|
Dexamethasone Alone or in Combination with
Ondansetron for the Prevention of Delayed Nausea and Vomiting Induced by
Chemotherapy
The Italian Group for Antiemetic Research
|
ABSTRACT
Background The
prevention of delayed nausea and vomiting caused by moderately
emetogenic chemotherapy for cancer has not been studied
systematically.
Methods We enrolled
patients who were scheduled to receive chemotherapy for the first
time in a double-blind, randomized, multicenter study. All the
patients received ondansetron combined with dexamethasone for
prophylaxis against emesis that might occur within 24 hours after
the start of chemotherapy (acute emesis). They were then divided
into two groups: patients who did not have either vomiting or
moderate-to-severe nausea (the low-risk group) and patients who had
one or both (the high-risk group). Patients in the low-risk group
were then randomly assigned to one of the following regimens, given
on days 2 through 5 after the start of chemotherapy: oral placebo, 4
mg of dexamethasone given orally twice daily, or 8 mg of ondansetron
in combination with 4 mg of dexamethasone, given orally twice daily.
Patients in the high-risk group were randomly assigned to receive
oral dexamethasone alone or in combination with ondansetron at the
same doses as those used in the low-risk group.
Results Among the
618 patients in the low-risk group, there was a complete absence of
both delayed vomiting and moderate-to-severe nausea in 91.8 percent
of those who received ondansetron combined with dexamethasone, 87.4
percent of those who received dexamethasone alone, and 76.8 percent
of those who received placebo. The proportions of patients who were
protected by dexamethasone combined with ondansetron or by
dexamethasone alone were significantly greater than the proportion
protected by placebo (P<0.001 and P<0.02, respectively). Of
the 87 patients in the high-risk group, complete protection was
achieved in 40.9 percent of those treated with ondansetron and
dexamethasone and in 23.3 percent treated with dexamethasone alone
(P not significant).
Conclusions The best
way to prevent delayed nausea and vomiting in patients receiving
moderately emetogenic chemotherapy is to control these complications
within the first 24 hours after the start of chemotherapy.
Dexamethasone alone provides adequate protection against delayed
emesis in patients at low risk (those who have not had acute
emesis).
The administration of moderately
emetogenic drugs such as cyclophosphamide, doxorubicin, epirubicin,
and carboplatin often causes nausea and vomiting within 24 hours
after the start of chemotherapy (acute emesis) or two to five days
later (delayed emesis). The combination of a 5-hydroxytryptamine3–receptor
antagonist, such as ondansetron, with dexamethasone can protect 90
percent of patients from acute emesis,1,2 but
these drugs given alone or in combination protect only 40 to 60
percent of patients from delayed emesis.3,4,5
Previous studies of therapy to prevent delayed emesis have not
accounted for the possibility that the occurrence of this
complication may be influenced by the control of emesis during the
first 24 hours.
In the present study, we investigated whether all patients who
receive moderately emetogenic chemotherapy need prophylaxis against
delayed emesis. We also compared the efficacy of several regimens
for the prevention of delayed emesis. Patients receiving chemotherapy
for cancer were given prophylaxis against emesis during the first 24
hours and then divided into two groups: those who did not have acute
emesis and were therefore at low risk for delayed emesis and those
who did have acute emesis and were therefore at high risk for
delayed emesis. Patients in these two groups were then randomly
assigned to one of three treatments for the prevention of delayed
emesis.
Methods
Patients
From September 1997 to June 1999, all adult patients scheduled
to receive moderately emetogenic chemotherapy for the first time
at 1 of 23 medical-oncology divisions (22 in Italy and 1 in
Yugoslavia) were asked to enter the study. They were to receive
cyclophosphamide (600 to 1000 mg per square meter of body-surface
area), doxorubicin (50 mg per square meter),
epirubicin (75 mg per square meter),
or carboplatin (300 mg per square meter),
either alone or in combination.
