Stratified, Randomized, Double-Blind Comparison of Intravenous Ondansetron Administered as a Multiple-Dose Regimen Versus Two Single-Dose Regimens in the Prevention of Cisplatin-Induced Nausea and Vomiting By Thomas M. Beck, Paul J. Hesketh, Stefan Madajewicz, Rudolph M. Navari, Kelly Pendergrass, Eric P. Lester, Julie A. Kish, William K. Murphy, John D. Hainsworth, David R. Gandara, Leslie J. Bricker, Alan M. Keller, Joanne Mortimer, Daniel V. Galvin, Karen W. House, and Judy C. Bryson Purpose: This study compares the efficacy and safety of two single-dose regimens with the approved threedose regimen of ondansetron in the prevention of cisplatin-induced emesis. Patients and Methods: This multicenter study was a stratified, randomized, double-blind, and parallel group design. Chemotherapy-naive inpatients were randomized to receive intravenous (IV) ondansetron (Zofran; Glaxo Inc, Research Triangle Park, NC) 0.15 mg/kg times three doses, every 4 hours or a single 8-mg or 32-mg dose followed by two saline doses that began 30 minutes before cisplatin administration. Cisplatin (high-dose -a 100 mg/m2 or medium-dose 50 to 70 mg/m 2 ) was given as a single infusion (s 3 hours). Patients were monitored for emetic episodes, adverse events, and laboratory safety parameters for 24 hours after cisplatin administration. Results: A total of 699 patients (359 high-dose, 340 medium-dose) were enrolled. Of these, 618 were assessable for efficacy (15 ineligible, 66 protocol deviations). The 32-mg dose was superior to the 8-mg single dose with regard to total number of emetic episodes (highdose, P = .015; medium-dose, P < .001), complete response (no emetic episodes: high-dose, 48% v 35%;
CISPLATIN
IS A widely used chemotherapeutic agent that has demonstrated activity in a number of malignancies; however, severe nausea and emesis are the most frequent side effects of this treatment.1 The amount of nausea and emesis experienced varies with the dose of cisplatin; however, all patients who receive > 100 mg/m 2 will vomit during the first 24 hours if they do not receive effective antiemetic therapy. 2 Failure to control nausea and emesis produces obvious deleterious effects on the patient's quality of life and increases the likelihood that patients may refuse further beneficial or potentially curative chemotherapy. Hence, improved control of nausea and emesis is of particular importance in patients who receive adjuvant and outpatient chemotherapy. Recent evidence suggests that serotonin and serotonin 5-HT 3 receptors play an important role in the mechanism by which chemotherapeutic agents produce nausea and emesis. 3 -5 Ondansetron is a selective 5-HT 3 receptor antagonist that can control the nausea and vomiting produced by cisplatin effectively in both laboratory animals6 and in patients who receive various doses
P = .048; medium-dose, 73% v 50%; P = .001) and failure rate (> 5 emetic episodes, withdrawn or rescued: high-dose, 20% v 34%; P = .018; medium-dose, 9% v 23%; P = .005). The 32-mg single dose was also superior to the 0.15 mg/kg times three dose regimen with regard to total number of emetic episodes (medium-dose, P = .033) and failure rate (high-dose, 20% v 36%; P = .009; medium-dose, 9% v 22%; P = .011). Ondansetron was well tolerated. The most common adverse event was headache. An approximate 10-fold increase in the incidence of clinically significant transaminase elevations was observed in the high-dose versus medium-dose cisplatin strata (asparate aminotransferase [AST], 6.5% v 0.7%; serum alanine aminotransferase [ALT], 5.0% v 0.3%). Conclusion: A 32-mg single dose of ondansetron is more effective than a single 8-mg dose and is at least as effective as the standard regimen of 0.15 mg/kg times three doses in the prevention of cisplatin-induced acute emesis. J Clin Oncol 10: 1969-1975. © 1992 by American Society of Clinical Oncology.
