Principles and Practice of Pharmaceutical Medicine

meta-analytical comparisons of placebo-controlled
studies of different drugs, which were conducted at
different times and in different places. The general
aim is to compare the new drug with a widely
recognized ‘gold standard’. This ‘gold standard’
might be the prototypical drug in the same
pharmacological class (e.g. a clinical trial compar-
ing a new cephalosporin with an old one), or it could
be an hitherto dominant therapy or procedure (e.g.
comparing a proton pump inhibitor with an H 2 an-
tagonist, or conservative management with a new
drug versus surgery). Sometimes, a change in phar-
maceutical formulation may have occurred, and,
even after approval, there may be questions over
its superiority, patient preference or economic
advantage compared with the formulation that was
initially approved (see Makuch and Johnson, 1986,
1989).

Open-label studies

Conducting open-label studies can be a liberating
and fascinating experience. When both the patient
and the prescriber know the treatment being admi-
nistered, many of the complexities of early-phase
studies go away. Furthermore, when it is appre-
ciated that double-blind clinical trials are always
an abstraction from the ordinary clinical situation,
to observe how one’s new drug actually works in
that latter environment is often eye-opening; one
common and pleasant experience is to see with
one’s own eyes how conservative was the estimate
of product efficacy prior to its approval.
This ‘real-world’ environment can be studied at
length and relatively cheaply, too. Longitudinal
study designs (e.g. the Framingham Study or the
UK Physicians Cohort Study) can assess multiple
effects of treatment: pathological, economical,
quality of life and even epidemiological impacts
can be assessed. One can also find out what sort of
patient one’s drug will be prescribed to, which may
or may not resemble the patient population pre-
PLA/NDA, and which may suggest unknown ben-
efits and hazards of the new therapy.
The open-label trial approach is, however, not
without its critics. Friedmanet al. (1985) drew
attention to the need to observe whether

the cohort being followed represents the larger

population for whom the drug is being pre-

scribed;

the treatment groups are truly comparable, as

patients are often matched on only one or at

most a small number of clinical characteristics.

the need to check that randomization, or at

least patient allocation, has not become unba-

lanced or biased as a result of some unspecified

factor.

Another difficult aspect in the design of open-label

studies is how one assesses those patients who

withdraw from the study. The reasons for with-

drawal can be at least as varied as in double-blind

studies (intolerability, administrative difficulties,

coincidental emergent disease or concomitant

therapies, etc.). However, in addition, in an open-

label design, patients may develop an opinion on

the superiority of one or other treatment for reasons

that may or may not be explicit. If completion of a

course of therapy is one end point of the study, then

all withdrawals can be accounted treatment fail-

ures, and the statistical handling is fairly straight-

forward. However, if there is another end point, and

if withdrawals are imbalanced between the treat-

ment groups and unrelated to product intolerabil-

ity, then the situation becomes a lot more clouded.

Under these latter conditions, the entire trial may

have to be abandoned when it becomes apparent

that the trial design cannot answer the hypothesis

under test one way or the other.

On the positive side, open-label trials are usually

easy to administer, and patient recruitment and long-

evity within each treatment group can easily be mon-

itored as the study progresses. Investigators have

greater freedom in entering and allocating patients,

and this is often more comfortable than a placebo-

controlled situation in the ordinary clinical setting.

Equivalence trials

Sometimes, the demonstration of equivalency is suf-

ficient, especially when the competing product can-

not be expected to be inferior, or when a successor

10.1 OBJECTIVES OF THE PHASE IV CLINICAL DEVELOPMENT PROGRAM 121