findings ofno differencesare rarely reported, but
sometimes this finding may just be a function of
small sample size for each individual study or
the small degree of difference to be found. It
must also be recognized that many drugs were
introduced into medicine prior to the current
modern-day comprehensive testing programs.
Nonetheless, after many years and millions of
prescriptions, it is of reassurance that few have
shown significant clinically important gender-
related differences.
Differences in disease presentations
A report from theNational, Heart, Lung and Blood
Institute (NHLBI, 1996), showed that the age
and incidence (1988–1993) of onset of heart
disease between genders were different; 24% of
the 65–74 year-old males compared to about 18%
of females in the same age group. This incidence
rose in both genders at 75–84 years to about 28%
males and 30% females.
Not only do women develop heart disease later
but they also present differently. The signature
symptom of a heart attack, severe chest pain, is
often absent in women, and pain in the upper back
or neck, or breathlessness and nausea, may present
either as a single symptom or as multiple symp-
toms. The American Heart Association states that
44% of women are likely to die in the first year of
their heart attack, compared to 27% of men.
It is not surprising that heart attack and angina
are misdiagnosed more commonly in women than
men during emergency room visits. The range
between hospitals of misdiagnosis was 0–11%,
with an average 2.3% for angina and 2.1% for
heart attacks. The diagnosis was missed in 7% of
women under 55 years (Popeet al., 2000).
Finally a large NIH study (The Women’s Health
Study of 40 000 women), just completed in 2005,
showed that aspirin gave no cardiac protection to
women as had previously been assumed, though it
did reduce the incidence of ischemic strokes in
women over 65 years. This compared to the reduc-
tion of heart attacks in men. A subsequent meta-
analysis of six studies, including the Women
Health Study, confirmed this, and in addition,
showed no benefit in reducing ischemic strokes in
men (Bergeret al., 2006).What’s representative?
An additional dilemma is, what population
is ‘representative’ for female dose and efficacy
determination? Women of childbearing potential
(54%)? These will have possible hormonal cycling
changes and those on contraceptive hormones will
have even greater changes, added to a possible
basic gender difference, either amplifying or even
suppressing effects.
The needs of women aged 66 years or more are
already represented in regulatory drug testing
guidelines in the elderly, Federal Register (Federal
Register, 1990), but women 50–65 years old also
can lay claim to special consideration, given the
special problems associated with combined hor-
monal loss and age changes (e.g. osteoporosis, loss
of possible cardiac estrogen protection and
changes in body fat composition and its distribu-
tion). Pregnant women, already isolated from drug
development by fear of legal tort laws and, indeed,
by their physicians’ reluctance to even prescribe in
early pregnancy, can also stake a claim to require
additional studies. Finally, when studying females
of childbearing potential, should we include
patients on oral contraceptives (OCs), with their
large levels of regulated fluctuating but synthetic
hormones, or rely on females not taking OCs? The
latter option will increase the risk of potential fetal
exposure.
It must now be apparent that the female popula-
tion (51%) containsmanypotential subgroups,
none truly ‘representative’, for all have major phy-
siological differences from each other. For industry
to study all groups would be impractical, unecono-
mical and would gravely slow the drug develop-
ment process and compromise the number of
agents placed into development. To include all
groups within one all-encompassing study, unless
extremely large, offends a basic research nostrum –
that is ‘stabilize, reduce or remove all the variables
except the one to be measured’, or the signals many
be lost in the static. This is especially true in phase
II studies.16.2 THE DILEMMAS 205