materials which are indispensable for making dosage
`forms.
`
`Lupin Ex. 1041 (Page 3 of 18)
`
`

`

`266
`
`During the past 25 years or so, knowledge about drug-excipient
`interactions has increased tremendously. In the following, I shall
`try to give a systematic view of this area spiced with some per-
`sonal experiences~ We shall also have to give answers regarding
`the question about the importance of these interactions.
`
`TYPES OF INTERACTIONS
`According to their nature, the interactions between drugs and
`
`excipients may be classified into
`
`physical and physicochemical
`chemical
`biological
`
`interactions. Table 1 gives a synoptic view over these classes
`and their distinguishing features.
`
`TABLE i
`CLASSES AND FEATURES OF DRUG - EXCIPIENT INTERACTIONS
`
`
`
`Interaction ~Action Drug
`
`Normal drug action
`
`No excipient action
`
`Drug A
`
`Excipient E
`
`Weak physicochemical
`interaction
`
`A + E
`
`Physicochemical inter.action
`
`A+E
`
`Chemical interaction
`
`A + E
`
`Biological interaction
`
`A + E
`
`in vitro I found by method
`
`in vivo
`
`n
`®
`
`=!"
`
`~=
`
`~"
`
`m
`
`physicochemical
`
`physicochemical
`
`; l~
`
`analytical
`
`~
`
`~
`
`I ~
`
`normal
`
`no activity
`
`normal
`
`decreased
`increased
`no action
`
`decreased
`side effects
`toxic effects
`
`side effects
`toxic effects
`
`PHYSICAL AND PHYSICOCHEMICAL INTERACTIONS
`To begin with, physical and physicochemial interactions to a
`large extent not only constitute the basis of development and
`manufacture of dosage forms but also of absorption. Some typical
`
`Lupin Ex. 1041 (Page 4 of 18)
`
`

`

`267
`
`TABLE 2
`SOME PHYSICAL AND PHYSICOCHEMICAL INTERACTIONS
`
`Absorption
`
`Deliquescence
`
`Salt Formation
`
`Ad hesion
`
`Adsorption
`
`Binding
`
`Dissolution
`
`Segregation
`
`Emulsification
`
`Solubilization
`
`Inclusion
`
`Solvatization
`
`Coagulation
`
`Ionisation
`
`Complexation
`
`Precipitation
`
`Spreading
`
`Swelling
`
`interactions are listed in Table 2. It is almost commonplace to
`mention the dissolution of a drug in water or in another solvent
`as an example of a desired interaction. On the other hand, there
`are interactions which delay the rate and reduce the extent of
`release and absorption. Here it is of minor importance where to
`draw the border-line between physical and physicochemical inter-
`actions. Hydrophobization of solid sulfadiazine by admixture of
`magnesium stearateI is regarded as a physical process whereas for-
`mation of a sparingly soluble salt of chloroquine with carboxyme-
`thylcellulose2 is a physicochemical process because ionic inter-
`
`actions take place. Both these interactions reduce the release
`rate in vitro, and possibily the rate and extent of absorption.
`Two other examples illustrate the many faces of physical and
`physicochemical interactions. They demonstrate that by the same
`interactive principle the bioavailability is reduced in one case
`but is improved in the other.
`Boman et al.3 encountered an interference of a preparation with
`p-aminosalicylic acid (PAS) with another drug preparation contain-
`ing rifampicin (RMP) as the active principle, by monitoring the
`blood levels of the latter in humans. The reduced blood levels of
`RMP in patients treated with both these drugs simultaneously could
`be traced back to the adsorption of RMP onto bentonite which served
`as an excipient in the PAS granules. This adsorption process takes
`place in the stomach. The figures of Table 3 show a reduction of
`the available RMP by 1/3. No RMP could be recovered from RMP-loaded
`bentonfte when desorption was attempte~ with 0.i N HCl i__n yitro.
`This is also an example of a cross-interaction between the ac-
`tive substance of a product A with an excipient in product B; sueh
`a situation could not have been foreseen by the manufacturers ofthe
`
`Lupin Ex. 1041 (Page 5 of 18)
`
`

