Optimal Use of Antiemetics in the Outpatient Setting
Optimal Use of Antiemetics in the Outpatient Setting
It is said that "necessity is the mother of invention." By the same
token, advances in symptom management have become necessary due to the
toxicities associated with advances in antitumor therapy. This is particularly
true in the area of antiemetics. As chemotherapeutic agents of greater
therapeutic and emetogenic potential came into common use, it became necessary
to improve our understanding of the mechanisms of chemotherapy-induced emesis
and to develop more potent and more effective antiemetics.
In recent years, the administration of chemotherapy has rapidly changed from
a highly technical hospital-based procedure to a routine outpatient procedure.
However, every advance has a price. On the one hand, more effective supportive
care agents eliminated the necessity for intensive in-hospital care for the
patient receiving chemotherapy. On the other hand, this change in the venue of
chemotherapy administration now requires a capacity for effective administration
of these protective agents in the outpatient setting as well.
Fortunately in the case of antiemetics, this transition has been relatively
unremarkable. Improved understanding of the dose-response characteristics of
antiemetics and the time course of various types of emesis led to the
realization that antiemetic management is virtually identical in the outpatient
and inpatient settings. The one caveat is that health-care professionals need to
be more vigilant in the outpatient setting, because much of the potential emetic
response will occur at home, away from the medical facility.
As patient preferences and concerns regarding quality-of-life issues have
been taken into greater consideration, the importance of effective antiemetic
control has been further appreciated. In a survey of patient concerns published
in 1983, nausea and vomiting were found to be the two toxicities of
chemotherapy most feared by patients. One could object that this survey was
conducted well before the era of modern antiemetics and, therefore, this ranking
might be attributed to ineffective antiemetic control. However, when the survey
was repeated a decade later, nausea was still ranked the first and vomiting
the fifth most significant side effect of chemotherapy.
To devise effective strategies for the prevention of chemotherapy-induced
emesis, one must first understand the causes and mechanisms of emesis. It is
particularly important to remember that vomiting is a physiologic rather than a
pathologic process. Vomiting is the body’s natural defense against the
ingestion of toxic substances and is the most effective method to rid oneself of
such toxins. Vomiting only appears pathologic in the clinical setting because
many oncologic emetogenic agents are administered parenterally, while the body’s
defenses remain the same.
Emetic Reflex Arc
Since emesis is a basic defensive mechanism, it is not surprising to find
that this response is controlled by a basic reflex arc. The emetic reflex arc
is a multiafferent system with potential stimulation through neuroreceptors
located in the chemoreceptor trigger zone in the brain stem, the upper
gastrointestinal tract wall (peripheral pathway), the cerebral cortex (learned
or anticipatory vomiting), and the vestibular organs (motion sickness).
The emetic response is coordinated through the emetic center located in the
brain stem in the region of the nucleus tractus solitarius. Efferent neural
impulses are then transmitted to the multiple effector organs of the
gastrointestinal, respiratory, and musculoskeletal systems that must be
coordinated for an effective emetic response. This transmission of neural
impulses passes through numerous synapses between the initial stimulatory event
and the ultimate emetic response.
Identification and blockade of relevant neurotransmitters and
neurotransmitter receptors within the emetic reflex arc has proven to be the
cornerstone of antiemetic therapy. As the list of potentially relevant
neurotransmitters has expanded, the flexibility and effectiveness of
antiemetic therapy has expanded as well.
The first formal clinical trials of antiemetics were performed in the 1960s
and concentrated on dopamine (D2)-receptor antagonists. The phenothiazines
and the butyrophenones were shown to be effective antiemetics against
chemotherapy-induced emesis and became the mainstays of antiemetic therapy.
However, such therapies became inadequate as new chemotherapeutic agents with
greater emetogenic potential were developed. Against a highly emetogenic agent
such as cisplatin, which was introduced in the mid-1970s, traditional
standard-dose antiemetics were little better than placebo. This situation
changed with the publication of a single clinical trial.
