A multicentre, double-blind study comparing placebo, ondansetron and ondansetron plus dexamethasone for the control of cisplatin-induced delayed emesis

Annals of Oncology!: 945-952, 1996. O 1996 Kluwer Academic Publishers. Printed in the Netherlands.

Original article A multicentre, double-blind study comparing placebo, ondansetron and ondansetron plus dexamethasone for the control of cisplatin-induced delayed emesis I. Olver,1 W. Paska,2 A. Depierre,3 J.-F. Seitz,4 D. J. Stewart,^ L. Goedhals,6 B. McQuade,2 J. McRae2 & J. R. Wilkinson2 on behalf of the Ondansetron Delayed Emesis Study Group 1

Royal Adelaide Hospital, Adelaide, Australia; 2 Glaxo Wellcome Research and Development, Greenford, UK; 3 Centre Hospitalier Universitaire, Besancon; AInstitutJ. Paoli I. Calmette, Marseille, France; 3Ottawa Regional Cancer Centre, Ottawa, Canada; 6National Hospital, Bloemfontein, South Africa Summary

Background: The purpose of this study was to investigate the efficacy and safety of oral ondansetron, given alone or in combination with dexamethasone in the control of cisplatininduced delayed emesis. Patients and methods: This was an international, multicentre, double-blind, randomised, placebo-controlled, parallel group study. A total of 640 chemotherapy-naive patients received ondansetron 8 mg i.v. and dexamethasone 20 mg i.v. for the control of acute emesis prior to cisplatin (> 70 mg/m2) on day 1. Patients who were not rescued or withdrawn on day 1 were to be randomised 24 hours after the start of cisplatin administration to one of four groups; group I placebo oral (p.o.), twice daily (bd) on days 2-6 (n — 125); group II ondansetron (8 mg p.o. bd) on days 2/3 followed by placebo (p.o. bd) on days 4-6 (n — 199); group HI ondansetron (8 mg p.o. bd) on days 2-6 (n - 214); group IV ondansetron (8 mg p.o. bd) plus dexamethasone (4 mg p.o. bd) on days 2-6 (n - 66). Results: On day 1, 81% of patients had complete control of acute emesis, with 68% having no emesis and no nausea. Over days 2/3 and over days 2—6, significantly more patients receiving ondansetron plus dexamethasone (group IV) reported no emesis and no nausea (49% and 45%, respectively) compared to ondansetron alone (32% and 27%, respectively) or placebo (group I; 33% and 27%, respectively; P< 0.05 for all pairwise comparisons). There were no significant Key words: cisplatin, delayed emesis and nausea, dexadifferences in the control of emesis over days 2/3, where methasone, ondansetron, 5-HT3 receptor antagonist

Introduction The pattern of emesis following various chemotherapy regimens differs. With cisplatin treatment the pattern appears to be biphasic [1], the first peak corresponding to acute emesis which occurs within 3-5 hours after cisplatin then gradually subsides by 24 hours [2,3], the second peak corresponding to delayed emesis, although often less severe, may last for up to one week but is usually most intense between 24 and 72 hours [1, 6]. This contrasts with many other chemotherapy

agents such as cyclophosphamide [4, 5] which induce a different pattern of emesis giving a monophasic or single prolonged phase extending over several days. Delayed emesis following cisplatin, although often less intense than acute emesis, is still a major problem for many patients particularly as they may be at home when it occurs, and therefore it may affect the quality of life of the patient The incidence of delayed emesis and nausea varies but it can be as high as 80% [1,6]. Ondansetron, a highly selective 5-HT3 receptor antagonist, has been shown to be superior to conven-

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

61% of patients had complete emetic control (0 emetic episodes) with ondansetron plus dexamethasone (group IV), 54% with ondansetron (groups II + HI) and 49% with placebo (group I). In the distribution of nausea grades, ondansetron plus dexamethasone (group IV) was significantly superior to ondansetron (groups II + IH; P - 0.037) and placebo (group I; P - 0.013) over days 2/3. Over days 2-6 there were no significant differences in the control of emesis, however a comparison of the distribution of nausea grades over days 2-6 showed ondansetron plus dexamethasone (group TV) to be significantly superior to ondansetron (group HI; P - 0.043) and placebo (group I; P - 0.024). All treatments were well tolerated and no unexpected drug-related adverse events were reported. There were no differences in the overall incidence of adverse events between the active treatment groups or placebo. Constipation and headache, recognised side effects of 5-HT3 receptor antagonists, were the most commonly reported adverse events with the incidence of constipation with ondansetron alone (group HI) being significantly greater than with placebo over days 2-6 (14% vs. 6%; P - 0.030). Conclusion: In contrast to some previous investigations, in this study, ondansetron alone appears to have a minor role in the control of cisplatin-induced delayed emesis and nausea. In conclusion, the combination of oral ondansetron plus dexamethasone is superior to ondansetron and to placebo.