`

`268
`
`TABLE 3
`REDUCTION OF RIFAMPICIN BIOAVAILABILITY
`in the presence of bentonite in p-aminosalicylic acid granules
`
`RMP
`simultaneously
`administered with
`
`RMP in plasma
`AUC~ (#g ¯ ml-~" h)
`5 subjects, mean +_- s.d.
`
`PAS granules
`(with bentonite)
`
`PAS-Na tablets
`(without bentonite)
`
`P~acebo granules
`(with I~entonite)
`
`67 + 15 (control, 107 .+_ 11)
`
`105 +27 (control, 107 + 11)
`
`73 + 7 (control, 108 + 19)
`
`RMP, rifampicine; PAS, p-aminosalicylic acid
`
`individual drug preparations. In addition, such a case may well
`serve as an argument to justify the development of combinations at
`least of drugs which therapeutically are used together.
`On the other hand, the same type of interaction, i.e. a drug
`adsorbed on bentonite, was successfully employed as a retarding
`principle without loss of bioavailability. Utilizing an adsorbate
`~i
`
`, : 20 McGinity
`of amphetamine sulfate on bentonite, weight ratio
`and Lach4 performed a bioavailability study on 6 individuals. As
`shown in Figure 2 this preparation yielded a marked slow-release
`effect while retaining complete absorption as measured by urinary
`excretion.
`
`o = 15 mg ofdrug alone
`
`e= 15 mg of drug as 1 : 20 adsorbate
`
`1,2-
`
`E
`
`o_ 0.6_
`
`Fig. 2. Amphetamine excretion
`(urine) in a single subject
`following a single dose o.f am-
`phetamine sulfate and amphet-
`amine bentonite adsorbate, resp.
`(McGinity and Lach, 1977)
`
`5
`
`10
`
`15
`
`HOURS
`
`Lupin Ex. 1041 (Page 6 of 18)
`
`

`

`269
`
`Examples of complexation are to be found within the group
`of tetracycline antibiotics which are very fond of this passion
`(Table 4). A look at their structure with four rings featuring
`several functional substituents with acid and basic as well as
`chelating properties makes this easy to understand (Fig. 3). Here,
`chelating with metal ions will lead to reduced bioavailability so
`%hat excipients have to be chosen carefully.
`
`R1
`
`R2
`
`R3
`
`H
`
`H
`
`H
`
`Cl
`
`OH
`
`OH
`
`H
`
`H
`
`0
`
`OH
`
`Tetracycline
`
`Oxytetracycline
`
`Doxycycline
`
`OH
`
`H
`
`Chlortetracycline
`
`Fig. 3. Structure of some important tetracycline antibiotics.
`
`TABLE 4
`COMPLEXING AGENTS FOR TETRACYCLINE ANTIBIOTICS
`
`Metal Cations:
`
`Fe2+, Fe~+, Cu2+, Ni2+, Co2+,
`Zn2+, Mn2+
`
`Anions:
`
`Mg2+, Ca2+
`
`A~3+, Zr4+, 99Te4+
`
`Phosphate, Citrate
`Salicylate, p-Hydroxybenzoate
`Saccharate
`
`Neutral Substances:
`
`Caffeine, Urea, Thiourea
`
`Synthetic Polymers:
`
`Polyvinylpyrrolidone
`
`Biopolymers:
`
`Serum albumin, Globulins,
`Lipoproteins, RNA
`
`(DiJrckhe[mer, 1975)
`
`5
`
`Lupin Ex. 1041 (Page 7 of 18)
`
`

