Anti-emetic efficacy and toxicity of nabilone, a synthetic cannabinoid, in lung cancer chemotherapy.
Abstract
Nabilone, a synthetic cannabinoid, and Prochlorperazine were compared in a double-blind crossover study of 34 patients with lung cancer undergoing a 3-day schedule of chemotherapy with Cyclophosphamide, Adriamycin and Etoposide. Symptom scores were significantly better for patients on nabilone for nausea, retching and vomiting (P less than 0.05). Fewer subjects vomited with nabilone (P = 0.05) and the number of vomiting episodes was lower (P less than 0.05); no patients on nabilone required additional parenteral anti-emetic. More patients preferred nabilone for anti-emetic control (P less than 0.005). Adverse effects common with nabilone were drowsiness (57%), postural dizziness (35%) and lightheadedness (18%). Euphoria was seen in 14% and a "high" in 7%. Erect systolic blood pressure was lower in nabilone patients on Day 1 (P = 0.05) but postural hypotension was a major problem in only 7%. Nabilone is an effective oral anti-emetic drug for moderately toxic chemotherapy, but the range and unpredictability of its side-effects warrant caution in its use.
Br. J. Cancer (1983), 48, 657-663
Anti-emetic efficacy and toxicity of nabilone, a synthetic
cannabinoid, in lung cancer chemotherapy.
S. Ahmedzai, D.L. Carlyle, I.T. Calder & F. Moran
Centre for Respiratory Investigation, and Department of Pharmacy, Glasgow Royal Infirmary,
Alexandra Parade, Glasgow G31.
Summary Nabilone, a synthetic cannabinoid, and Prochlorperazine were compared in a double-blind
crossover study of 34 patients with lung cancer undergoing a 3-day schedule of chemotherapy with
Cyclophosphamide, Adriamycin and Etoposide. Symptom scores were significantly better for patients on
nabilone for nausea, retching and vomiting (P<0.05). Fewer subjects vomited with nabilone (P=0.05) and
the number of vomiting episodes was lower (P<0.05); no patients on nabilone required additional parenteral
anti-emetic. More patients preferred nabilone for anti-emetic control (P<0.005). Adverse effects common
with nabilone were drowsiness (57%), postural dizziness (35%) and lightheadedness (18%). Euphoria was
seen in 14% and a "high" in 7%. Erect systolic blood pressure was lower in nabilone patients on Day 1
(P=0.05) but postural hypotension was a major problem in only 7%.
Nabilone is an effective oral anti-emetic drug for moderately toxic chemotherapy, but the range and
unpredictability of its side-effects warrant caution in its use.
Gastrointestinal toxicity is frequently a limiting
factor in the acceptability of cancer chemotherapy
to both patients and physicians (Anonymous,
1979). Nausea and vomiting may be so severe as to
adversely affect quality of life and reduce
compliance with treatment (Laszlo & Lucas, 1981).
Anecdotal reports that smoking marijuana could
reduce nausea caused by cytotoxics prompted
studies in the United States which showed that A9-
THC, the major psychoactive constituent of
cannabis, possessed significant anti-emetic
properties (Sallan et al., 1975; Laszlo, 1979).
However, the incidence of psychotropic and to a
lesser extent cardiovascular side-effects, and its
poor oral efficacy, have limited the use of A9-THC
(Frytak et al., 1979).
Nabilone, a synthetic derivative of cannabis, is
structurally different from natural cannabinoids,
possessing a dimethyl-heptyl side-chain which
prevents its chemical conversion into A9-THC
(Lemberger & Rowe, 1975). Early studies of
nabilone demonstrated useful anti-emetic activity
but with a high incidence of cannabinoid side-
effects; however these were conducted in mixed
groups of patients with various malignancies and
cytotoxic regimes and who may also have had prior
experience of both emesis-inducing chemotherapy
and marijuana (Herman et al., 1979; Steele et al.,
1980). The present study was designed to evaluate
the efficacy and toxicity of nabilone compared with
oral prochlorperazine in a histologically
homogeneous group of new patients with lung
cancer receiving identical chemotherapy.
Methods
Subjects
Thirty-four consecutive patients, including 15
females, with small cell bronchial carcinoma who
were eligible for chemotherapy were studied. The
median age was 58 years and all but one had an
admission ECOG performance status of 3 or less.