946

Patients and methods

dose of ondansetron, they had vomited, retched or had chronic nausea, or if they had received anti-emetics. Patients were also excluded if they had abdominal or pelvic irradiation within 48 hours prior to inclusion into the study or were scheduled to receive radiotherapy during the study or cisplatin on days 2-6 of the study or moderately to highly emetogenic chemotherapy prior to or during the study. Treatment Acute phase (day 1) All patients were to be hospitalised during the first 24 hours of the study. Approximately 45 minutes prior to receiving cisplatin all patients received dexamethasone 20 mg (as sodium phosphate) as a slow intravenous injection over 15 minutes. Approximately 30 minutes prior to cisplatin all patients received ondansetron 8 mg (as hydrochloride dihydrate) as a slow intravenous injection over 15 minutes. Cisplatin >70 mg/m2 was given as an intravenous infusion over a period of not more than three hours. Administration of cisplatin was to begin no later than 14.00 hours and where cyclophosphamide (<1000 mg/m2), doxorubicin (<35 mg/m2), epirubicin (<50 mg/m2) or methotrexate (<200 mg/m2) were administered with cisplatin, these were also to be started no later than 14.00 hours. Delayed phase (day 2-6) Patients who were not rescued with other anti-emetics or withdrawn for any reason on day 1 were randomised 24 hours after starting cisplatin treatment to one of four oral treatment regimens for five consecutive days of the delayed phase (Table 1). Patients were allocated to one of the treatment groups according to a pre-determined computer-generated randomisation code. Administration of oral study drug was double-blind. Medication was dispensed in a blister pack, the first dose being taken 24 hours after the initiation of cisplatin, the second dose being taken in the evening of day 2 at approximately 20.00 hours. For days 3-6, each dose of oral medication was to be taken every 12 hours. The times of administration of oral study medication were recorded by the patient on the diary card. All drug supplies were accounted for on a study administration log kept by the pharmacist.

Conduct of the study This was a multicentre, randomised, double-blind, parallel group study involving 41 centres in 12 countries (Austria, Australia, Belgium, Canada, France, Germany, Ireland, Israel, The Netherlands, Portugal, South Africa and the UK). The study was carried out according to Good Clinical Practice Guidelines and to the principles of the Declaration of Helsinki as modified by the 41st World Medical Assembly, Hong Kong 1989. The protocol received approval from all Regulatory Authorities and local Ethics Committees as appropriate to the countries in which the study was carried out Written informed consent was obtained from patients prior to entry into the study.

Assessments Efficacy On each of days 1-7, the number and timing of emetic episodes was recorded on a 'vomit/retch' card in hospital on day 1 and by the patient on a diary card on days 2-7. An emetic episode was defined as a single vomit or retch or any number of continuous vomits and/ or retches where a retch was a heave not producing liquid. Emetic episodes were by definition separated by absence of vomiting or Table 1. Treatment groups-days 2-6.

Patients Randomisation Chemotherapy-naive cancer patients, male and female, aged at least 12 years (18 years in France), scheduled to receive their first course of cisplatin chemotherapy given as a single dose of >70 mg/m2 over a period of up to three hours, either alone or in combination with other cytotoxic agents, were included. Patients were to be excluded if they were pregnant, had received an investigational drug in the previous 30 days, had contraindications to corticosteroid use or had other aetiologies for vomiting or nausea (e.g., gastrointestinal obstruction, hypercalcaemia, active peptic ulcer disease or central nervous system metastases). In addition, patients were to be excluded if, in the 24 hours prior to the first

groups Group I Group II

Group m Group IV

Days 2

3

4

5

6

P O O O+D

P O O O+D

P P O O+D

P P O O+D

P P O O+D

P - placebo p.o. bd; O - ondansetron (as the hydrochloride dihydrate salt) 8 mg p.o. bd; O + D ondansetron 8 mg p.o. bd plus dexamethasone 4 mg p.o. bd.

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

tional anti-emetics such as high-dose metoclopramide for the control of acute emesis induced by cisplatin [2, 3]. In addition, dexamethasone has been shown to enhance the efficacy of both ondansetron and metoclopramide in this situation [7-9]. Complete control of acute emesis (0 emetic episodes) can be obtained in up to 90% of patients with the combination of ondansetron and dexamethasone [8]. Ondansetron plus dexamethasone has recently been shown to be superior to metoclopramide, dexamethasone and diphenhydramine in the control of acute emesis over three courses of cisplatin therapy [10] and has also been shown to be significantly better than the combination of metoclopramide, dexamethasone plus lorazepam in the control of emesis and nausea on days 1-5 over three consecutive courses of cisplatin chemotherapy [11]. Several randomised studies have investigated conventional anti-emetics in the control of cisplatininduced delayed emesis [1-3,6,12-14]. Many of these have been inconclusive, furthermore, some studies have not been adequately designed, in that small numbers of patients were used, patients were randomised before chemotherapy and received randomised treatment throughout the acute and delayed phases. Such study designs do not take into account possible differential drop-out and failure rates on day 1, thus leading to potential bias during the delayed phase. In the present study, all patients received the same treatment on day 1 for acute emesis i.e., ondansetron plus dexamethasone. Those who were not rescued or withdrawn on day 1 were to be randomised after 24 hours to assess the control of delayed emesis and nausea