`

`270
`
`Nevertheless, chelate formation may occasionally provide a means
`of improving the bioavailability of a drug. According to Lach and
`coworkers6 the absorption of dicumarol is facilitated after treat-
`
`ment with magnesium compounds prior to administration, or by ad-
`ministration together with magnesium compounds. This absorption-
`enhancing effect of magnesium ions is due to chelate formation,
`Figure 4. Figure 5 shows the blood levels of dogs that were trea-
`
`2 Na~
`
`Dicumarol, Bishydroxycoumarin
`
`Dicumarol-Magnesium-Chelate
`
`Fig. 4. Structures of dicumarol and its magnesium chelate.
`
`..... Dicumarol + Mg(OH)2, 6+94
`
`..... DicumaroI-Mg-Chelate
`
`Control
`
`~ Control
`
`-~ 4o
`
`~ 2o
`
`~ 30-
`
`~ 20-
`
`~ 10_
`
`20 410 60 80 1 O0 0 20 40 60 80 100
`HOURS HOURS
`
`Fig. 5. Plasma concentration in dogs after oral administration
`of dicumarol (left and right, solid lines), dicumarol mixed with
`magnesium hydroxide 6 + 94 (left, dashed line), and dicumarol
`magnesium chelate (right, dashed line), resp. Mean of four dogs.
`(Akers,’ Lach and Fischer, 1973)
`
`Lupin Ex. 1041 (Page 8 of 18)
`
`

`

`271
`
`ted with neat dicumarol, with a mixture of dicumarol and magnesium
`hydroxide, and with the magnesium chelate of dicumarol. The enhan-
`7
`ced absorption was also confirmed in humans
`From these examples it is obvious that the term ’physical in-
`compatibility’ should be used with caution. We should not use it
`simply as a synonym for ’physical (or physicochemical) interaction’
`~s is frequently the case. ’Incompatib~y’ merely constitutes a
`relative assessment of an interaction with regard to a certain ex-
`pectation of stability, release rate or biovailability. What in
`one case happens to be an interfering precipitation, complexation,
`or adsorption may be utilized as a principle of controlled release
`of the same drug, as soon as the objective of dosage form design
`changes.
`
`I should now like to comment on the most frequently used exci-
`pient - water. The interactions of drug substances with water and
`water vapour play an outstanding role in pharmacy. This is certain-
`ly widely accepted as far as water is considered as solvent and
`vehicle in liquid and semi-solid dosage forms. Its role as an exci-
`pient in the preparation of solid dosage forms on the other hand
`is a more transient and hidden one.
`The interaction of drugs with water often results in the forma-
`tion of hydrates. According to our own experience about one third
`of drug substances is capable of hydrate formation. A few hydrates
`and their anhydrous counterparts are remarkably stable. Their rates
`of conversion are slow enough even in aqueous suspension. This
`applies, for example, to ampicillin and its trihydrate. I.n such a
`case, the hydrated and the anhydrous species may be characterized
`and treated just as though they were different salts. Problems
`arise, however, when the transition between anhydrous and hydrated
`form, and vice versa, takes place within the range of climatic
`conditions during manufacture and storage of the drug product.
`Drugs which interact only weakly with water, thus forming less
`stable hydrates, are therefore likely to cause more trouble.
`The well-known anticonvulsive carbamazepine forms a dihydrate
`which c~ystallizes as long needles. This dihydrate also forms
`slowly when carbamazepine is stored at a very high relative humi-
`dity. Figure 6 shows a record of the weight gain during the uptake
`of two mols of water. The water is lost again at 43 % r.h. or
`
`Lupin Ex. 1041 (Page 9 of 18)
`
`