The criteria for exclusion before randomization were the presence
of nausea and vomiting or the use of antiemetic agents during the
24 hours before the administration of chemotherapy, a severe concurrent
illness other than neoplasia, other causes of vomiting (e.g.,
gastrointestinal obstruction, central nervous system metastases, or
hypercalcemia), concurrent treatment with glucocorticoids (unless
given as supplements) or benzodiazepines (unless given at night for
sedation) or abdominal radiotherapy, and a white-cell count of less
than 3000 per cubic millimeter or a platelet count of less than
70,000 per cubic millimeter. Also excluded from the study were
pregnant women, patients in whom the administration of dexamethasone
was contraindicated, patients scheduled to receive highly emetogenic
chemotherapy (e.g., dacarbazine, cisplatin, or mechlorethamine) on
the same day as moderately emetogenic chemotherapy, and patients
scheduled to receive any other cytotoxic agent from day 2 to day 5,
with the exception of fluorouracil, etoposide, teniposide,
vincristine, vinorelbine, vindesine, and vinblastine.
Design of the Study
Twenty-four hours after the start of moderately emetogenic
chemotherapy, patients were divided into two groups according to the
effectiveness of prophylaxis against emesis during the first 24
hours: a low-risk group, which included patients who had neither
vomiting nor moderate-to-severe nausea within the first 24 hours,
and a high-risk group, which included patients who had one or both
of these symptoms. The patients in the low-risk group were randomly
assigned to receive placebo, dexamethasone, or dexamethasone in
combination with ondansetron to prevent delayed nausea and vomiting,
and the patients in the high-risk group were randomly assigned to
receive dexamethasone or dexamethasone plus ondansetron for this
purpose.
Patients and personnel engaged in the study were blinded to the
assigned treatment for delayed emesis. The study was approved by the
ethics committee of each participating institution; all the patients
gave written informed consent. Patients were evaluated according to
the intention-to-treat principle.
Antiemetic Therapy
For the prevention of nausea and vomiting during the first 24
hours, all the enrolled patients received a combination of 8 mg
of dexamethasone, diluted in 100 ml of saline and administered intravenously
over a 15-minute period 30 minutes before the beginning of
chemotherapy, and 8 mg of ondansetron, diluted in 100 ml of saline
and administered intravenously over a 15-minute period 15 minutes
before chemotherapy. In addition, immediately before the start of
chemotherapy and then at six-hour intervals, 4 mg of dexamethasone
was administered orally, for a total of four doses.
Twenty-four hours after the start of chemotherapy, the patients
were randomly assigned to take one of the following regimens of
oral antiemetic drugs on days 2 through 5 after chemotherapy. Patients
in the low-risk group were to take placebo, 4 mg of dexamethasone
twice daily, or the combination of 8 mg of ondansetron and 4 mg of
dexamethasone twice daily. Patients in the high-risk group were to
take 4 mg of dexamethasone twice daily or the combination of 8 mg of
ondansetron and 4 mg of dexamethasone twice daily. A block-balanced
randomization list (with 12 patients per block) was used. To ensure
that the oral treatment could not be identified, the drugs were put
into identical capsules, and each patient received two pill
containers. The day after chemotherapy was given, the investigators
called each patient to remind him or her of the correct pill
containers to be used, according to whether patients were in the
high-risk or the low-risk group. Compliance was checked by counting
the remaining pills at the end of treatment. Eating was not
permitted until six hours after the administration of chemotherapy.
Clinical Assessment
Episodes of nausea and retching or vomiting were recorded by the
patients on diary cards for the first 24 hours after chemotherapy (acute
emesis) and for the following eight days (delayed emesis). Nausea
and vomiting were defined according to previously published criteria.1
The primary end point of this study was complete protection from
both vomiting and moderate-to-severe nausea on days 2 through 5
(complete protection from delayed emesis). The secondary end points
were complete protection from delayed vomiting, complete protection
from delayed moderate-to-severe nausea, the time to the first
episode of acute vomiting, and the mean number of episodes of
vomiting on days 2 through 5 among the patients who vomited. Adverse
events other than episodes of vomiting or nausea were also recorded
on the diary cards by the patients.