3 8 of this chemotherapeutic agent. ,7, Ondansetron has
been shown to be superior to high-dose metoclopramide 78, in the control of emesis in these patients. A number of clinical trials have shown that three
0.15-mg/kg doses of ondansetron provide complete control of emesis during the first 24 hours after high-
dose cisplatin administration (2 100 mg/im 2) in 40% to
From Mountain States Tumor Institute, Boise, ID; Boston University Medical Center, Boston, MA; State University of New York at Stony Brook, Stony Brook, NY; Simon-Williamson Clinic, Birmingham, AL; Research Medical Center, Kansas City; Washington University Hospital, St Louis, MO; University of Tennessee, Memphis, and The Vanderbilt Clinic, Nashville, TN; Wayne State University, and Henry Ford Hospital, Detroit,MI; MD Anderson Hospital, Houston, TX; University of California, Davis, and Veterans Administration Medical Center, Martinez, CA; St FrancisHospital, Tulsa, OK; Glaxo Inc, Research TrianglePark, NC. Submitted April 6, 1992; acceptedJuly 30, 1992. Supported by a grantfrom Glaxo Inc, Research TrianglePark, NC. Address reprintrequests to Thomas M. Beck, MD, Mountain States Tumor Institute, 151 E Bannock, Boise, ID 82712-6297. © 1992 by American Society of ClinicalOncology. 0732-183X/92/1012-0021$3.00/0
Journalof Clinical Oncology, Vol 10, No 12 (December), 1992: pp 1969-1975
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1969
1970
BECK ET AL
45% of patients. 7,9 A variety of dosing schedules have been tested and similar response rates were obtained with intermittent dosing at 2-, 4-, 6-, or 8-hour intervals. 9-11 Additionally, ondansetron administered as a 24-hour continuous infusion (total dose, 32 mg) has been shown to provide comparable, but not superior, response rates. 8 These observations, together with the fact that most patients who fail antiemetic therapy with metoclopramide do so within the first 6 to 8 hours,8' 12 have suggested that 5-HT3 antagonism may be most beneficial during the initial hours of treatment and that shorter treatment schedules might be just as effective in the control of cisplatin-induced emesis. The present study was undertaken to evaluate the efficacy of two fixed single-dose ondansetron regimens (8 and 32 mg) by comparing them with the approved dose and schedule (0.15 mg/kg every 4 hours times three) in patients receiving cisplatin chemotherapy. The 32-mg dose was chosen because this would be comparable to the dose that a person of average weight (70 kg) would receive if given 0.15 mg/kg times three doses of intravenous (IV) ondansetron. The 8-mg dose was chosen because preliminary data from European studies suggested that this dose may be effective in patients who have received lower doses of cisplatin (personal communication, Glaxo Group Research Ltd, July 1992). PATIENTS AND METHODS
Study Design This randomized, double-blind, parallel-group, multicenter study was stratified by cisplatin dose and was designed to enroll a total of 600 assessable chemotherapy-naive patients who were scheduled to receive cisplatin at a dose of either 50 to 70 mg/m2 or > 100 mg/m 2.
PatientEligibility Chemotherapy-naive cancer patients, aged 18 years or older, were eligible for the study if they had a Karnofsky performance status of at least 60%. All patients were hospital inpatients during treatment. Patients were excluded from the study if they had impaired renal function (serum creatinine concentration > 2 mg/dL or creatinine clearance < 50 mL/min), a serum alanine aminotransferase (ALT concentration of > twice the upper limit of normal or if they had vomited or retched 24 hours before the study). Patients could not have received any antiemetic medication during the 24 hours preceding the study period, radiation therapy to the abdominal or pelvic region within 48 hours before the study period, or any of the previously mentioned treatments during the study period. The protocol was reviewed and approved by the institutional review board at each institution and written informed consent was obtained from each patient.
Chemotherapy Patients received cisplatin at a dose of either 50 to 70 mg/m2 or > 100 mg/m 2 that was administered as a single IV infusion
during a period of 3 hours or less. Use of other chemotherapy agents was allowed concurrently, with the exception of cyclophosphamide (> 500 mg/m 2), nitrogen mustard (mechlorethamine), dacarbazine (DTIC), procarbazine, carmustine (BCNU), ifosfamide (> 1.5 g/m 2), or carboplatin. Antiemetic Treatment Patients were stratified according to their dose of cisplatin and then randomized (1:1:1) to receive either three 0.15-mg/kg doses, a single 8-mg dose, or a single 32-mg dose of IV ondansetron hydrochloride (Zofran; Glaxo Inc, Research Triangle Park, NC), according to a computer-generated scheme from the Department of Biostatistics at Glaxo Inc. Administration of the study drug was double-blinded. Patients assigned to receive three doses of ondansetron were given their first dose 30 minutes before the start of cisplatin and their second and third doses 4 and 8 hours, respectively, after the initial dose. Patients who were assigned to receive a single dose of ondansetron were given the dose of ondansetron 30 minutes before the start of cisplatin, followed by doses of saline administered 4 and 8 hours after the initial dose. Assessment ofAntiemetic Efficacy and Safety The primary efficacy parameter was the number of emetic episodes that occurred during the 24-hour study period. Patients were observed for the occurrence of emetic episodes during the 24 hours after initiation of cisplatin chemotherapy, and adverse events during the 24.5 hours after the administration of the first dose of the study drug. An emetic episode was defined as a single episode of vomiting (expulsion of stomach contents through the mouth), a single episode of retching (an attempt to vomit not productive of stomach contents), or any number of continuous vomits and/or retches. Emetic episodes by definition were separated by an at least 1-minute absence of both vomiting and retching. The number of emetic episodes was used to define the treatment response: complete response (CR), no emetic episodes; major response, one to two emetic episodes; minor response (MR), three to five emetic episodes; and failure, more than five emetic episodes, requirement of rescue antiemetic therapy, or withdrawal from the study. The severity of nausea was assessed by patients, independently from emesis, using a 100-mm visual analog scale (0, no nausea; 100, nausea as bad as it could be). Food intake (unless physically unable to ingest liquids and/or solids) was assessed using the following scale: full meals, light snacks, liquids only, or nothing by mouth. Both assessments were made immediately before the first dose of the study drug and again 24 hours after cisplatin initiation. The change in nausea was calculated by subtracting the baseline score from the posttreatment score. Blood samples, to monitor complete blood cell counts and biochemistry, were obtained 48 hours before the first dose of study drug and at the end of the 24-hour study period. Abnormal values considered related to ondansetron were followed up until they returned to normal or were otherwise explained. StatisticalMethods Treatment groups were compared with regard to the number of emetic episodes experienced during the 24-hour study period using the Wilcoxon rank-sum test. As in previous studies, for this analysis, patients who reported more than five emetic episodes or who were rescued or withdrawn for any reason were assigned the same arbitrarily high value (> five) for number of emetic episodes.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1971
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN Treatments were also compared with respect to the proportion of patients with a CR and the proportion of patients who were considered to have undergone unsuccessful treatment using the Mantel-Haenszel test. Additionally, treatment groups were compared with respect to time to first emetic episode (hours from the start of cisplatin infusion) using the Wilcoxon rank-sum test. Patients with no emetic episodes were assigned the same arbitrary time (> 24 hours). The Wilcoxon rank-sum test also was used to compare treatment groups with respect to severity of nausea. Treatment groups were compared with regard to food intake using the Mantel-Haenszel test.