947 retching for at least one minute. Nausea was assessed on each day by the patient using the scale, none - no nausea; mild - did not interfere with normal daily life; moderate - interfered with normal daily life; or severe - bedridden due to nausea. Patients' Global Satisfaction with the control of nausea and vomiting was assessed using a visual analogue scale. The patient was asked to mark on a 100 mm line where 0 mm was 'not at all satisfied' and 100 mm was 'totally satisfied'. Safety All adverse events were documented throughout the study period. The severity of each adverse event and the relationship to the study treatments was assessed by the investigator. Pairwise Fisher's exact tests were used to compare the proportions of patients experiencing adverse events. Routine haematological tests were taken pre- and post-treatment Statistical methods

Duration of control of delayed emesis To determine the duration of treatment required to control delayed emesis and nausea comparisons between groups were made for days 4, 5 and 6. For each day, only patients who successfully completed the previous days (i.e., not rescued or withdrawn before the start of that day) were considered. Comparisons were also made between treatment groups during the overall delayed phase i.e., days 2-6. Analyses were limited to comparisons of group I, group HI and group IV. The proportions of patients with a complete emetic response (0 emetic episodes) or no emesis and no nausea were compared between each pair of treatments using two-sided Mantel-Haenszel chisquared tests, stratified by cluster of centres. For the purposes of the analysis, centres were grouped into geographical clusters such that each cluster had between 40-80 patients. The distribution of nausea grades and the change in global satisfaction scores from day 1 value were compared between treatment groups using Wilcoxon rank sum tests, stratified by cluster of centres using the van Elteren procedure.

Sample size

Populations The safety population included all patients who entered the study and received at least one dose of study medication. The intent-totreat population comprised all patients in the safety population who received cisplatin-containing chemotherapy, who were randomised into the delayed phase of the study and took at least one dose of study medication in the delayed phase. The efficacy population included all patients in the intent-to-treat population who had no major protocol violations (i.e., violations affecting efficacy for each analysis period). The efficacy population was used for a separate analysis of the emesis and nausea data. The results shown are based on the intent-to-treat population, since this was the primary analysis population. Control of delayed emesis The primary objective of this study was to compare the efficacy of ondansetron alone or combined with dexamethasone, with placebo from the start of day 2 (24 hours after cisplatin administration) until midnight on day 3. Days 2/3 were grouped together as one time period. Groups n and HI (see Table 1 for details of treatments) were combined for this treatment period, since all patients had received the same treatment for that study period. All pairwise treatment comparisons were made between the combined groups U and HI, group I and group IV over days 2/3.

Results Patient characteristics A total of 642 patients were recruited into the study on day 1 and 640 received study treatment. Of these, 604 (intent-to-treat population) were randomised into the delayed phase, 125 in group I, 199 in group n, 214 in group HI and 66 in group IV. The main reasons for exclusion from the efficacy population were incorrect timing of the first oral dose (8%), incorrect number of tablets taken (3%) and administration of excluded chemotherapy/doses (3%). All 604 patients who received randomised study treatments were included in the analysis of safety and all patients in the intent-to-treat population were included in the analysis of efficacy. The treatment groups were well matched with respect to the demographic factors and chemotherapy details for the intent-to-treat population as shown in Table 2 and Table 3, respectively. The most common primary tumour site was lung (29%). Control of emesis and nausea on day 1 (prior to randomisation) Overall, 81% of patients reported no emesis, 93% experienced <2 emetic episodes and 68% reported no emesis and no nausea. There was an imbalance in the control of emesis in treatment groups I-IV on day 1. Patients subsequently randomised to placebo had greater control of emesis on day 1 (Figure 1). Control over days 2/3 A higher proportion of patients in the ondansetron plus dexamethasone group (group IV) experienced a complete emetic response (0 emetic episodes) compared to the placebo group (group I) (61% vs. 49%, respectively; P = 0.084, Figure 2). In the ondansetron

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

In order to calculate the sample sizes, it was assumed that 35% patients given placebo, 55% patients given ondansetron and 75% patients given ondansetron plus dexamethasone would experience complete control of emesis during day 2-6. With 128 patients in group I, 192 patients in each of groups II + III and 64 patients in group IV, the powers to detect treatment differences at the 5% level of significance were 94% (ondansetron vs. placebo), >99% (ondansetron plus dexamethasone vs. placebo) and 83% (ondansetron plus dexamethasone vs. ondansetron). In addition, it was assumed that 10% of ondansetron patients would withdraw over days 2/3 and 5% on each days 4 and 5. Assuming these withdrawal rates and given the size of each of the groups, this would give the study 84% power to detect treatment differences between ondansetron (groups II + III) and placebo (group I) on days 2/3, 81% for ondansetron (group III) vs. placebo (group n) on day 4, 80% for ondansetron (group IH) vs. placebo (group D) on day 5 and 81% for ondansetron (group IH) vs. placebo (group II) on day 6. No additional allowance was made for any differential drop-out rate after day 4 between groups n and HI, so treatment comparisons on days 5 and 6 may be biased.