`

`272
`
`2O
`
`2 rnol H20
`
`÷o
`
`0
`
`"
`
`~"~ I
`
`’
`
`5
`
`’ ’ ’ [
`- 10
`
`TIME {weeks}
`
`’
`
`’
`
`~
`
`’
`
`’
`
`’
`
`’
`
`I
`15
`
`Fig. 6. Formation of carbamazepine dihydrate from anhydrous
`carbamazepine in humid air, 23°C.
`
`lower at room temperature; at elevated temperature, the loss of
`water takes place at higher humidities as well.
`Carbamazepine tablets may be prepared by a wet granulation pro-
`cess. If the granules are not dried very thoroughly, fractions of
`the active ingredient may retain the water of crystallization° Any
`remaining traces of the dihydrate will then function as nuclei for
`
`ready rehydration as soon as the water vapour pressure is high
`enough or liquid water comes in contact with the tablet. The pre-
`sence of dihydrate nuclei may have two practical consequences. Re-
`hydration then occurs at a much lower humidity (Figure 7), which
`means that unfavourable storage conditions give rise to a slow
`recrystallization of the drug throughout the tablet. This results
`in a plaster-like solidification of the tablet by the inter-
`locking needles of the dihydrate. Otherwise, rehydration starts
`at the very first contact with aqueous liquids at the time of ad-
`ministration. If disintegration is not very fast, dihydrate cry-
`stallization may overtake disintegration and thus again render
`the tablet non-disintegrating. It is clear that 200 mg of a drug
`with an equilibrium solubility of 24 mg per i00 ml at 37°C is un-
`available when locked within such a tablet. Figure 8 shows the
`hydrate-covered surface of a tablet of neat carbamazepine after
`
`Lupin Ex. 1041 (Page 10 of 18)
`
`

`

`2O
`
`|
`
`~
`
`| I ,," 93%
`
`273
`
`2 mol H20
`
`IIi ..."
`o
`8o~ ...........
`/I ./ ,,,"
`......................
`
`0 I
`0
`
`’
`
`’ ’ , ,
`50
`100
`
`TIME (days)
`
`,
`
`,
`
`,
`
`1 50
`
`Fig. 7. Rehydration of carbamazepine after drying the dihydrate
`to 0.05 mol residual water content at 23°C, at 23°C and various
`relative humidities.
`
`Fig. 8. Surface of tablets of neat carbamazepine (SEM micro-
`photograhs). Left, fresh surface after compression; right, sur-
`face covered .with needles of the dihydrate which formed after
`wetting with water.
`
`Lupin Ex. 1041 (Page 11 of 18)
`
`

`

`274
`
`wetting with water, and, for comparison, the fresh surface of such
`a tablet.
`
`CHEMICAL INTERACTIONS
`It is easy to distinguish chemical from physicochemical inter-
`actions by defining that the latter leave the drug molecule chemi-
`cally intact whereas the former do not. Generally speaking, chemi-
`cal interactions must be avoided because they reduce the amount of
`drug. Products of reactions and degradation are at best biological-
`ly inactive° But they may also exhibit biological effects of their
`own and their presence in a drug preparation may then cause unwan-
`.ted effects in vivo.
`chemical interactions with excipients should be recognized in
`an early stage of dosage form development. Testing the chemical
`compatibility of a drug substance with excipients therefore con-
`stitutes an essential part of a preformulation programme. This is
`done by exposing mixtures of the drug with various excipients to
`stress Conditions such as heat, humidity, and light. After certain
`periods of time, the samples are inspected visually and analyzed
`for degradation products. Semi-quantitative thin-layer chromato-
`graphy will do the job quite effectively in most cases.
`In the class of chemical interactions, the following groups
`have been encountered:
`
`TABLE 5
`CHEMICAL DRUG-EXCIPIENT INTERACTIONS
`
`General Catalysis
`- Acceleration of a degradation reaction typica!
`of the drug
`
`Specific Catalysis
`- Formation of a specific degradation product of
`the drug which is not formed in the absence of
`the excipient
`
`Reaction
`- Formation of a reaction product between
`drug and excipient
`
`Degradation of the drug due to a certain quality
`of the excipient
`
`Interactions of the first type are common and include hydroly-
`sis of ester and amide links in drug substances. Depending on the
`
`Lupin Ex. 1041 (Page 12 of 18)
`
`