There were no patients with active psychiatric
disease. Patients were informed of the nature and
potential adverse-effects of the chemotherapy and
that two different anti-emetic agents were under
assessment. The relation of nabilone to cannabis
was not disclosed, and it was thought that none of
the patients had prior experience of marijuana.
Details of the patients are given in Table I.
Chemotherapy
All patients received two 21-day cycles of
combination chemotherapy comprising
Cyclophosphamide (CTX) 1 gm-2, Adriamycin
(ADR) 40mgm-2 and Etoposide (VP-16)
JOOmgm-2 on Day 1; VP-16 lOOmgm-2 on Days
2 and 3; and Vincristine 2mg with Methotrexate
50mgm-2 on Day 10, followed by folinic acid
rescue. The Day 1-3 chemotherapy pulses were
given on an in-patient basis, with CTX and ADR
administered as i.v. boluses and VP-16 as an i.v.
infusion over 1-2 h.
? The Macmillan Press Ltd., 1983
Correspondence: S. Ahmedzai.
Received 25 May 1983; accepted 1 August 1983.
658 S. AHMEDZAI et al.
Table I Characteristics of patients entering the study
Total number entered:
Sex distribution:
Age:
Stage:
Performance Status (ECOG):
Anti-emetics
The anti-emetics under study were restricted to the
Day 1-3 pulses as the Day 10 drugs were not
thought to be unduly toxic. For each cycle, patients
were admitted to hospital on the day preceding
chemotherapy (Day 0), and started on the anti-
emetic regime at 10.00 pm. The chemotherapy drugs
were given at 10.00 am or shortly thereafter on each
of the following 3 days with anti-emetics continued
throughout, the patients were usually discharged
late on Day 3 or on Day 4.
A double-blind, double-dummy design was used,
with patients receiving by random allocation either
nabilone or prochlorperazine on the first cycle, and
crossing over on the second course. Nabilone
dosage was 2 x 1 mg capsules at 10.00 am and
10.00pm; prochlorperazine dosage was *2 x 5mg
tablets at 6.00am, 2.00pm and 10.00pm. On the
third or subsequent cycles of chemotherapy patients
were treated with the anti-emetic agent of their
choice. If a patient experienced severe nausea or
vomiting in spite of the study drug, parenteral
treatment with metoclopramide 10mg or
chlorpromazine 50mg was given as required and
the number of doses recorded.
Assessments
Symptoms With each anti-emetic course the
subjects completed a self-rating symptom
questionnaire, covering anorexia, nausea, retching
and vomiting for the week before chemotherapy
and for each of the 3 treatment days. Each
symptom was graded into 4 categories, scored on a
0-3 scale with 3 representing the worst category. In
addition patients were asked to report any side-
effects and if these were not spontaneously offered,
specific questions were asked about them on
completion of the questionnaire. At the end of the
second cycle patients were asked their preference
for the first or second regime, taking into account
both anti-emetic and any side-effects.
Physical Blood pressure in the erect and supine
positions and pulse rate were recorded just before
the first dose of anti-emetic at 10.00pm on Day 0,
34
Male 19: Female 15
Median 58 years (range 27-72)
Limited 18: Extensive 16
Median 2
("0"=2, "I"= 10, "2"= 14, "3"=7, "4"= 1)
1 h afterwards and therafter twice daily. The
number of vomiting episodes was recorded but the
volume of vomitus was not routinely measured.
Statistical methods
Symptom scores were analysed by the Mann-
Whitney U test. The binomial test was used to
analyse preferences between the two active drugs,
and the number of vomiting episodes on either.
Blood pressure -data was analysed by the
independent t-test. The effect of order of drug
administration on preference was studied using
Fisher's exact test. All probability values given are
for two-tailed tests; significance accepted when
P ( 0.05.
Results
Of the 34 patients entered, 6 dropped out after the
first course (5 died during the first cycle of
chemotherapy, one was withdrawn from
chemotherapy after review of histology-see Table
II), and 2 patients did not complete a course
because of adverse effects, leaving 26 patients who
completed the crossover. All but 4 of these entered
the study on their first cycle of chemotherapy; the 4
had received one prior cycle with standard
phenothiazine anti-emetics and had all experienced
mild to moderate gastro-intestinal toxicity and so
were considered suitable for inclusion in the
analysis.