948 D Complete emetic response (0 emetic episodes)

Table 2. Patient demography (intent-to-treat population).

• No emetic episodes and no nausea

Randomisation groups Placebo (group I)

125 78 (62%) 47 (38%) 59 11 32-82 123 64 12 39-98 125

413

66

261 (63%) 152 (37%)

37 (56%) 29 (44%)

59 11 23-84 407

59 12 35-81 64

67 14 37-112 413

64 14 30-120 66

66 (53%) 37 (30%) 22 (18%) 23 (18%)

216 (52%) 132 (32%) 62 (15%) 63 (15%)

39 (59%) 21 (32%) 6 (9%) 9 (14%)

36 (29%) 26 (21%) 27 (22%) 17 (14%) 15 (12%) 7 (6%)

123 (30%) 99 (24%) 74 (18%) 62 (15%) 51 (12%) 32 (8%)

16 (24%) 18 (27%) 13 (20%) 11 (17%) 8 (12%) 6 (9%)

79% 73%

n»124 n«123 Not recorded (n)

1 2

Rwdomisition

* Unit of alcohol - one measure of spirit, one glass of wine or 250 ml of beer. b Patients can have more than one primary tumour site.

Group I

n=198 n-198

n*214 n«212

1 1

Group II

0

2

Group III

n*66 n«$4 0

2

Group IV

group! futknu uibiequcntly assigned to on Days 2-6

Figure 1. Control of emesis and nausea on day 1 (prior to randomisation). Group I received placebo on days 2-6; group II received ondansetron on days 2-3 and placebo on days 4-6; group III received ondansetron on days 2-6; group IV received ondansetron plus dexamethasone on days 2-6.

Table 3. Most common* chemotherapy regimens (intent-to-treat population). Randomisation groups treatment11

Number of patients Cisplatin alone Cisplatin with 5-fluorouracU Cisplatin with cyclophosphamide Cisplatin dose (mg/m2) Median Dose <90 mg/m2 Dose >90 mg/m2

Group I

Group II

Group III

Group IV

125 21(17%)

199 30(15%)

214 29(14%)

66 9(14%)

35 (28%)

59 (30%)

55 (26%)

20 (30%)

18(14%)

30(15%)

38(18%)

12(18%)

80 81(65%) 44 (35%)

80 128(64%) 71 (36%)

80 140(65%) 74 (35%)

80 44 (67%) 22 (33%)

* Most common denned as > 10% of patients receiving the cytotoxic agent Between 5%—10% of patients received bleomycin, doxorubicin, etoposide, methotrexate and vincristine in combination with cisplatin. b Group I received placebo on days 2-6; group II received ondansetron on days 2-3 and placebo on days 4-6; group HI received ondansetron on days 2-6; group IV received ondansetron plus dexamethasone on days 2-6.

group (group II and HI), 54% of patients had a complete emetic response. None of the pairwise treatment comparisons of efficacy of emesis was statistically significant. No nausea was reported by 52% of patients who

n«121 Not recorded (n) 4 Groups

n-123 2 I

II & III I I I

IV

Figure 2. Control of delayed emesis (0 emetic episodes) on days 2/3 and days 2-6. Group I received placebo on days 2-6; group n received ondansetron on days 2-3 and placebo on days 4-6; group HI received ondansetron on days 2-6; group IV received ondansetron plus dexamethasone on days 2—6. No significant difference between any pair of treatments.

received ondansetron plus dexamethasone (group IV), 35% who received ondansetron (groups II + HI) and 34% of patients who received placebo (group I; Table 4). Of those patients who reported no nausea on day 1, 66% of patients who received ondansetron plus dexamethasone compared to 45% who received ondanse-

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

Number of patients Sex Male Female Age (years) Mean SD Range n Weight (kg) Mean SD Range n Alcohol consumption' Non-user <4 units per day >4 units per day Prior heavy use Type of tumourb Lung Gynaecological Head and neck Gastrointestinal Genito-urinary Other

+ m)

Ondansetron + dexamethasone (group IV)

Ondansetron (groups n

949 Table 4. Control of delayed nausea - days 2/3 (intent-to-treat population).

Table 5. Control of delayed nausea - worst day of days 2-6 (intentto-treat population).