`

`275
`
`shape of the hydrolysis-rate - pH-profile of the individual active
`substance the degradation will be accelerated by excipients either
`of acid or of basic nature. A classical example is the hydrolysis
`of acetylsalicylic acid catalyzed by the lubricant magnesium stea-
`rate.
`My colleagues who have to formulate drugs are mainly inter-
`ested in stable compositions and fast results, and the question
`of why a certain excipient impairs the stability of a drug is
`of minor importance. Insight into the nature of excipient-indu-
`ted degradation, however, is useful to determine the boundary
`conditions of such reactions as well as of the use of the exci-
`pient. It is often rather difficult to pinpoint the cause of
`reaction. I should like to present some cases where this has been
`possible.
`Recently, exploratory compatibility tests were performed with
`CGP 4540, a yellow water-insoluble isothiocyanate (I) which is
`being investigated as an anthelminthic. Mixtures with tricalcium
`phosphate and with polyvinylpyrrolidone, respectively, under heat
`and humidity stress changed to an intense brick red. This red sub-
`stance was isolated and turned out to be the corresponding sym.
`thiourea (II).
`
`2 R-N=C=S ~ R-NH-C-NH-R
`[Caa(P04) 2] II
`S
`
`II
`
`Subsequently, we successfully employed tricalcium phosphate as a
`catalyst in benzene suspension to prepare more of this thiourea.
`In another case, we experienced an even more surprising example
`of a specific catalytic effect of an excipient. During a compati-
`bility study with an experimental antidepressant, CGP 6085 A (III),
`we oberserved, in the chromatograms, a new spot which consistently
`appeared in mixtures containing dicalcium hydrogenphosphate dihy-
`drate. This finding was somewhat unexpected since this tablet fil-
`ler had, in our experience, a rather good compatibility record.
`
`Lupin Ex. 1041 (Page 13 of 18)
`
`

`

`276
`
`A, O2, H~O
`[CaHPO4]
`
`x HCI
`
`TTI
`
`TV
`
`The reaction product (IV) turned out to be the result of an oxi-
`dative process that was gentle enough to leave the bas±c struc-
`ture of the active substance intact. This catalytic process can
`again be used to prepare enough of this material for any biologi-
`cal testing.
`Reactions between drug and excipient may result in a new sub-
`stance in which parts of the drug as well as of the excipient are
`joined. We observed the formation of esters of stearic acid with
`a drug having aliphatic hydroxyls. Ethylene glycol monosalicylate
`underwent transesterification with fatty acid esters of a cream
`base, forming the ethylene glycol myristate, palmitate and stea-
`rate.
`One of the great favourites among the fillers for solid dosage
`forms is lactose. However, as a sugar with a hemiacetal group it
`is capable of reacting with primary and secondary amines. Secon-
`dary and aromatic amines are able to form stable lactosides. With
`primary aliphatic amines, the reaction proceeds further on a com-
`plicated route including an Amadori rearrangement and finally arri-
`ves at brown-coloured polymers. These are formed from the amine
`with carbonyl compounds generated by the decomposition of the
`Amadori products. Since the majority of drugs are nitrogen bases
`or their salts, the formulator must carefully check whether or not
`the use of lactose - as well as of other sugars - is possible.
`The already mentioned antidepressant CGP 6085 A, the hydrochlo-
`ride of a secondary amine, may serve as an example. A mixture of
`this drug with lactose remained stable. The same was found with
`samples of the drug mixed with magnesium stearate. However, if all
`three were combined and exposed to the same stress conditions, the
`mixture became discoloured and the chromatogram revealed degrada-
`tion of the drug substance. The weakly alkaline magnesium stearate
`obviously liberates some free drug base which, in turn, r~adily
`reacts with the sugar. Free CGP 6085 base with lactose added indeed
`gives rise to a degradation similar to that of the ternary mixture
`
`Lupin Ex. 1041 (Page 14 of 18)
`
`