RESULTS
Seven hundred three patients were enrolled onto this trial from 26 different centers, and 699 patients received active treatment. Three hundred fifty-nine patients received high-dose cisplatin (scheduled dose, > 100 mg/ m2; actual range, 94 to 200 mg/m 2), and 340 received medium-dose cisplatin (scheduled dose, 50 to 70 mg/m 2; actual range, 45 to 77 mg/m 2). Of the 699 patients randomized and treated with study medication, 15 (2%) were ineligible and 66 (9%) were unassessable for treatment efficacy because of protocol deviations. Table 1 lists the reasons for the exclusion of these patients. The most common reasons were receipt of an excluded medication (any agent with potential antiemetic effect) and ondansetron- and cisplatin-dosing errors. Exclusions were distributed evenly to the treatment groups and strata. The efficacy data presented are for the 618 assessable patients; however, an intent-to-treat analysis was also performed on all 699 patients who received active study drug. This resulted in a similar outcome (ie, all results that were statistically significant in the analysis presented below were also significant in the intent-toTable 1. Protocol Deviations Resulting in Patient Exclusion From
Efficacy Evaluation No. of Patients (N = 699)'
Deviation No. of ineligible patients (%) Received antiemetic within 24 hours before the study Vomited within 24 hours before the study
15 (2) 8 4
Previously received chemotherapy Abdominal radiation before or during study Deviations during study (%) Received excluded concomitant medication Ondansetron dosing error Cisplatin dosing error Withdrawn due to administrative error Received alternate antiemetic by mistake Blind broken by study staff Received cyclophosphamide dose > 550 mg/m Missing dosing and efficacy data
2 1
2
*Nineteen patients had > one reason for exclusion.
66 (9) 30 24 16 7 5 3 3 1
Table 2. Patient Characteristics Ondansetron Dose 0.15 mg/kg x 3 No. No. of patients 234 Sex Male 157 Female 77 Age, years Median 61 Range 20-87 Alcohol use* Nonuser or occasional use 162 Moderate use 19 Current or prior heavy use 53 Type of cancer Lung 107 Head and neck 41 Gastrointestinal 23 Genitourinary 20 Gynecologic 14 Bone and soft tissue 8 Other 21 Chemotherapy Cisplatin alone 47 Cisplatin plus other agents 187
32 mg x 1
8mgxl
%
No.
%
No.
%
100
245
100
220
100
67 33
160 85
65 35
148 72
67 33
62 21-82
62 22-82
69 8 23
160 31 52
65 13 21
158 16 46
72 7 21
46 18 10 9 6 3 9
117 45 34 16 14 2 17
48 18 14 7 6 1 7
108 41 13 21 12 4 21
49 19 6 10 5 2 10
20 80
55 190
22 78
57 163
26 74
NOTE. Nonuse and/or occasional use are defined as < seven drinks per week. Moderate use is defined as one to four drinks per day. Heavy use is defined as five or more drinks per day. *Data were unavailable for two patients in the 8 mg x I group.
treat population, and the P values obtained were very similar). Finally, because 26 institutions participated in this study, the data were also analyzed stratified by institution, using the Cochran-Mantel-Haenszel test, to control for variability amongst institutions. Once again the outcome was unchanged and all statistically significant results are comparable with those presented in later tables. PatientCharacteristics
Patient groups were similar with regard to age and sex distribution. Patient characteristics for both dose levels of cisplatin are listed in Table 2. Approximately two thirds of the patients in each treatment group were male; the ratio of male:female patients was slightly higher in the high-dose cisplatin stratum (70% male) compared with the medium-dose stratum (60% male). The percentage of patients with previous or current heavy alcohol use was also slightly higher in the highdose cisplatin stratum (26% v 17%). Lung cancer was the most common malignancy in both strata (50%), followed by head and neck cancer (18%) and gastrointes-
tinal malignancies (10%). Twenty-three percent (23%)
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1972
BECK ET AL
of patients received cisplatin alone, whereas 77% received cisplatin in combination with other chemotherapeutic agents. The agents most frequently used in combination were etoposide (32%) and fluorouracil (26%). Combination chemotherapy regimens were similar for both treatment groups.
