Principles and Practice of Pharmaceutical Medicine

isoenzymes simultaneously. Barbiturates (barbi-
tals), rifampicin and cigarette smoking are all
well-known enzyme inducers, and can affect the
metabolism of a widevariety of drugs. A redundant
nonsteroidal agent (known as antipyrine), was the
classical probe drug for enzyme induction, and its
metabolism is increased compared to baseline
when a 14-day challenge by an enzyme-inducing
drug is administered. Modern ‘cocktail’ studies
have now superseded antipyrine.
Substrate ‘cocktails’ are now used so as to effi-
ciently study the effect of the test drug on several
CYP450 isoenzymes at once. The cocktail com-
prises a mixture of drugs where each is metabo-
lized wholly by a sole and different isoenzyme.
Several established ‘cocktails’ have been pub-
lished. For example, the ‘Indiana Cocktail’ con-
tains (isoenzyme) caffeine (1A2), tolbutamide
(2C9), dextramethorphan (2D6) and midazolam
(3A4). All such studies need to have adequate
monitoring for safe administration of drugs in
place – such as oxygen saturation monitoring
when midazolam is given. Acute attention to detail
and timing is vital in order to obtain reliable,
interpretable results. In particular, the timing of
blood sampling for drug concentrations must be
carefully designed in accordance with the known
pharmacokinetic profiles of each drug adminis-
tered and executed with the greatest precision.
For studies of CYP450 isoforms involving large
phenotypical differences in humans, prescreened

volunteers, with known isoenzyme activity and

capacity, are needed. In this way, experimentally

induced extreme plasma concentrations and the

consequent clinical hazard can be avoided (e.g.

due to preexisting slow metabolism).

Two typical phase I drug interaction clinical trial

designs are shown in Figure 20.4. Pharmacokinetic

profiles are found within each subject for each

substrate with and without concomitant exposure

to the study drug. These designs are applicable

to both enzyme induction and inhibition effects

(Figure 20.4, upper half). The same study sche-

matic can be used to study the effects of inhibitors

or inducers on the study drug as a substrate for

CYP450 systems (as illustrated in the lower half of

Figure 20.4).

Some drug interaction studies must be done in

patients and use longer durations of exposure, and

these are usually conducted in small groups of

patients (rarely more than 12). These investigate

not only pharmacokinetic interaction considera-

tions but also the potential for interference with

the efficacy of a proven agent. For example, drugs

in development for rheumatoid arthritis must

undergo an interaction study with methotrexate

prior to the execution even of phase II clinical

trials, because it is essential to assure that the

new agent neither interferes with the therapeutic

effect of methotrexate nor potentiates its adverse

effects. These studies are more complex and need

to be designed carefully on a case-by-case basis.

Substrate Study drug

Study drug Inducer

Study drug

and Substrate

Inducer/Inhibitor

and Study drug

Figure 20.4 Two typical designs for a phase I drug interaction clinical trial. The horizontal arrow represents time, at a
scale of several days. Above the horizontal line is a typical screening study, where the test medication (Study drug) is
being screened for any sort of interaction (inhibitory or inducing) with a known isoenzyme substrate; note that a
cocktail of several substrates can also be used with this design. Below the horizontal arrow is a study design testing
whether elimination of the test medication is itself susceptible to enzyme induction or inhibition by some other drug
(inducer/inhibitor); note that the roles of the known and unknown drugs have essentially been reversed

262 CH20 DRUG INTERACTIONS