`

`277
`
`of CGP 6085 hydrochloride, lactose, and magnesium stearate. Similar
`results had been found with amphetamine sulfate and lactose8. In
`accordance with the results of the compatibility study, it was pos-
`sible to formulate tablets of CGP 6085 A that were chemically
`stable even with lactose as filler provided that the tablet mixture
`was lubricated with the neutral hydrated castor .oil instead of mag-
`nesium stearate.
`This case shows that a test of chemical compatibility may fur-
`nish misleading predictions if only binary drug-excipient mixtures
`are employed. In due, recognition of the limitations of binary mix-
`tures and the possib~ity of multiple interactions, a system of
`testing the influence of excipients on the stability of drugs had
`been developed. It features factorial design and a ’stepping up’
`9
`procedure as a possible route to stable solid dosage forms
`A very annoying type of incompatibility may be noticed only
`later during the production stage of a drug speciality. It is the
`influence of variations of excipient quality. Impurities introdu-
`ced by an excipient, or degradation products thereof, may decrease
`the stability of a drug. This should b~ considered when any changes
`occur such as modifications of the excipient manufacturing process~
`or changes of. the excipient source or of storage conditions. Exam-
`pies are
`
`- traces of formaldehyde in starch
`- traces of iron in talc
`- peroxides in propylene glycol and in polyglycols
`- benzaldehyde in benzylic alcohol.
`
`We have observed significant differences in the stability of an
`ester-type drug when two different qualities of tricalcium phos-
`phate were used in a solid preparation. The phosphate, the aqueous
`suspension of which was neutral, increased the rate of hydrolysis
`of the drug much more than did the slightly acidic quality (pH = 5).
`In the case of foreseeable changes it is therefore a good pre-
`ventive measure against unpleasant surprises to compare the che-
`mical compatibility of old and new qualities of an excipient.
`
`BIOLOGICAL INTERACTIONS
`The interactions we have dealt with so far are by definition
`those that can be found by physical, physiochemical and analy-
`
`Lupin Ex. 1041 (Page 15 of 18)
`
`

`

`278
`
`tical methods in vitro. The third class - the biological inter-
`actions - can be detected only by using living substrate.
`Let us consider a drug A with its typical pharmacokinetic and
`pharmacodynamic performance, and an excipient E which is without
`any effect when administered alone. In vitro, we cannot find any
`interaction by solubility, dissolution rate, dialysis, partition
`methods, etc., and yet we find the performance of drug A altered:
`we are then dealing with a biological interaction. I shall give
`an example to illustrate this point.
`i0
`Following up on earlier experiments of Levy with goldfish,
`GoudaII worked with pentobarbital and sodium dioctylsulfosuccinate
`
`(DOSS). In Figure 9 his results are depicted diagrammatically. The
`fish when put into water containing a low concentration of the bar-
`biturate turns over after some time as a result of the absorption
`and action of the barbiturate. T e time elapsing until the fish
`turns over is considerably short@r when a minute concentration
`
`120 min
`
`D
`
`Fig. 9. Effect of simultaneous (top) and subsequent (bottom
`treatment of goldfish with 0.003 % dioctylsulfosuccinate (D) on
`the action of 0.02 Z pentobarbital (P) in water. Times elapsed
`until the fish turn upside down (’overturn time’) are given.
`
`Lupin Ex. 1041 (Page 16 of 18)
`
`