32 mg x 1 v8 mg x 1 32 mg x 1 v0.15 mg/kg x 3 8mg x 1 vO.15 mg/kg x 3
CR
EE
FR
0.048* 0.315 0.349
0.015* 0.095 0.590
0.018* 0.009* 0.749
Abbreviation: FR, failure (> 5 EE, rescue, or withdrawal). *P < .05.
Antiemetic Efficacy High-dose cisplatin stratum (> 100 mg/m2). Efficacy data for the 317 assessable patients who received highdose cisplatin are listed in Table 3. The primary efficacy variables listed in Table 3 include CR with no emetic episodes, major response with one to two emetic episodes, MR with three to five emetic episodes, and failure with more than five emetic episodes, withdrawal, or rescue. Statistical test results are listed in Table 4. Patients who received 32 mg of ondansetron as a single dose experienced significantly fewer emetic episodes than those who received 8 mg (P = .015). Additionally, patients who received the 32-mg dose experienced a higher CR rate (no emetic episodes; P = .048) as well as a lower failure rate than those who received 8 mg (P = .018).
With regard to secondary efficacy variables, the time to the first emetic episode was significantly longer (P = .015) in the 32-mg versus 8-mg group of patients. The 32-mg single-dose group also experienced less nausea and a lower failure rate than those patients who received standard doses (0.15 mg/kg times three) of ondansetron (P = .036 and .009, respectively). In no instance among the efficacy criteria evaluated was a single 32-mg dose of ondansetron inferior to the standard 0.15-mg/kg three-dose regimen of ondansetron in those patients who received high-dose cisplatin. No Table 3. Antiemetic Efficacy in the High-Dose Cisplatin Stratum (> 100 mg/m2): Primary Efficacy Variables
No.
%
100
100
8 mgxl No.
%
Safety Assessments All 699 patients who received active study medication were included in the evaluation of safety. Safety results are shown in Table 7 for the two strata combined. The most common events reported were headache, diarrhea,
Ondansetron Dose
32mg xl No.
115 100 102
0.15mg/kgx3
%
100
41
41
40
35
49
48
14 5
14 5
16 9
14 8
14 11
14 11
4
4
11
10
8
8
36
36
39
34
20
20
Abbreviation: EE, emetic episode. *Number of patients assessable for antiemetic efficacy.
significant differences in food intake were observed between the three treatment groups. Medium-dose cisplatinstratum (50 to 70 mg/m 2). The antiemetic efficacy data for the 301 assessable patients who received medium doses of cisplatin are presented in Table 5. Statistical test results are summarized in Table 6. In this group of patients, the 32-mg single-dose of ondansetron showed even greater statistical superiority than in the group of patients who received higher doses of cisplatin. The 32-mg dose was significantly superior to the 8-mg single dose for all primary and secondary efficacy variables evaluated. Additionally, the 32-mg single dose was significantly superior to the 0.15-mg/kg times three dose regimen with regard to both total number of emetic episodes and the number of patients who underwent unsuccessful treatment (> five emetic episodes, rescue, or withdrawal). Once again, the 32-mg single dose was not inferior to the standard 0.15-mg/kg times three dose regimen in any respect. Food intake with the 32-mg dose was superior to that with both the standard regimen and the 8-mg single dose.
Table 5. Antiemetic Efficacy in the Medium-Dose Cisplatin Stratum (50 to 70 mg/m2): Primary Efficacy Variables
Ondansetron Dose 0.15 mg/kg x 3
No. of patients* Complete response 0EE Major response 1 EE 2 EE Minor response 3-5 EE Failure > 5 EE or withdrawn/rescued
Table 4. Antiemetic Efficacy in the High-Dose Cisplatin Stratum 2 (>100 mg/m ): Statistical Test Results
No. of patients* Complete response 0 EE Major response 1 EE 2 EE Minor response 3-5 EE Failure > 5 EE or withdrawn/rescued
8mgxl
32mgxl
No.
No.
%
93
100
No.
%
101
100
62
61
54
50 68
73
7 4
7 4
12 8
11 7
10 4
11 4
6
6
8
7
3
3
22
22
25
23
8
9
*Number of patients assessable for antiemetic efficacy.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
%
107 100
1973
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN Table 6. Antiemetic Efficacy in the Medium-Dose Cisplatin Stratum (50 to 70 mg/m2): Statistical Test Results
32 mg x 1 v8mg x 1 32 mg x 1 v 0.15 mg/kg x 3 8 mg x 1 v 0.015 mg/kg x 3
CR
EE
FR
0.001" 0.083 0.114
0.001" 0.033* 0.201
0.005* 0.011* 0.786
Abbreviation: CR, complete response. *P < .05.