Statistical Analysis
In a previous study of patients given moderately emetogenic chemotherapy
who did not receive any prophylaxis against delayed emesis, we found
that the incidence of delayed vomiting or moderate-to-severe nausea
was about 90 percent among patients who had nausea, vomiting, or
both during the first 24 hours after the start of chemotherapy and
42 percent among those who did not.6 The
number of patients to be enrolled in the current trial was
calculated as follows. It was assumed that among patients with
either vomiting or moderate-to-severe nausea within 24 hours after
the start of chemotherapy, prophylaxis with dexamethasone would
reduce the incidence of delayed vomiting or nausea from 90 percent
to 75 percent when used alone and to 50 percent when combined with
ondansetron. To detect a significant difference between these two
antiemetic treatments with a probability of 80 percent (with a P
value of 0.05 by one-tailed testing
defined as indicating statistical significance), 106 patients (53
in each treatment group) would be required. If only 15 percent of
enrolled patients had vomiting or moderate-to-severe nausea within
the first 24 hours, it would be necessary to enroll at least 706
patients in order to include 106 patients with these symptoms.
Therefore, 600 patients who did not have vomiting or
moderate-to-severe nausea in the first 24 hours (200 in each treatment
group and 200 in a placebo group) would be needed to perform a
placebo-controlled evaluation of these treatments for the prevention
of delayed emesis. Assuming that the incidence of delayed vomiting
or nausea among patients taking placebo is 40 percent, and
hypothesizing that this incidence would be reduced to 25 percent
with the use of ondansetron combined with dexamethasone (and to 32.5
percent with dexamethasone alone), we estimated the power of the
current study to detect a difference between treatment groups to be
nearly 90 percent.
Fisher's exact test (two-tailed) was used for the analysis of
outcomes in the high-risk group. For the analysis of outcomes in
the low-risk group, Fisher's exact test was generalized with use of
the Freeman–Halton test7 to
evaluate the balance of prognostic factors and to compare the
proportions with each adverse event or with complete protection from
delayed vomiting and moderate-to-severe nausea among the three
treatment subgroups. When the difference in proportions was
significant, three pairwise comparisons of subgroups were performed,
with adjustment of the significance level according to Bonferroni's
inequality. Logistic models were used to evaluate the efficacy of
the antiemetic treatments in providing complete protection from
delayed vomiting and moderate-to-severe nausea, both in a
unifactorial analysis and in multifactorial analyses in which all
the prognostic factors available on case-record forms were
considered; therefore, an overall G test and the Wald test for each
factor were used.8
The 95 percent confidence intervals for the differences between treatment
subgroups in the proportions of patients with protection from
delayed vomiting and moderate-to-severe nausea were also calculated.
Among the patients who vomited, we compared the mean times to
the first episode of vomiting as well as the mean numbers of days
with vomiting, moderate-to-severe nausea, or both with the use of
nonparametric tests (the Kruskal–Wallis test and Wilcoxon's rank-sum
test).
Results
A total of 708 patients entered the study, of whom 1 died and
2 were lost to follow-up. The remaining 705 patients were evaluated according
to the intention-to-treat principle. Among these 705 patients were 5
who refused prophylaxis against delayed emesis because of adverse
events during the first 24 hours after the start of chemotherapy and
14 who took the wrong study capsules. The low-risk group consisted
of 618 patients who were protected against both vomiting and
moderate-to-severe nausea during the first 24 hours. They were
randomly assigned to receive placebo (203 patients), dexamethasone
(207 patients), or the combination of ondansetron and dexamethasone
(208 patients). The high-risk group consisted of 87 patients, all of
whom had vomiting, moderate-to-severe nausea, or both within 24
hours after the start of chemotherapy. They were randomly assigned
to receive dexamethasone (43 patients) or the combination of
ondansetron and dexamethasone (44 patients). The characteristics of
the patients were similar in all subgroups within the low-risk and
high-risk groups (Table
1).
|
During the first 24 hours after the start of moderately emetogenic chemotherapy,
the standard antiemetic combination of ondansetron and dexamethasone
completely protected 634 of the 705 patients from vomiting (89.9
percent) and 499 of the 705 patients from any nausea (70.8 percent).