`

`279
`
`0.003 Z) of DOSS is added to the barbiturate solution. The ’over-
`turn time’ is equally short when the fish is at first immersed ina
`solution of the surfactant, then in water for rinsing and finally
`in the drug solution. The shortening effect of pretreatment of the
`fish on the ’overturn time’ is independent of the duration of the
`exposure to the surfactant. The solubility of pentobarbital was
`not influenced by surfactant concentrations up to 0.i ~. Thus a
`physical interaction could be ruled out and the increased membrane
`permeability to the drug may be defined as a biological interaction.
`Higher concentrations, however, have a retarding effect on barbitu-
`rate absorption in rats and mice. This may be explained by the in-
`clusion of the drug in nonabsorbable micelle.s of the surfactan~.
`When talking about biological interactionsi we verge upon the
`area of drug-drug interactions or therapeutical interactions. How-
`ever, excipients are either biologically inactive or, if they are
`not free of effects of their own, their use is restricted in res-
`pect of concentration, quantity and route of administration in
`order to prevent such effects from becoming apparent. Therefore
`biological drug - excipient interactions are relatively rare
`events. They may be found especially in the group of surface acti-
`ve substances. Some of the typical effects of surfactants may be
`explained on a physicochemical basis. One of them is the well-
`known improvement in absorption by enhanced wettability and des-
`aggragation of hydrophobic and sparingly soluble drugs~ provided
`that submicellar concentrations of the surfactant are applied.
`Reduced absorption of sparingly soluble drugs in spite of enhanced
`apparent solubility and improved in vitro dissolution rate may take
`place in the case of concentrations higher than the critical micel-
`lar concentration of the surfactant. In addition to these physico-
`chemical interactions, the occurrence of biological interactions
`was demonstrated, as was shown in the example.
`
`Experiments with animals or animal organs may furnish important
`clues for interactions of drugs. But such cases may also be of
`rather an academic nature, and often enough the in Vivo pheno-
`menon boils .down to a physicochemical interaction. I feel that the
`limited transferability of results obtained in animals, let alone
`isolated organs, does not justify the use of still more animal ex-
`periments of doubtful value for dosage form development and for
`
`Lupin Ex. 1041 (Page 17 of 18)
`
`

`

`280
`
`demonstrating excipient effects; bioavailability studies in man
`will have to follow anyway. We, in the industry, are anyway forced
`to carry out more and more animal experiments. Instead, advantage
`should be taken of ~n vitro methods as much as possible.
`
`CONCLUSION
`I have tried to show that there are useful as well as unfavour-
`able interactions between drugs and excipients. Well planned ex-
`periments make the interactive potential of a drug evident. It is
`important, however, to assess the varying significance of inter-
`actions carefully. Putting the experimental results to work,
`taking advantage of useful interactions and avoiding adverse ones
`is one of the necessary prerequisites for the achievement of safe
`and reliable drug products.
`
`ACKNOWLEDGEMENTS
`I wish to thank Miss Ch. Br~cher for the preparation of the
`SEM microphotographs (Figure 8) and Dr. H. Hess for helpful discus-
`sions and suggestions. Also the advice of Mrs. E.-S. Krebs during
`the preparation of the manuscript is gratefully acknowledged.
`
`REFERENCES
`i. Ahmed, M. and Enever, R.P. (1978) Pharm. Acta Helv., 53, 358.
`2. Stampf, G. (1978) Pharmazie, 33, 356.
`3. Boman, G. Lundgren, P. and Stjernstr6m, G. (1975) Europ.
`J.clin. Pharmacol., ~, 293.
`4. McGinity, J. W. and Lach, J. L. (1977) JoPharm. Sci., 66, 63.
`5. D~rckheimer, W. (1975) Angew. Chem., 87, 751.
`6. Akers, M. J., Lach, J. L. and Fischer, L. J. (1973) J.Pharm.
`Sci., 62, 391.
`7. Ambre, J. J. and Fischer, L. J. (1978) Clin.Pharmac. Ther.,
`l_!, 2 1.
`8. Castello, R. A. and Mattocks, A. M. (1962) J.Pharm. Sci.,
`51, 106.
`9. a) Leuenberger, H. and Becher, W. (1975) Pharm. Acta Helv. ,
`88;
`b) Becher, W. (1978) in Sucker, Fuchs, H. and Speiser, P.
`Pharmazeutische Technologic, Thieme, Stuttgart, pp. 797 -
`801; 812.
`i0. Levy, G. and Anello, Jo A. (1969) J.Pharm. Sci., 58, 494.
`ii. Gouda, M. W. (1974) Can. J.Pharm. Sci. , ~, 37.
`
`Lupin Ex. 1041 (Page 18 of 18)
`
`