and fever. The incidence of headaches was similar in each stratum; 21% of patients in the high-dose cisplatin stratum and 20% of patients in the medium-dose stratum experienced headaches. Within the medium-dose cisplatin stratum, significantly more patients who received the 32-mg single dose (26%) experienced headaches than those who received the 8-mg single dose (15%; P = .046). No other significant differences were observed in the incidence of headache between the treatment groups. Headaches were generally mild or moderate in nature and treatable with nonnarcotic analgesics. Five patients (two medium-dose stratum and three high-dose stratum) had severe headaches, which lasted from 1 to 30 hours. The incidence of diarrhea was greater in the high-dose cisplatin stratum (11%) than in the medium-dose cisplatin stratum (6%). Within the medium-dose stratum, significantly more patients who received the 0.15-mg/kg times three doses (9%) experienced diarrhea than those who received the 8-mg single dose (3%; P = .046). No other significant differences were observed in the incidence of diarrhea between the treatment groups. The incidence of fever was the same Table 7. Safety Results
DISCUSSION
The efficacy, safety, and simplicity of ondansetron as a single, effective antiemetic agent has been demonstrated in numerous clinical trials. 3,7,9,11 During the clinical development of ondansetron, a variety of multidose schedules were evaluated. The schedules studied were based on traditional methods of administering antiemetics, the pharmacokinetics of ondansetron (half-life of 3 to 4 hours), and what was believed to be the minimum effective plasma concentration necessary to block 5-HT3 receptors. The phase I and II studies evaluated three IV doses of ondansetron ranging from 0.01 mg/kg to 0.48 mg/kg per dose, with antiemetic activity observed even at the lowest dose tested. 13 ,14 These studies demon-
Ondansetron Dose 0.15 mg/kg x 3
in each stratum (9%), and no significant differences were observed between treatment groups. No significant differences were observed between the three treatment groups with respect to laboratory indices of safety, which included transaminase elevations. However, there was an approximate 10-fold increase in the incidence of clinically significant transaminase elevations between the two strata. In the high-dose cisplatin stratum, 6.5% of patients who had normal or belownormal baseline values of aspartate aminotransferase (AST) had increases to at least twice the upper limit of the normal range during the study compared with 0.7% of patients in the medium-dose cisplatin stratum. Additionally, 5% of patients in the high-dose cisplatin stratum who had normal or below normal baseline values of ALT had increases to at least twice the upper limit of the normal range during the study compared with 0.3% of patients in the medium-dose cisplatin stratum.
8 mg x 1
32 mg x 1
No.
%
No.
%
No.
%
No. of patients Most common events Headache Fever Diarrhea Transaminase elevations* High-dose cisplatin 2 (>100 mg/m )
234
100
245
100
220
100
43 26 25
18 11 11
44 19 16
18 8 7
55 16 18
25 7 8
ASTt ALTt Medium-dose cisplatin 2 (50-70 mg/m ) ASTt ALTt
8/106 4/105
8 4
6/112 5/110
5 7/105 5 7/102
7 7
1/97 1/100
1 1
1/98 0/99
1 0/91 0 0/91
0 0
*Transient increases to at least twice the upper limit of normal ranges in patients who had normal or below-normal baseline values. tSGOT. tSGPT.
strated that ondansetron had a wide therapeutic index with minimal toxicity. Efficacy performance was similar over the dose range of 0.10 mg/kg to 0.18 mg/kg. Although slightly greater antiemetic activity was observed at the higher doses tested, a traditional doseresponse relationship was not clearly established, and it seemed that no additional efficacy was obtained with doses higher than 0.18 mg/kg. Therefore, a midrange dose of 0.15 mg/kg (total daily dose, 32 mg in a 70-kg person) was selected for further clinical development. A variety of dosing schedules were then tested and similar response rates were obtained with intermittent dosing at 2-, 4-, 6-, or 8-hour intervals. 9 " With the dosing schedule of every 2 hours for three doses, the last dose of ondansetron was given only 3.5 hours after the initiation of treatment with cisplatin. That this regimen had similar response rates to the other regimens studied, including continuous infusion, raised questions with
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1974
BECK ET AL
regard to the need for repeated dosing during the 24 hours after cisplatin administration. The results of the current study demonstrate that the acute nausea and vomiting induced by cisplatin, in both high and medium doses, can be prevented effectively with a single, 32-mg dose of ondansetron. In this study, a single 32-mg IV dose was clearly superior to a single 8-mg dose, and was at least as effective as, if not superior to, three 0.15 mg/kg doses of ondansetron given every 4 hours. Unexpectedly, there was no difference between the standard schedule of 0.15 mg/kg every 4 hours for three doses and a single 8-mg IV dose. These data suggest that a sustained circulating plasma level of ondansetron may not be necessary to control emesis in the first 24 hours after administration of cisplatin. This leads us to speculate on what may be happening at the receptor level to explain the effectiveness of a single dose of ondansetron. The release of 5-HT from enterochromaffin cells is thought to be a crucial factor in the initiation and maintenance of the vomiting reflex. 