These results are similar to those obtained in our previous study.1 The
median time to the first episode of vomiting was 10 hours.
Delayed Nausea and Vomiting
All of the following results were obtained by the evaluation of
patients according to the intention-to-treat principle.
In the low-risk group, the efficacy of the combination of
ondansetron and dexamethasone was significantly superior to that of
placebo according to three measures of complete protection: an
absence of both vomiting and moderate-to-severe nausea during days 2
through 5 (protection in 91.8 percent of patients in the
dexamethasone-plus-ondansetron subgroup vs. 76.8 percent in the
placebo subgroup; P<0.001 for the pairwise comparison with
adjustment for Bonferroni's inequality); an absence of vomiting
alone (95.2 percent vs. 87.2 percent, P<0.02); and an absence of
moderate-to-severe nausea alone (93.3 percent vs. 81.8 percent,
P<0.002) (Table
2). By contrast, dexamethasone alone was significantly better
than placebo only in providing protection from both vomiting and
moderate-to-severe nausea (protection in 87.4 percent of the
patients in the dexamethasone subgroup vs. 76.8 percent in the
placebo subgroup; P<0.02). There were no statistically significant
differences between ondansetron and dexamethasone in combination and
dexamethasone alone in the rate of protection from vomiting,
moderate-to-severe nausea, or both vomiting and moderate-to-severe
nausea.
|
Among the patients in the low-risk group who had delayed vomiting or
moderate-to-severe nausea, the mean number of days on which these
complications occurred did not differ significantly among the three
subgroups (2.2 days in the placebo subgroup, 1.9 in the
dexamethasone subgroup, and 2.2 in the dexamethasone-plus-ondansetron subgroup).
After the prophylaxis against delayed emesis was stopped, only nine
patients in the low-risk group began to vomit again or had
moderate-to-severe nausea (eight patients for one day and one
patient for two days).
The antiemetic treatment provided much less protection against
delayed emesis in the high-risk group than it did in the low-risk group.
Among the patients in the high-risk group, all of whom vomited or
had moderate-to-severe nausea during the first 24 hours after
chemotherapy, the combination of ondansetron and dexamethasone was
not significantly more effective than dexamethasone alone in
preventing delayed vomiting and moderate-to-severe nausea (Table 3). The delayed
complications were prevented in 18 of the 44 patients taking the
combination of the two drugs (40.9 percent) and in 10 of the 43
patients taking dexamethasone alone (23.3 percent). Among the
patients in the high-risk group who had vomiting or moderate-to-severe
nausea, the mean number of days with these complications in the
dexamethasone subgroup (2.3 days) was not significantly different
from the number of days in the subgroup taking ondansetron plus
dexamethasone (2.4).
|
Multifactorial Analyses
Because of the small number of patients in the high-risk group,
we found no significant difference between the treatment subgroups with
regard to outcome. In the low-risk group, in addition to treatment,
only whether or not the full dose of chemotherapy was received had a
significant effect. The incidence of delayed emesis was greater
among patients who received a full course of chemotherapy than among
those who did not (P<0.02) (data not shown). In addition,
differences among the chemotherapy regimens had no significant
influence on the rate of complete protection from delayed emesis.