One agent had been of particular interest to researchers in the antiemetic
field. Metoclopramide, which was used for the treatment of gastroparesis and for
the performance of radiologic procedures, was known to be a D2 antagonist and a
prokinetic agent, both properties that should theoretically be valuable in the
prevention of chemotherapy-induced emesis. However metoclopramide at standard
doses was as ineffective as other agents of the time against cisplatin-induced
In 1981, Gralla et al took advantage of the concept of dose-response,
commonly used as a basis for dose escalation in phase I studies of cytotoxic
agents, and applied this concept to the field of supportive care. When the dose
of metoclopramide was escalated 25-fold, the number of acute vomiting episodes
after administration of cisplatin decreased by over 90% (from a median of 11 to
a median of 1). Addition of a corticosteroid increased the antiemetic efficacy
The next major advance in antiemetic therapy resulted from the identification
of another important neurotransmitter receptor. Antidopaminergic toxicity (akathisia,
oculogyric crisis, anxiety, depression) had become a significant problem in
the administration of high-dose metoclopramide. However, unlike
standard-dose metoclopramide, which effectively blocked only the D2 receptor,
high-dose metoclopramide blocked both the D2 receptor and the serotonin (5-HT3)
receptor. Since the 5-HT3 receptor also appeared to have a role in the
induction of emesis, the separation of antiemetic efficacy from antidopaminergic
toxicity through the use of specific 5-HT3 antagonists became theoretically
Ondansetron (Zofran) was the first 5-HT3 antagonist studied and introduced
into clinical practice in the United States. In early dose-ranging
studies,[9,10] this agent was found to have activity equivalent to that of
high-dose metoclopranide over a wide range of doses, with an extremely mild
toxicity profile. Granisetron (Kytril) and dolasetron (Anzemet) were
later found to have similar properties. As with the earlier antiemetic families,
the addition of a corticosteroid markedly enhanced the antiemetic efficacy of
the 5-HT3 antagonists.
Cost-effective, efficient outpatient use of the 5-HT3 antagonists for
prevention of acute chemotherapy-induced emesis was originally hampered by
inconvenient dosing schedules and high cost. Further steps in development
therefore concentrated on questions of schedule, route, and dose.
Ondansetron, granisetron, and dolasetron all have serum half-lives ranging
from 4 to 10 hours. Ondansetron, with the shortest half-life, was originally
administered in three divided doses over 8 hours in the belief that such a
schedule would be necessary to sustain antiemetic protection.[9,10] Such a
schedule would have been impractical for use in a busy outpatient setting.
However the extended schedule also proved to be unnecessary.
In 1990, Marty and colleagues treated 305 patients receiving cisplatin
with ondansetron, 32 mg, given intravenously either as a single bolus or as an
8-mg loading dose followed by a 1-mg/h continuous infusion for 24 hours.
Prevention of nausea and vomiting was identical in both groups, suggesting that
treatment with a single dose during the initial hours after chemotherapy was
sufficient for 24-hour protection despite the short half-life.
Route of Administration
Although the intravenous route is the route most commonly used, attempts have
also been made to deliver antiemetics by the respiratory, transdermal, buccal,
rectal, nasal, and oral routes. Greatest attention has been focused on oral
administration because of its convenience, patient satisfaction with
self-administration, and cost savings due to the minimal pharmacy preparation
and nursing administration time involved. That said, the oral route would not
have gained general acceptance if a decrement in antiemetic efficacy had been
In two studies, each with more than 1,000 patients, Gralla et al (Table
1) and Perez et al (Table 2) compared oral granisetron to intravenous
ondansetron in patients receiving the highly emetogenic agent cisplatin or
moderately emetogenic chemotherapy, respectively. Patients could receive
corticosteroids at investigator discretion, and approximately two-thirds of the
patients did receive this additional therapy. Efficacy was identical between the
two regimens in both the overall groups and the large subsets receiving
corticosteroids as well.