3 Measurements of 5-hydroxyindole acetic acid in the urine 3 suggest a large amount of 5-HT is released during the first 2 to 6 hours after administration of cisplatin. This observation might explain why patients who received cisplatin without antiemetics experienced the most intense emesis during the first 6 hours after treatment. 15 Also, in most cases, lack of response to antiemetic therapy with metoclopramide was apparent within the first 6 to 8 hours, 7,8,12 in sharp contrast to patients who received ondansetron, for which median time to first emesis has consistently been longer than 20 hours. Thus although recent speculation on the mechanism of antiemetic action of metoclopramide has expanded from dopaminergic blockade to also include 5-HT 3 receptor blockade, these observations would suggest that the schedules and/or doses of metoclopramide used to date have been inadequate to achieve optimal 5-HT 3 receptor blockade. One may speculate that after a large initial release of 5-HT, the rate of release then decreases to one that corresponds to the synthesis of new 5-HT. In this case, provided that there is adequate blockade of the 5-HT 3 receptors during these first few hours after the emetic stimulus, a major or complete suppression of emesis may be achieved for up to 24 hours. Systemic exposure (area-under-the-curve [AUC]) of ondansetron during the first 4 hours after administration has been correlated positively with antiemetic efficacy. 16,17 The single 32-mg dose may offer greater protection against emesis than the other two dosing regimens
examined in this study because there is greater systemic exposure (ie, greater AUC) in the first 4 hours after the 32-mg single dose than the other two schedules, which is when the emetic stimulus is thought to be the greatest. The standard dosing regimen and a single 8-mg dose of ondansetron may have produced similar efficacy results because systemic exposure of ondansetron during the first 4 hours would be similar for both regimens. Clearly, this is an area that requires further investigation. Ondansetron was safe and well tolerated in all patients in this study. The most frequent adverse effect was headache, which is in line with previously reported studies. 8,9,12 Significantly more patients who received the 32-mg single dose in the medium-dose cisplatin stratum experienced headaches than those who received the 8 mg dose; however, no difference in the incidence of headache was observed in the high-dose stratum. The second most common adverse event reported was diarrhea, the incidence of which was greater in the high-dose cisplatin stratum. This is again consistent with earlier observations 9 and suggests that diarrhea is related to cisplatin administration. However, the greatest difference between the two cisplatin strata in terms of safety assessments was the increased incidence of clinically significant transaminase elevations in the high-dose cisplatin stratum. Transient transaminase elevations have been observed frequently in patients who received ondansetron and cisplatin. However, no ondansetron dose-related effect was observed in this study, nor was an ondansetron dose-related effect observed in a previous ondansetron dose comparison study9 in regard to either the proportion of patients with elevations or the extent of the increase. Furthermore, cisplatin previously has been shown to have a cumulative-dose effect on transaminase values.'8 This study seems to confirm a cisplatin dose-related effect on transaminase elevations. However, transaminase elevations are transient and produce no apparent clinical sequelae. The demonstration of efficacy and safety of ondansetron when given as a single prophylactic dose before an emetogenic stimulus has significant economic implications. Pharmacy and nursing costs can be reduced. Increased consideration can be given to the administration of highly emetogenic drugs on an outpatient basis. The results of this study establish that a single 32-mg prophylactic dose of ondansetron is more effective than a single 8-mg dose and at least as effective as intermittent dosing for antiemetic protection against emesis induced by moderately high to highly emetogenic chemo-
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1975
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN therapy regimens. This suggests that sustained plasma levels of ondansetron producing continuous blockade of 5-HT 3 receptors are not necessary to maximize efficacy.
We gratefully acknowledge the following investigators for patient recruitment: William Graydon Harker, MD, Walter H.
Harvey, MD, Gerald J. Kallas, MD, Basil S. Kasimis, MD, Ali Khojasteh, MD, Saeeda Kirmani, MD, Montague Lane, MD, Ronald D. Lawson, MD, Greg G. Monaghan, MD, George A. Omura, MD, Alan L. Rosenblum, MD, Tariq Siddiqui, MD, Martin R. York, MD. We also gratefully acknowledge the help of Susan E. Patrick, PhD and Susan G. Dickerson in conducting this trial and Elaine M. Anderson, PhD, in the preparation of the manuscript.