Treatments

Treatments

Number of patients" None Mild Moderate Severe/rescued/withdrawn due to lack of response Number of patients for whom data were missing

Placebo (group I)

Ondansetron (group II + ffl)

Ondansetron + dexamethasone (group IV)

125 41 (34%) 26 (21%) 21 (17%)

405 139 (35%) 90 (22%) 72 (18%)

64 33 (52%) 10 (16%) 6 (10%)

33 (27%)

101 (25%)

14 (22%)

4

3

1

Ondansetron (group III)

Ondansetron + dexamethasone (group TV)

125 34 (27%) 24 (19%) 23 (19%)

208 61 (29%) 39 (19%) 32 (15%)

64 29 (45%) 9 (14%) 8 (13%)

43 (35%)

76 (37%)

18(28%)

1

0

0

Group in vs. group I: not significant; group HI vs. group IV: P — 0.043; group IV vs. group I: P — 0.024. Wilcoxon rank sum test for treatment differences, stratified by cluster of centres using the van Elteren procedure. Values are numbers of patients with percentages in parentheses. * Only includes patients who were not rescued or withdrawn prior to day 2. 100

tron and 41% who received placebo reported no nausea over days 2/3. In the analysis of the distribution of nausea grades (Table 4), significant differences were found in favour of the ondansetron plus dexamethasone group compared to the ondansetron (P — 0.037) and the placebo (P-0.013) group. Control over days 2-6 Over days 2-6, no emesis was experienced by 55% of patients who received ondansetron plus dexamethasone (group IV) compared with 47% who received ondansetron (group in) and 42% who received placebo (group I; Figure 2). None of these pairwise treatment differences were significant. A comparison for the distribution of nausea grades showed ondansetron plus dexamethasone to be significantly superior to both ondansetron (P - 0.043) and placebo (P = 0.024) in the control of nausea over days 2-6 (Table 5).

so

f 6° a

I 40

49% 33%

32%

n=120

n-401

20 •

Not recorded (n)

5

Groups

4

II & III

n=63 1

rv

Figure 3. Control of delayed emesis and nausea (no emesis and no nausea) on days 2/3. Mantel-Haenszel chi-squared test for treatment difference stratified by cluster of centres. Groups II + m vs. group I: not significant; groups n + in vs. group IV: P— 0.006; group IV vs. group h P— 0.018. Group I received placebo on days 2-6; group H received ondansetron on days 2-3 and placebo on days 4-6; group III received ondansetron on days 2-6; group IV received ondansetron plus dexamethasone on days 2-6.

Combined emesis and nausea Global satisfaction There were significantly more patients reporting no emesis and no nausea over days 2/3 (Figure 3) and days 2-6 (Figure 4) in the ondansetron plus dexamethasone group (group IV) compared to the ondansetron (groups II + IE days 2/3 and group ITJ days 2-6) or placebo (group I) groups. The proportion of patients experiencing neither emesis nor nausea in the ondansetron plus dexamethasone group were also significantly higher than for those patients receiving ondansetron on each of days 4 and 5 and placebo on each of days 4,5 and 6 (Figure 5).

The satisfaction with the control of emesis and nausea over days 2-6 was significantly superior with ondansetron plus dexamethasone (group IV) than with ondansetron (group HI; P = 0.031) or placebo (group I; P - 0.011). Tolerability Ondansetron was well tolerated in this study both when given intravenously in combination with dexametha-

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

Groups II + HI vs. group I: not significant; groups II + HI vs. group IV: P - 0.037; group IV vs. group I: P - 0.013. Wilcoxon rank sum test for treatment differences, stratified by cluster of centres using the van Elteren procedure. Values are numbers of patients with percentages in parentheses. ' Only includes patients who were not rescued or withdrawn prior to day 2.

Number of patients1 None Mild Moderate Severe/rescued/withdrawn due to lack of response Number of patients for whom data were missing

Placebo (group I)

950 sone on day 1 or orally over days 2-6 (Table 6). There was no statistical difference in the overall incidence of adverse events between the active treatment groups or placebo over days 2-6. The incidence of individual adverse events was also similar for all treatments with the exception of constipation. The incidence of constipation with ondansetron alone (group HI) was higher than in the placebo group (14% vs. 6%, P - 0.030, Table 6). Constipation and headache, well

D Placebo BOndansatron • Ondaraatron plus d«x»mtha»ono

100

80

60

Ia.n 40

n=164n=167 n=52

27%

27%

Not recorded (n) 3

20

2

1

3

1

0

n=152 n=152 n=49 3

1

1

Figure 5. Daily control of delayed emesis and nausea (no emesis and no nausea) on days 4-6. Ondansetron plus dexamethasone vs. placebo: * P— 0.011, ** P — <0.001, ' f - 0.004; ondansetron vs. ondansetron plus dexamethasone: *P- 0.009, ' P- 0.010, ns- not significant

n=64 0

n=208 0

n=123 Not recorded (n)

1

n=159 n=157 n=52

recognised side effects of 5-HT3 receptor antagonists, were the most commonly reported adverse events. There were no reports of headaches by patients receivFigure 4. Control of delayed emesis and nausea (no emesis and no ing ondansetron plus dexamethasone over days 2-6. nausea) on days 2-6. Mantel-Haenszel chi-squared test for treatSerious adverse events were rare and the majority of ment difference stratified by cluster of centres. Groups II + III vs. group I: not significant; groups II + III vs. group IV: P — 0.005; these were assessed as due to the patient's underlying group IV vs. group I: P - 0.006. Group I received placebo on days cancer or the chemotherapy regimen and none was as2-6; group II received ondansetron on days 2-3 and placebo on sessed by the investigator to be related to the study days 4-6; group III received ondansetron on days 2-6; group IV remedication. Withdrawals due to adverse events were ceived ondansetron plus dexamethasone on days 2-6. diverse in nature and reported in similar and low incidence in all groups (< 3%). There was no evidence of any Table 6. Summary of the most common adverse events'-day 1 and significant treatment effect on any laboratory parameter. days 2-6.