Symptoms
Twenty-six patients completed questionnaires for
one or both parts of the crossover. Symptom scores
prior to chemotherapy (Day 0) were similar for
nabilone and prochlorperazine subjects, except for
a higher proportion of the latter who reported mild
retching. Table III summarizes the scores for each
symptom during chemotherapy, giving the daily
mean scores, and the proportion of patients with
the maximum score, on either anti-emetic. Mean
symptom scores were always higher and the
proportion of patients with the worst scores was
NABILONE IN LUNG CANCER THERAPY 659
Table II Details of patients who failed to complete the crossover
Anti-emetic
Patient No. Age Sex Stage ECOG received Reason
7 43 M E 3 Prochlorperazine Died after 1st cycle of
chemo, on Day 41
16 57 F E 2 Nabilone Died during 1st cycle
of chemo, on Day 12
26 41 M L 3 Nabilone Died during 1st pulse
of chemo, on Day 12*
28 62 M L 2 Prochlorperazine Histology reviewed--
large cell anaplastic;
withdrawn and died on
Day 22
31 65 M E 3 Prochlorperazine Died during 1st pulse
of chemo, on Day 14*
35 49 M E 3 Nabilone Died during 1st pulse
of chemo, on Day 13*
*Neutropenia (WCC < 1000) at time of death; patient No 31 was also septicaemic.
Table III Symptom scores during chemotherapy with
relation to nablione (N) and prochlorperazine (P) courses.
% Patients with
(Mean scores)* maximum (worst) scores
Symptom Day (N P) (N P)
1 0.8 1.2 4 8 NS
Anorexia 2 0.8 1.3 0 8 NS
3 0.7 1.1 0 4 P<0.05*
1 0.3 1.0 0 16 P<0.005
Nausea 2 0.4 1.1 0 12 P<0.01
3 0.1 0.6 0 4 P<0.05
1 0.1 0.9 0 16 P=0.001
Retching 2 0.2 0.9 0 4 P<0.01
3 0.1 0.5 0 0 NS
1 0.3 0.7 4 16 NS
Vomiting 2 0.3 0.9 0 0 P<0.05
3 0 0.6 0 0 P<0.001
*All significance values are two-tailed (Mann-Whitney
U test).
NS = not significant.
660 S. AHMEDZAI et al.
also higher for each day in prochlorperazine-treated
patients. Nabilone was superior to prochlorperazine
on Day 1 for nausea (P =0.005) and retching
(P=0.001); on Day 2 for nausea (P<0.01),
retching (P<0.01) and vomiting (P<0.05); and on
Day 3 for anorexia (P<0.05), nausea (P<0.05) and
vomiting (P<0.001). On Day 1, 71% of nabilone
subjects experienced no nausea, compared to 36%
on prochlorperazine, and on Day 2 the relative
proportions with no retching, and no vomiting were
96% and 56%, and 79% and 68% respectively. By
Day 3, the relative proportions for nabilone and
prochlorperazine subjects with no nausea were 79%
and 40%, with no retching 83% and 48%, and
with no vomiting 100% and 60% respectively.
Vomiting episodes
Although the worst symptoms were reported on
Day 1, the greatest number of patients vomited on
Day 2-see Table IV. There was no statistical
difference between nabilone and prochlorperazine
patients' vomiting on Day 1, but a significantly
Table IV Proportion of patients vomiting in
relation to anti-emetic
Number of patients vomiting (%)
Day Nabilone Prochlorperazine
1 6/27 (22) 9/30 (30) NS
2 4/26 (15) 13/30 (43) P=0.05*
3 0/26 (0) 8/30 (27) P<0.01*
*Binomial test (two-tailed)
NS = not significant
higher proportion of the latter vomited on Day 2
(P=0.05) and on Day 3 (P<0.01). The total
number of vomiting episodes was higher on each
day for the prochlorperazine group: on Day 1 there
were 1.3 episodes per patient vomiting on nabilone
and 3.0 episodes per patient vomiting on
prochlorperazine (difference not significant); on
Day 2 the respective vomiting rates were 2.0 and
2.6 episodes per patient vomiting (P<0.05: Mann-
Whitney U test); on Day 3 when no patients on
nabilone vomited, there were 2.0 episodes per
patient vomiting on prochlorperazine (P<0.005).
Additional parenteral anti-emetics
No patient on nabilone required extra parenteral
anti-emetic. Four out of 30 patients receiving
prochlorperazine had symptoms severe enough to
require i.v. or i.m. metoclopramide 10mg or
Chlorpromazine 50mg. Of these, all received one
dose each on Day 1, and a mean of 2.5 doses on
Days 2 and 3.