REFERENCES 1. Von Hoff D, Schilsky R, Reicher CM, et al: Toxic effects of cis-Dichlorodiammineplatinum (II) in man. Cancer Treat Rep 63:1527-1531, 1979 2. Gralla RJ, Itri LM, Pisko SE, et al: Antiemetic efficacy of high dose metoclopramide; randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. N Engl J Med 305:905-909, 1981 3. Cubeddu LX, Hoffmann IS, Fuenmayor NT, et al: Efficacy of ondansetron (GR38032F) and the role of serotonin in cisplatininduced nausea and vomiting. N Engl J Med 322:810-816, 1990 4. Fozard JR: 5-HT 3 receptors and cytotoxic drug-induced vomiting. Trends Pharmacol Sci 8:44-45, 1987 5. Stables R, Andrews PL, Bailey HE, et al: Antiemetic properties of the 5HT 3-receptor antagonist, GR38032F. Cancer Treat Rev 14:333-336, 1987 6. Costall B, Domeney AM, Gunning SJ, et al: GR38032F: A potent and novel inhibitor of cisplatin-induced emesis in the ferret. Br J Pharmacol 90:90, 1987 (suppl; abstr) 7. Hainsworth J, Harvey W, Pendergrass K, et al: A single-blind comparison of intravenous ondansetron, a selective serotonin antagonist, with intravenous metoclopramide in the prevention of nausea and vomiting associated with high-dose cisplatin chemotherapy. J Clin Oncol 9:721-728, 1991 8. Marty M, Pouillart P, Scholl S, et al: Comparison of the 5-hydroxytryptamine 3 (serotonin) antagonist ondansetron (GR38032F) with high-dose metoclopramide in the control of cisplatin-induced emesis. N Engl J Med 322:816-821, 1990 9. Lane M, Grunberg S, Lester EP, et al: A double-blind comparison of three dose levels of IV ondansetron (OND) in the prevention of cisplatin (DDP)-induced nausea and vomiting. Proc Am Soc Clin Oncol 9:329, 1990 (abstr)
10. Kris MG, Gralla RJ, Clark RA, et al: Phase II trials of the serotonin antagonist GR38032F for the control of vomiting caused by cisplatin. JNCI 81:42-46, 1989 11. Hesketh PJ, Murphy WK, Lester EP, et al: GR38032F (GR-C507/75): A novel compound effective in the prevention of acute cisplatin-induced emesis. J Clin Oncol 7:700-705, 1989 12. DeMulder PHM, Seynaeve C, Vermorken JB, et al: Ondansetron compared with high-dose metoclopramide in prophylaxis of acute and delayed cisplatin-induced nausea and vomiting. A multicenter, randomized, double-blind, crossover study. Ann Intern Med 113:834-840, 1990 13. Kris MG, Gralla RJ, Clark RA, et al: Dose-ranging evaluation of the serotonin antagonist GRC507/75 (GR38032F) when used as an antiemetic in patients receiving anticancer chemotherapy. J Clin Oncol 6:659-662, 1988 14. Grunberg SM, Stevenson LL, Russell CA et al: Dose ranging phase I study of the serotonin antagonist GR38032F for prevention of cisplatin-induced nausea and vomiting. J Clin Oncol 7:1137-1141 15. VonHoff DD, Schilsky R, Reichert CM, et al: Toxic Effects of cis-Dichlorodiammineplatinum (II) in Man. Cancer Treat Rep 63:1527-1531, 1979 16. Pritchard JF, Wells CD: Relationship between ondansetron systemic exposure and antiemetic efficacy and safety in cancer patients receiving cisplatin. Pharmacology (in press) 17. Grunberg SM, Groshens, Robinson DC et al: Correlation of anti-emetic efficacy and plasma levels of ondansetron. Eur J Cancer 26:879-882, 1990 18. Hesketh PJ, Twaddell T, Finn A: A possible role for cisplatin (DDP) in the transient hepatic enzyme elevations noted after ondansetron administration. Proc Am Soc Clin Oncol 9:323, 1990 (abstr)
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
CISPLATIN
IS A widely used chemotherapeutic agent that has demonstrated activity in a number of malignancies; however, severe nausea and emesis are the most frequent side effects of this treatment.1 The amount of nausea and emesis experienced varies with the dose of cisplatin; however, all patients who receive > 100 mg/m 2 will vomit during the first 24 hours if they do not receive effective antiemetic therapy. 2 Failure to control nausea and emesis produces obvious deleterious effects on the patient's quality of life and increases the likelihood that patients may refuse further beneficial or potentially curative chemotherapy. Hence, improved control of nausea and emesis is of particular importance in patients who receive adjuvant and outpatient chemotherapy. Recent evidence suggests that serotonin and serotonin 5-HT 3 receptors play an important role in the mechanism by which chemotherapeutic agents produce nausea and emesis. 3 -5 Ondansetron is a selective 5-HT 3 receptor antagonist that can control the nausea and vomiting produced by cisplatin effectively in both laboratory animals6 and in patients who receive various doses
P = .048; medium-dose, 73% v 50%; P = .001) and failure rate (> 5 emetic episodes, withdrawn or rescued: high-dose, 20% v 34%; P = .018; medium-dose, 9% v 23%; P = .005). The 32-mg single dose was also superior to the 0.15 mg/kg times three dose regimen with regard to total number of emetic episodes (medium-dose, P = .033) and failure rate (high-dose, 20% v 36%; P = .009; medium-dose, 9% v 22%; P = .011). Ondansetron was well tolerated. The most common adverse event was headache. An approximate 10-fold increase in the incidence of clinically significant transaminase elevations was observed in the high-dose versus medium-dose cisplatin strata (asparate aminotransferase [AST], 6.5% v 0.7%; serum alanine aminotransferase [ALT], 5.0% v 0.3%). Conclusion: A 32-mg single dose of ondansetron is more effective than a single 8-mg dose and is at least as effective as the standard regimen of 0.15 mg/kg times three doses in the prevention of cisplatin-induced acute emesis. J Clin Oncol 10: 1969-1975. © 1992 by American Society of Clinical Oncology.