Groups

Adverse event

IV

III

I

Treatments Day l d

Number of patients 640 Any adverse event 53 (8%) Constipation 10(2%) 19 (3%) Headache

Discussion Days 2-6 b

Group I

Group U

Group UI

Group IV

125

199

214

66

32 (26%) 7 (6%) 8(6%)

54 (27%) 21 (11%) 13(7%)

65 (30%) 22 (33%) 29(14%) c 9(14%) 13(6%) 0(0%)

" Most common defined as >3% of patients for day 1 and >5% of patients for days 2-6. Some patients experienced more than one adverse evenL b Only includes patients randomised into the delayed phase. Group I received placebo on days 2-6; group LI received ondansetron on days 2-3 and placebo on days 4-6; group III received ondansetron on days 2—6; group IVreceivedondansetron plus dexamethasone on days 2-6. c Group HI vs. group I; P- 0.030. i Day 1 - patients received ondansetron 8 mg i.v. and dexamethasone 20 mg i.v. Values are number of patients with percentage in parentheses.

The control of emesis and nausea following chemotherapy, and the maintenance of control over subsequent cycles has a profound effect on patients' quality of life and willingness to complete their course of treatment [16-18]. As the majority of patients return home after receiving their chemotherapy, the need for continued protection from these symptoms is paramount, particularly when highly emetogenic agents such as cisplatin are given. Most anti-emetic trials employing cisplatin chemotherapy have only evaluated the acute phase of emesis. Some studies have reported results from the delayed phase by continuing the anti-emetic treatment from day 1 [2, 3, 15]; however, very few have randomised treatments at the start of the delayed phase (i.e., 24 hours

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

5?

45%

951

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

after cisplatin), thereby avoiding any bias or imbalance results from these two studies are unclear. The study from the control of emesis and nausea on day 1. designs are similar except that ondansetron alone (0.15 Recently, two large cisplatin-induced delayed emesis mg/kg i.v. x 3) was given on day 1 to control acute studies employing the 5-HT3 receptor antagonists emesis and patients were randomised after 24 hours to granisetron [19] and ondansetron [20] have been re- oral ondansetron (8 mg bd day 2/3 or days 2-6) or ported. The granisetron study showed there was no sig- placebo bd. The difference in the results between the nificant difference in the control of emesis and nausea two studies may in part be attributed to the control of between the granisetron groups and the placebo group emesis produced by the different ondansetron region days 2-4, although a difference was seen over days mens employed during the acute phase. The proportion 5-7. In contrast, a recently completed study with of patients with a complete emetic response on day 1 in ondansetron [20] demonstrated a statistically signifi- this present study was 81% compared with 64% in the cant effect for ondansetron over placebo for both study by Navari et al. [20]. It may be that the better emesis and nausea over days 2/3, a time when the inci- control of acute emesis seen in this study leads to a dence of delayed emesis is at its highest, and also on lower incidence of delayed emesis. Furthermore, padays 4 and 5 for emesis and on day 4 for nausea. tients who failed treatment during the acute phase were The 5-HT3 receptor antagonists such as ondanse- not eligible for randomisation into the delayed phase, tron are highly effective in the control of acute cis- therefore, the incidence of emesis and nausea in the platin-induced emesis, particularly when used in com- delayed phase may be artificially low. The incidence of bination with dexamethasone [8,9]; however, their role emesis in the placebo group on days 2/3 in the present in delayed emesis is less clear. In this present study, all study was 51% compared to 74% in the study by patients received ondansetron plus dexamethasone to Navari et al. [20]. As a consequence, there was more control acute emesis. Patients were then randomised at opportunity for improvement in the control of delayed the start of the delayed phase, i.e., 24 hours after cis- emesis in the study by Navari et al. [20], which may platin administration, to one of four treatment groups explain the significant superiority of ondansetron alone (Table 1). A numerical trend for superior control of compared to placebo. delayed emesis and a significant effect for nausea was Recently, Roila et al. [22] showed that patients who observed for ondansetron plus dexamethasone com- experienced cisplatin-induced emesis in the acute pared to the other treatment groups over days 2/3 and phase, achieved significantly better protection from days 2-6. Nausea as well as emesis has been ranked by delayed emesis when subsequently treated with onpatients as the worst side effect associated with chemo- dansetron plus dexamethasone compared to those therapy treatment [21]. The majority of patients will be treated with metoclopramide plus dexamethasone. All treated on an out-patient basis during the delayed patients received ondansetron plus dexamethasone in phase, therefore, effective control of both emesis and the acute phase. These findings suggest that ondansenausea is of paramount importance for the patient The tron has a role to play in delayed emesis. A dexamethapresent study shows that a significantly greater propor- sone alone treatment arm was not included in this pretion of patients receiving ondansetron plus dexametha- sent study, therefore, no definitive conclusion can be sone had neither emesis nor nausea throughout the made on the efficacy steroids alone would have in the delayed phase compared to patients receiving placebo. delayed phase. Dexamethasone has been shown to have A recent study by Roila et al. [15] has shown that the some activity on delayed emesis/nausea [6, 23]. There incidence of delayed emesis appears to be influenced by is the possibility there is a synergistic effect between the control of acute emesis. The proportion of patients ondansetron plus dexamethasone. Studies have shown experiencing no emetic episodes in the acute phase of that dexamethasone given with ondansetron enhances this study was higher in those patients subsequently ran- significantly the efficacy of ondansetron in the acute domised to placebo treatment compared to those ran- phase [8, 9]. The findings from this study and the two domised to ondansetron plus dexamethasone (Figure 1). recent studies [20, 22] indicate that 5-HT3 receptor Therefore, the incidence of delayed emesis and nausea antagonists have a role in the prevention of cisplatinin the placebo group may be low relative to the other induced delayed emesis and nausea, either alone [20] treatment groups. As a consequence, the results may be or in combination with dexamethasone as shown in this biased towards the placebo arm. present study and recently by Roila et al. [22]. In contrast to the findings in this study, the study by All treatment regimens were well tolerated. There Navari et al. [20] showed that ondansetron alone was was no statistical difference in incidence of adverse significantly more effective than placebo in controlling events between the treatment groups. There was also delayed emesis and nausea over days 2/3, 4 and 5. A little difference in the incidence of individual adverse complete emetic response (0 emetic episodes) was events between the four treatment groups. The only achieved in 36% of patients receiving ondansetron notable adverse drug reactions reported were constipacompared to 26% of patients receiving placebo over tion and headache, both recognised side-effects of days 2/3 with a complete or major response (<2 emetic 5-HT3 receptor antagonists. Interestingly, no headaches episodes) achieved in 56% and 37% of patients, were reported over days 2-6 in patients receiving respectively (P - 0.001). The reasons for the different ondansetron plus dexamethasone, although headache