Adverse Events
No severe or unexpected adverse events were reported. On days
2 through 5 after the start of chemotherapy, adverse events occurred
in 331 (53.6 percent) of the 618 patients who had not had vomiting
or moderate-to-severe nausea during the first 24 hours and in 55
(63.2 percent) of the 87 patients who did (P not significant). The
most important adverse events are listed in Table 4. Among
the patients in the low-risk group, constipation was significantly
more frequent among those taking ondansetron and dexamethasone
combined (25.0 percent) than among those taking placebo (8.9
percent, P<0.001) or dexamethasone alone (8.7 percent,
P<0.001). In the high-risk group, there were no significant
differences in the rates of adverse events between the two treatment
subgroups.
|
Discussion
This randomized study shows that the efficacy of prophylaxis against
delayed vomiting or moderate-to-severe nausea due to moderately
emetogenic chemotherapy for cancer is strongly influenced by the
occurrence of these complications during the first 24 hours after
the start of chemotherapy. The combination of ondansetron and
dexamethasone, which controlled delayed vomiting and nausea in more
than 90 percent of patients who did not have these symptoms initially,
protected only slightly more than 40 percent of patients who did
have vomiting and nausea during the first 24 hours. This finding
indicates that an effective way to prevent delayed emesis in
patients receiving moderately emetogenic chemotherapy is to control
emesis during the first day. The drug combination that we used in
this study, a 5-hydroxytryptamine3–receptor antagonist
and dexamethasone, is probably the best choice for preventing acute
emesis.1
Two limitations of our study may hinder the evaluation of
antiemetic prophylaxis in the group of patients at high risk for
delayed emesis. First, in the high-risk group, we did not include a
placebo subgroup. We did this for ethical reasons, because delayed nausea
and vomiting would have developed in a high proportion of these
patients if they had not been given prophylaxis.6 For
this reason, we cannot evaluate precisely the benefit conferred by
antiemetic prophylaxis in these patients. Second, for unknown reasons,
the full course of chemotherapy called for in the protocol was
administered to only 386 of the 705 patients (54.8 percent); 37 of
these 386 patients (9.6 percent) received less than 90 percent of
the dose prescribed in the protocol. As a result of this shortfall
and the lower doses for some patients, there were fewer cases of
emesis than expected. Therefore, the number of patients in the
high-risk group was too small to allow the study to reach a
statistical power of 80 percent for detecting a significant
difference between outcomes with the two antiemetic treatments
(dexamethasone alone or dexamethasone in combination with
ondansetron). Given these limitations, the best choice for
preventing delayed nausea and vomiting in patients at high risk for
these complications remains to be identified.
Among the patients in the low-risk group, both dexamethasone alone
and dexamethasone in combination with ondansetron were significantly
better than placebo in providing protection against delayed emesis,
and no significant difference in efficacy was found between the two
regimens. It is noteworthy that high percentages of the patients at
low risk who were given placebo did not have delayed vomiting (87.2
percent) or moderate-to-severe nausea (81.8 percent). These
percentages are similar to those we found in a previous study of
patients who received neither prophylaxis nor placebo against
delayed emesis, in which the same regimen for prophylaxis against
emesis during the first 24 hours was used. In our opinion, the high
rate of protection from delayed emesis among patients who received
placebo should not discourage physicians from using antiemetic
prophylaxis, especially because an episode of delayed emesis
predicts the occurrence of both acute and delayed emesis during
subsequent cycles of chemotherapy.6
In patients at low risk for delayed nausea and vomiting, dexamethasone
alone seems preferable to the combination of dexamethasone and ondansetron
because its efficacy is similar to that of the combination, it is
better tolerated (specifically, it causes less constipation), and it
is less costly.
In conclusion, we believe that prophylaxis against delayed emesis
is warranted both in patients at low risk for this complication and
in those at high risk, although fewer than half the patients at high
risk will be protected and a remarkable number of patients at low
risk would be protected without active treatment. Future studies of
treatments to prevent delayed emesis must be designed to take into
account the influence of complete protection from emesis during the
first 24 hours on the control of delayed emesis.9
We are indebted to the Department of Clinical and Biological Sciences,
University of Turin, for enrolling their patients in this study; to
Glaxo Wellcome, Verona, Italy, for furnishing the antiemetic drugs;
and to G. Grandolini, Ph.D., and M. Marani, Ph.D., of the Department
of Chemistry and Pharmaceutical Technology, University of Perugia,
for preparing the placebo for the study.