Spector et al reversed the antiemetic agents, comparing intravenous
granisetron to oral ondansetron in the treatment of patients receiving cisplatin,
and achieved similar results. Thus, the oral route has equivalent efficacy to
the intravenous route for 5-HT3 antagonists that are administered at appropriate
The best dose of a drug is the lowest dose that is fully effective, as this
will be the most cost-effective strategy and the strategy most likely to avoid
unnecessary dose-related side effects. However, identification of the lowest
fully effective dose is not always straightforward. The fact that approved doses
of various 5-HT3 antagonists differ between countries, with the approved dose of
ondansetron being higher and the approved dose of granisetron being lower in the
United States than in Europe, suggests significant flexibility in dosing.
In 1994, Hesketh et al classified chemotherapeutic regimens as having
high, moderate-high, or moderate emetogenicity and treated patients receiving
these regimens with ondansetron at 32, 24, or 8 mg, respectively (all in
combination with dexamethasone, 20 mg). No loss of antiemetic effectiveness was
noted, providing documentation that dose de-escalation under certain
circumstances is feasible. Nevertheless, because the dose of ondansetron varied
with the emetogenicity of the regimen, the question of whether the standard dose
of ondansetron (32 mg for patients receiving cisplatin) was appropriate or
whether efficacy within a category might be compromised by dose de-escalation
was not addressed.
DiPiro et al examined this question in patients receiving moderately
emetogenic chemotherapy, randomizing patients to receive ondansetron at either
32 or 20 mg (both with dexamethasone, 10 mg). No loss of efficacy attributable
to ondansetron dose de-escalation was noted. Pectasides et al looked at the
same question in patients receiving highly emetogenic (cisplatin) chemotherapy.
Patients were randomized to receive ondansetron, 24 or 8 mg (both with
dexamethasone 20 mg), and equivalent antiemetic protection was still seen in
Of note, ondansetron dose de-escalation in the Pectasides study was
greater than that in the DiPiro study despite the greater emetogenicity of
the challenge agent. This variation in fully effective doses can be understood
through review of the dose-response curves of the 5-HT3 antagonists
(Figure 1). When data from dose-ranging studies of ondansetron, granisetron,
or dolasetron are plotted, a logarithmic rather than linear dose-response
curve is seen. All of these curves have a steep slope at lower doses
followed by a therapeutic plateau once a threshold dose value has been
surpassed. Although the numeric doses of the various agents may differ, the
level of complete antiemetic protection at the plateau doses is the same.
Thus, dose de-escalation is possible as long as the dose remains above the
threshold value for that agent. By the same token, dose escalation beyond a dose
that surpasses the threshold value will not result in a significant increase in
efficacy (although some increase in toxicities such as headache and constipation
may be seen).
The threshold values for the commonly used 5-HT3 antagonists, when used
against the "gold standard" highly emetogenic agent cisplatin, are
still not fully defined. Definition of the threshold value for 5-HT3 antagonists
used against moderately emetogenic chemotherapy (for which the threshold value
will be lower) or against bone marrow transplant regimens (for which the
threshold value may be higher) present future challenges.
Potency and Specificity
While it is valuable to recognize the marked similarities between the 5-HT3
antagonists, it is also valuable to recognize the potential differences. Potency
and specificity are two areas that have been considered in this regard. Although
potency may define full dose and reflect receptor-binding affinity, no
clinically significant difference in level of efficacy has been noted between
agents used at or above their threshold efficacy dose values. In contrast,
differences in binding specificity have certainly been described. In
addition to marked affinity for the 5-HT3 receptor, compounds related to
dolasetron may also have measurable binding affinity for the 5-HT2 receptor,
whereas ondansetron has measurable binding affinity for the 5-HT1B, 5-HT1C, and
opiate mu receptors (Table 3).
Theoretically, binding specificity (or lack of specificity) should not be as
important as the function of the secondary receptor. If the secondary receptor
adds additional antiemetic efficacy, then the lack of specificity is
advantageous. If the secondary receptor adds only additional toxicity, then the
lack of specificity is disadvantageous. Practically, secondary neuroreceptor
binding has not yet been shown to provide additional clinically significant
efficacy or toxicity for any of the currently available 5-HT3 antagonists.