952 was reported on day 1 in 3% of patients receiving ondansetron plus dexamethasone combination. The incidence of headache in the placebo group (6%) was similar to that seen in the ondansetron alone treated groups (6%-7%) over days 2-6. These findings suggest that not all reported headaches are related to active treatment. Serious adverse events were reported rarely and the majority of these were assessed as due to the patient's underlying cancer or to the chemotherapy regimen; none were assessed by the investigator as related to study medication. In contrast to some previous investigations, in this study, ondansetron alone appears to have a minor role in the control of cisplatin-induced delayed emesis and nausea. In conclusion, the combination of oral ondansetron plus dexamethasone is superior to ondansetron and to placebo.

7. 8.

9. 10.

11.

12.

This clinical investigation was supported by Glaxo Wellcome Research and Development, Greenford, Middlesex, UK. (Australia) J. Bishop, R. M. Fox, I. Olver, R. StuartHarris; (Austria) H. Kienzer, R. Lenzhofer, D. Lutz, J. Schuller, P. Sevelda; (Belgium) H. Bleiberg, J. Michel, J. De Greve, G. De Wasch; (Canada) P. Klimo, D. J. Stewart; (France) A. Depierre, X. Froger, M. Giovannini, C. Martin, L. Mignot, J. Rimailho, J.-F. Seitz, N. Tubiana; (Germany) P. O. Augener, D. Christmann, M. Schmidt, R. Gurtler; (Ireland) D. Carney; (Israel) A. Figer, M. Inbar, B. Kaufman; (The Netherlands) A. C. Dullemond-Westland; (Portugal) H. Queiroga; (South Africa) L. Goedhals; (United Kingdom) R. Coleman, C. Collis, N. G. P. Davidson, E. D. Gilby, H. C. Kitchener, T. J. Priestman, J. T. Roberts, M. Soukop.