Preference
Sixteen patients preferred the anti-emetic control of
nabilone, compared to 3 preferring
prochlorperazine (P <0.005). However, taking
adverse effects into account, only 12 patients
wished to receive nabilone for subsequent
chemotherapy courses, and 7 requested
prochlorperazine (difference not significant). Seven
of the remainder expressed no preference and 8
were not evaluable by failing to complete the
crossover. There was no significant relation between
order of administration of active drugs and anti-
emetic preference (Table V), or overall preference
taking into account adverse effects.
Table V Anti-emetic
preferences of patients, with reference
administration of drug.
to the order of
Order of drug Preference
administration Nabilone* Prochlorperazine Neither Not Evaluable
First 8 1 4 4
Secondt 8 2 3 4
16 3 7 8 T=34
*Preference for nabilone to prochlorperazine is significant:
P<O.005, binomial test (two-tailed)
tThe 10 vs. 9 preference for the second drug is not significant (binomial test). There
is no significant relation between preference and order of administration of drug
(Fisher's exact test).
NABILONE IN LUNG CANCER THERAPY 661
The median ages of patients preferring nabilone
and prochlorperazine overall were 58 years and 61.5
years, and the median ECOG status for these
groups was 2 and 1 respectively. There was no
statistical association between sex or stage of
disease and preference.
Side-effects
Side-effects were commoner with nabilone than
prochlorperazine-see Table VI. The commonest
were drowsiness (57% on nabilone and 27% on
prochlorperazine), postural dizziness (35% and 4%
respectively) and light-headedness (18% of nabilone
subjects only). The severity of these symptoms
usually fell progressively over the 3 day period. A
"drunk" feeling was reported by 5 patients (18%)
on nabilone, of whom 2 found it pleasant and 3
unpleasant. Moderate euphoria occurred in 4
patients (14%) on nabilone and a "high" was
observed in 2 (7%); no patients on
prochlorperazine reported these sensations. Mild
confusion and disorientation were seen in 3
nabilone subjects (11%) and more upsetting
dysphoria in 2 (7%): however no patient had
hallucinations.
The side-effects which were severe enough to
cause a patient to be withdrawn from a course of
nabilone, or to prefer prochlorperazine for
subsequent chemotherapy in spite of better anti-
emetic control with nabilone, were extreme
drowsiness (2 patients), severe postural dizziness
(2), light-headedness (4) and unpleasant "drunk"
feeling (3). The median age of patients who
reported a "drunk" feeling on nabilone was 42
years; for those who experienced euphoria or a
"high" it was 49.5 years. The median ages of
patients who reported other side-effects were close
to the whole group's. There was no significant
association between anti-emetic preference and the
patients' previous alcohol consumption.
There were no major changes in mean blood
pressures or pulse rates in patients on either drug.
Nabilone patients had a slightly lower initial BP
than patients starting on prochlorperazine (mean
supine BP + sd = 130/79 + 22/14 mm Hg and
136/83 + 24/18mm Hg respectively); the only
statistically significant difference in the BPs
occurred on Day 1 when the mean erect systolic
pressure on nabilone was 122+17 mm Hg and on
prochlorperazine 133 +19mm Hg (P<0.05).
Diastolic pressures did not vary significantly.
The lowest blood pressures were recorded in
Patient No 16, a 57 year old female with extensive
disease, ECOG status 2, and previous mitral valve
replacement. She was withdrawn from the course of
Table VI Side-effects of drugs
Nabilone Prochlorperazine
Side-effects 28 Subjects 26 Subjects
Drowsiness-mild 12 (43)* 6 (23)
-severe 4 (14) 1 (4)
Postural dizziness-mild 8 (28) 1 (4)
-severe 2 (7) 0
Lightheadedness-mild 1 (4) 0
-severe 4(14) 0
Confusion/disorientation 3 (11) 0
Dysphoria 2 (7) 0
Drunk-feeling-pleasant 2 (7) 0
-unpleasant 3 (11) 0
Euphoria 4 (14) 0
"High" 7 (7) 0
Dry Mouth 3 (11) 1 (4)
Blurred vision 1 (4) 0
Paraesthesia/numbness 2 (7) 2 (8)
Vertigo 1 (4) 0
Nausea 1 (4) 0
Headache 0 1 (4)
Itch 0 1 (4)
*Figures in parentheses are percentages.