* Dr. Roila, one of the principal
investigators, assumes responsibility for the overall content of the
article. The participants in the study group are listed in the
Appendix.
Source Information
Address reprint requests to Dr. Fausto Roila at the Medical
Oncology Division, Policlinico Hospital, 06122 Perugia, Italy, or at [log in to unmask].
References
Appendix
The Italian Group for Antiemetic Research includes the following
(all in Italy unless otherwise specified): Principal investigators: Medical Oncology
Division, Policlinico Hospital, Perugia — F. Roila; Medical
Statistics Unit, Department of Internal Medicine and Public Health,
University of L'Aquila, L'Aquila — E. Ballatori and B. Ruggeri;
Medical Oncology Division, Policlinico Hospital, Perugia — V. De
Angelis and M. Tonato; Institute for Oncology and Radiology,
Belgrade, Yugoslavia — S. Bosnjak; and Department of Internal
Medicine and Oncologic Sciences, University of Perugia, Perugia — A.
Del Favero. Monitor: Medical
Oncology Division, Perugia — G. Ciccarese. Collaborating
centers and investigators: Medical Oncology Division, Perugia
— C. Basurto, G. Ciccarese, M.A. Palladino, and S. Porrozzi; Institute
for Oncology and Radiology, Belgrade, Yugoslavia — Z. Marinkovic, Z.
Neskovic-Konstantinovic, S. Susnjar, S. Vasovic, S. Colakovic, V.
Lukic, and D. Radosavljevic; Medical Oncology Service, Legnano
Hospital, Legnano — S. Fava, E. Grimi, A. Calcagno, A. De Paoli, M.
Luoni, and A. Tocci; Medical Oncology Division, Hospital Renzetti,
Lanciano — A. Nuzzo, L. Laudadio, A. Di Blasio, and M. Sacco;
Medical Oncology Service, Sassari — A. Contu, N. Olmeo, A. Pazzola,
and G. Baldino; Medical Oncology Division, Negrar Hospital, Verona —
V. Picece, M. Nicodemo, M. Cirillo, and E. Recaldin; Medical
Oncology Division, Ravenna — C. Dazzi, A. Cariello, P. Giovanis, and
F. Zumaglini; Medical Oncology Division, Potenza — G. Rosati, L. Manzione,
D. Bilancia, and A. Rossi; Medical Oncology Division, Arcispedale S.
Anna, Ferrara — D. Donati, R. Maccaferri, and P. Malacarne; Medical
Oncology Division, Bergamo — R. Labianca, A. Quadri, and M.A. Pessi;
Medical Oncology Division, La Sapienza University, Rome — E. Cortesi
and O. Martelli; Medical Oncology Service, Fabriano — L. Giuliodori,
R.R. Silva, and D. Mari; Medical Oncology Department, University of
Cagliari, Cagliari — B. Massidda and M.T. Ionta; Medical Oncology
Service, Pesaro — P. Alessandroni and A. Baldelli; Medical Oncology
Service, Hospital S. Eugenio, Rome — M. Antimi and M. Minelli; Medical
Oncology B Division, National Cancer Institute, Naples — C. Gridelli
and A. Rossi; Medical Oncology Service, Parma — R. Passalacqua and
M. Quarta; Medical Oncology Service, Foligno — M. Sassi and D.
Pinaglia; Medical Oncology Department, Internal Medicine Division,
V. Fazzi Hospital, Lecce — E. De Marino; Medical Oncology Service,
Anagni — M.A. Giampaolo and F. Ciancola; Medical Oncology Service,
Giulianova — A. Lalli and S. Di Felice; and Medical Oncology
Service, Erba — C. Casartelli.
Edward E.
Rylander, M.D.
Diplomat American
Board of Family Practice.
Diplomat American
Board of Palliative Medicine.