References 1. Kris MG, Gralla RJ, Clark RA et al. Incidence course and severity of delayed nausea and vomiting following the administration of high-dose cisplatin. J Clin Oncol 1985; 3:1379-84. 2. De Mulder PHM, Seynaeve C, Vermorken JB et al. Ondansetron compared with high-dose metoclopramide in prophylaxis of acute and delayed cisplatin-induced nausea and vomiting. Ann Intern Med 1990; 113: 834-40. 3. Marty M, Pouillart P, Scholl S et al. Comparison of the 5-hydroxytryptamine3 (serotonin) antagonist ondansetron (GR 38032F) with high-dose metoclopramide in the control of cisplatin-induced emesis. N Engl J Med 1990; 322: 816-21. 4. Morran C, Smith DC, Anderson DA, McArdle CS. Incidence of nausea and vomiting with cytotoxic chemotherapy: A prospective randomised trial of anti-emetics. BMJ 1979; 1879: 1323-4. 5. Martin Jimenez M, Diaz-Rubio E, Blazquez Encinar JC et al. Tolerancia, toxicidad y aceptacion del tratamiento quimioterapico FAC (5-fluorouracilo, adriamicina y ciciofosfamida) en el cancer de mama, Analsis de 100 pacientes consecutivas. Neoplasia 1988; 5: 8-14. 6. Kris MG, Gralla RJ, Tyson LB et al. Controlled delayed vomiting: Double-blind, randomised trial comparing placebo, dexa-

13. 14. 15.

16. 17. 18.

19.

20.

21. 22.

23.

Received 26 April 1996; accepted 2 October 1996. Correspondence to: Dr. Walter Paska Glaxo Wellcome Research & Development Greenford Road Greenford Middlesex UB6 0HE, UK

Downloaded from http://annonc.oxfordjournals.org/ by guest on June 6, 2016

Acknowledgements — Ondansetron Delayed Emesis Study Group

methasone alone and metoclopramide plus dexamethasone in patients receiving cisplatin. J Clin Oncol 1989; 7:108-14. Allan SG, Combleet MA, Warrington PS et al. Dexamethasone and high-dose metoclopramide: Efficacy in controlling cisplatin induced nausea and vomiting. BMJ 1984; 289: 878-9. Roila F, Tonato M, Cognetti F et al. Prevention of cisplatininduced emesis: A double-blind, multicentre, randomised, cross-over study comparing ondansetron and ondansetron plus dexamethasone. J Clin Oncol 1991; 9:675-8. Smyth JF, Coleman RE, Nicolson M et al. Does dexamethasone enhance control of acute cisplatin-induced emesis by ondansetron? BMJ 1991; 303:1423-6. Italian Group for Anti-emetic Research. Difference in persistence of efficacy of two anti-emetic regimens on acute emesis during cisplatin chemotherapy. J Clin Oncol 1993; 11(12): 2396-404. Cunningham D, Dicato M, Verweij J et al. Optimum antiemetic therapy for cisplatin-induced emesis over repeat courses: Ondansetron plus dexamethasone compared with metoclopramide, dexamethasone plus lorazepam. Ann Oncol 1996; 7: 277-82. Grunberg SM, Groshen S, Stevenson LL et al. Double-blind, randomised study of 2 doses of oral ondansetron (GR38032F) for the prevention of cisplatin-induced delayed nausea and vomiting. Proc Am Soc Clin Oncol 1990; 9: 327 (Abstr 1264). Gandara DR. Progress in the control of acute and delayed emesis induced by cisplatin. Eur J Cancer 1991; 27 (Suppl 1): S9-S11. Kris MG, Tyson LB, Clark RA et al. Oral ondansetron for the control of delayed emesis after cisplatin. Cancer 1992; 70: 1012-6. Italian Group for Anti-emetic Research. Ondansetron + dexamethasone vs. metoclopramide + dexamethasone + diphenhydramine in prevention of cisplatin-induced emesis. Lancet 1992; 340: 96-9. Laszlo J. Nausea and vomiting as major complications of cancer chemotherapy. Drugs 1983; 25(1): 1-7. Lindley CM, Hirsch JD, O'Neill CV et al. Quality of life consequences of chemotherapy-induced emesis. Qualify of Life Res 1992; 1:331-40. Soukop M, McQuade B, Hunter E et al. Ondansetron compared with metoclopramide in the control of emesis and quality of life during repeated chemotherapy for breast cancer. Oncology 1992; 49: 295-304. Smyth J. Delayed emesis after high-dose cisplain - the residual problem. Proceedings of the SmithKline Beecham Satellite Symposium to the XVJJ Congress of the European Society for Medical Oncology 1992; 24-26. Navari MN, Madajewicz S, Anderson N et al. Oral ondansetron for the control of cisplatin-induced delayed emesis: A large, multicentre, double-blind, randomised, comparative trial of ondansetron versus placebo. J Clin Oncol 1995; 13(9): 2408-16. Coates A, Abraham S, Kaye S et al. On the receiving end patient perception of the side-effects of cancer chemotherapy. Eur J Cancer Oncol 1983; 19: 203-8. Roila F, De Angelis V, Contu A et al. Ondansetron versus metoclopramide both combined with dexamethasone in the prevention of cisplatin-induced delayed emesis. Proc Am Soc Clin Oncol 1996; 15: 528 (Abstr 1705). Roila F, Boschetti E, Tonato M et al. Predictive factors of delayed emesis in cisplatin-treated patients and anti-emetic activity and tolerability of metoclopramide or dexamethasone. Am J Clin Oncol 1991; 14: 238-42.