662 S. AHMEDZAI et al.
nabilone on Day 2 because of severe drowsiness,
moderate postural dizziness and mild confusion: the
lowest recordings were 90/60 mm Hg supine,
80/55mm Hg erect, with a pulse rate of 110/min
(sinus rhythm). Within 3 days of discontinuing
nabilone her BP and pulse rate had returned to
normal pre-treatment levels and symptoms settled.
Pulse rates did not vary significantly with either
drug-the highest mean rates were 99/min for
prochlorperazine and 98/min for nabilone, on Day
2. Arrhythmias were not observed.
Discussion
There has been a surge of interest in anti-emetic
control for cytotoxic chemotherapy in recent years
(Anonymous, 1979; Laszlo & Lucas, 1981; Frytak
& Moertel, 1981). Newer agents being evaluated are
high-dose metoclopramide, dexamethasone and the
cannabinoids (Gralla et al., 1981; Trounce, 1982).
Most controlled studies of the latter have hitherto
been conducted in the United States; those of
nabilone have demonstrated useful oral efficacy and
superiority over oral prochlorperazine in mixed
groups of patients with various malignancies on
different chemotherapeutic regimes (Herman et al.,
1979; Steele et al., 1980). However a recent short
report from a British study has suggested less
satisfactory results and an unacceptably high
incidence of side-effects (Cornbleet et al., 1982).
Our study was conducted in patients with a
single tumour and histological type and all
undergoing identical chemotherapy of moderate
emetic potential. In addition the large majority of
our patients had not received prior chemotherapy,
and we are reasonably certain that they were not
experienced in marijuana nor were they prejudiced
by being alerted to the possibility that they may be
receiving a cannabis-like drug. For these reasons
this study contains a higher degree of control and
freedom from bias than the previously reported
ones, and may be of greater relevance to clinical
practice in Britain.
The results showed nabilone to be an effective
oral anti-emetic agent for a 3-day chemotherapy
regime containing Cyclophosphamide, Adriamycin
and Etoposide, completely obviating the use of
parenteral medication. For multiple-day schedules,
an oral anti-emetic is clearly to be preferred to
repeated parenteral doses of prophylactic or "on-
demand" anti-emetic drugs, or to daily courses of
high dose i.v. metoclopramide.
In this trial we placed emphasis on the patients'
subjective assessment of their gastro-intestinal
symptoms, including retching which is not usually
assessed but which some patients find as much if
not more distressing than vomiting. In the objective
assessment we did not find the measurement of
volume of vomitus to be sufficiently reliable to
include in the analysis, as others have attempted
(Gralla et al., 1981); the volume is also probably of
less relevance to the patients than the number of
vomiting or retching episodes.
Nabilone was superior in most of the evaluated
parameters to oral prochlorperazine, which
confirms the findings of two earlier studies. Unlike
the subjects of Herman et al. (1979), our patients
and those of Steele et al. (1980) did not
significantly prefer nabilone overall to
prochlorperazine for subsequent chemotherapy.
Since ours was a fixed-dose study, it is not possible
to say whether a reduction in dose for patients
experiencing troublesome side-effects would have
reduced these and improved subsequent
acceptability whilst retaining anti-emetic efficacy.
Single doses of nabilone in man have been shown
to have dose-related pharmacological effects, with
1 mg inducing relaxation and sedation but no dry
mouth or hypotension (Lemberger & Rowe, 1975).
The use of widely varying doses of both nabilone
and prochlorperazine in published studies has made
the interpretation of efficacy and toxicity more
difficult. Herman et al., (1979) actually reported on
two trials which were analysed together, in one of
which patients received 6mg of nabilone and 30mg
of prochlorperazine daily, and in the other 8mg of
nabilone and 40mg of prochlorperazine daily. Not
surprisingly, the reported toxicity of nabilone was
higher than that observed by Steele et al. (1980)
and our group who used a dose of 4 mg daily, and
prochlorperazine at 20 mg and 30 mg daily
respectively; the range and incidence of side-effects
in these trials are more comparable. The results
Steele et al. (1980) obtained for nabilone against
non-platinum agents and low-dose platinum were
also similar to ours. In their pilot study of nabilone
against a variety of chemotherapeutic drugs
including platinum, Cornbleet et al. (1982) used a
higher starting dose of 2mg 6-hourly for the first
12h, followed by 2mg 12-hourly. This increase in
dose was sufficient for the authors to find a "high
incidence (55%) of significant psychotropic side-
effects."
We did not find any statistical association
between age or sex and nabilone toxicity and so we
suggest that the dose of nabilone be restricted to
2mg 12-hourly for most patients for adequate anti-
emetic control aganist non-platinum drugs, with a
moderate but overall acceptable incidence of side-
effects. Indeed several patients at this dose reported
pleasant mental changes, and our limited experience
is that subsequent courses of nabilone at the same
or lower dose (1 mg 12-hourly) retained its efficacy
NABILONE IN LUNG CANCER THERAPY 663
with no increase in toxicity. However, the long-term
use of nabilone has yet to be studied.
Postural hypotension and tachycardia are
important pharmacological effects of the
cannabinoids (Anonymous, 1978), but this only
became a symptomatic problem in 7% of our
nabilone subjects. One of the 2 patients with severe
cardiovascular toxicity had a mitral valve
replacement; extra caution is therefore indicated
when using nabilone in patients with known cardiac
disease.
Furthermore, unpredictability of the adverse
effects discussed above demands the exercise of
caution in all patients receiving nabilone for the
first time, and we would recommend that they
receive the drug under in-patient supervision at
least for the first 24 hours.
We thank Drs G. Boyd, R.D. Stevenson and S.W.
Banham for allowing us to study their patients; Mr J.
Urwin of the Pharmacy Department of Glasow Royal
Infirmary for practical assistance; Miss Seonaid Buchanan
and Miss Irene Adams for secretarial services; and Mrs
Rosemary McCusker for the statistical computation.
Nabilone and placebo capsules were supplied by Lilly
Research Ltd.
References
ANONYMOUS (1978). Cannabis and the cardiovascular
system. Br. Med. J., i, 460.
ANONYMOUS (1979). Cancer chemotherapy-the inbuilt
deterrent. Br. Med. J., i, 1312.
CORNBLEET, M.A., HAMILTON, D.A., CHRISTIAN, P. &
SMITH, J.F. (1982). Evaluation of nabilone as an anti-
emetic. Br. J. Cancer, 46, 492.
FRYTAK, S. & MOERTEL, C.G. (1981). Management of
nausea and vomiting in the cancer patient. J. Am.
Med. Assoc., 245, 393.
FRYTAK, S., MOERTEL, C.G., O'FALLON, J.R. & 5 others.
(1979). Delta-9-tetrahydrocannabinol as an anti-emetic
for patients receiving cancer chemotherapy. Ann.
Intern. Med., 91, 825.
GRALLA, R.J., ITRI, L.M., PISKO, S.E & 6 others. (1981).
Anti-emetic efficacy of high-dose metoclopramide:
randomized trials with placebo and prochlorperazine
in patients with chemotherapy-induced nausea and
vomitting. N. Engl. J. Med., 305, 905.
HERMAN, T.S., EINHORN, L.H., JONES, S.E. & 8 others.
(1979). Superiority of nabilone over prochlorperazine
as an anti-emetic in patients receiving cancer
chemotherapy. N. Engl. J. Med., 300, 1295.
LASZLO, J. (1979). Tetrahydrocannabinol: From pot to
prescription? Ann. Intern. Med., 91, 916.
LASZLO, J. & LUCAS, V.S. (1981). Emesis as a critical
problem in chemotherapy. New Engl. J. Med., 305,
948.
LEMBERGER, L. & ROWE, H. (1975). Clinical
pharmacology of nabilone, a cannabinol derivative.
Clin. Pharmacol. Thera., 18, 720.
SALLAN, S.E., ZINBERG, N.E. & FREI, E. (1975). Anti-
emetic effect of delta-9-tetrahydrocannabinol in
patients receiving cancer chemotherapy. N. Engl. J.
Med., 293, 795.
STEELE, N., GRALLA, R.J., BRAUN, D.W. & YOUNG, C.W.
(1980). Double-blind comparison of the anti-emetic
effects of nabilone and prochlorperazine on
chemotherapy-induced emesis. Cancer Treat. Rep., 68,
219.
TROUNCE, J.R. (1982). Anti-emetics and cytotoxic drugs.
Br. Med. J., 286, 327.