3 8 of this chemotherapeutic agent. ,7, Ondansetron has
been shown to be superior to high-dose metoclopramide 78, in the control of emesis in these patients. A number of clinical trials have shown that three
0.15-mg/kg doses of ondansetron provide complete control of emesis during the first 24 hours after high-
dose cisplatin administration (2 100 mg/im 2) in 40% to
From Mountain States Tumor Institute, Boise, ID; Boston University Medical Center, Boston, MA; State University of New York at Stony Brook, Stony Brook, NY; Simon-Williamson Clinic, Birmingham, AL; Research Medical Center, Kansas City; Washington University Hospital, St Louis, MO; University of Tennessee, Memphis, and The Vanderbilt Clinic, Nashville, TN; Wayne State University, and Henry Ford Hospital, Detroit,MI; MD Anderson Hospital, Houston, TX; University of California, Davis, and Veterans Administration Medical Center, Martinez, CA; St FrancisHospital, Tulsa, OK; Glaxo Inc, Research TrianglePark, NC. Submitted April 6, 1992; acceptedJuly 30, 1992. Supported by a grantfrom Glaxo Inc, Research TrianglePark, NC. Address reprintrequests to Thomas M. Beck, MD, Mountain States Tumor Institute, 151 E Bannock, Boise, ID 82712-6297. © 1992 by American Society of ClinicalOncology. 0732-183X/92/1012-0021$3.00/0
Journalof Clinical Oncology, Vol 10, No 12 (December), 1992: pp 1969-1975
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1969
1970
BECK ET AL
45% of patients. 7,9 A variety of dosing schedules have been tested and similar response rates were obtained with intermittent dosing at 2-, 4-, 6-, or 8-hour intervals. 9-11 Additionally, ondansetron administered as a 24-hour continuous infusion (total dose, 32 mg) has been shown to provide comparable, but not superior, response rates. 8 These observations, together with the fact that most patients who fail antiemetic therapy with metoclopramide do so within the first 6 to 8 hours,8' 12 have suggested that 5-HT3 antagonism may be most beneficial during the initial hours of treatment and that shorter treatment schedules might be just as effective in the control of cisplatin-induced emesis. The present study was undertaken to evaluate the efficacy of two fixed single-dose ondansetron regimens (8 and 32 mg) by comparing them with the approved dose and schedule (0.15 mg/kg every 4 hours times three) in patients receiving cisplatin chemotherapy. The 32-mg dose was chosen because this would be comparable to the dose that a person of average weight (70 kg) would receive if given 0.15 mg/kg times three doses of intravenous (IV) ondansetron. The 8-mg dose was chosen because preliminary data from European studies suggested that this dose may be effective in patients who have received lower doses of cisplatin (personal communication, Glaxo Group Research Ltd, July 1992). PATIENTS AND METHODS
Study Design This randomized, double-blind, parallel-group, multicenter study was stratified by cisplatin dose and was designed to enroll a total of 600 assessable chemotherapy-naive patients who were scheduled to receive cisplatin at a dose of either 50 to 70 mg/m2 or > 100 mg/m 2.
PatientEligibility Chemotherapy-naive cancer patients, aged 18 years or older, were eligible for the study if they had a Karnofsky performance status of at least 60%. All patients were hospital inpatients during treatment. Patients were excluded from the study if they had impaired renal function (serum creatinine concentration > 2 mg/dL or creatinine clearance < 50 mL/min), a serum alanine aminotransferase (ALT concentration of > twice the upper limit of normal or if they had vomited or retched 24 hours before the study). Patients could not have received any antiemetic medication during the 24 hours preceding the study period, radiation therapy to the abdominal or pelvic region within 48 hours before the study period, or any of the previously mentioned treatments during the study period. The protocol was reviewed and approved by the institutional review board at each institution and written informed consent was obtained from each patient.
Chemotherapy Patients received cisplatin at a dose of either 50 to 70 mg/m2 or > 100 mg/m 2 that was administered as a single IV infusion
during a period of 3 hours or less. Use of other chemotherapy agents was allowed concurrently, with the exception of cyclophosphamide (> 500 mg/m 2), nitrogen mustard (mechlorethamine), dacarbazine (DTIC), procarbazine, carmustine (BCNU), ifosfamide (> 1.5 g/m 2), or carboplatin. Antiemetic Treatment Patients were stratified according to their dose of cisplatin and then randomized (1:1:1) to receive either three 0.15-mg/kg doses, a single 8-mg dose, or a single 32-mg dose of IV ondansetron hydrochloride (Zofran; Glaxo Inc, Research Triangle Park, NC), according to a computer-generated scheme from the Department of Biostatistics at Glaxo Inc. Administration of the study drug was double-blinded. Patients assigned to receive three doses of ondansetron were given their first dose 30 minutes before the start of cisplatin and their second and third doses 4 and 8 hours, respectively, after the initial dose. Patients who were assigned to receive a single dose of ondansetron were given the dose of ondansetron 30 minutes before the start of cisplatin, followed by doses of saline administered 4 and 8 hours after the initial dose. Assessment ofAntiemetic Efficacy and Safety The primary efficacy parameter was the number of emetic episodes that occurred during the 24-hour study period. Patients were observed for the occurrence of emetic episodes during the 24 hours after initiation of cisplatin chemotherapy, and adverse events during the 24.5 hours after the administration of the first dose of the study drug. An emetic episode was defined as a single episode of vomiting (expulsion of stomach contents through the mouth), a single episode of retching (an attempt to vomit not productive of stomach contents), or any number of continuous vomits and/or retches. Emetic episodes by definition were separated by an at least 1-minute absence of both vomiting and retching. The number of emetic episodes was used to define the treatment response: complete response (CR), no emetic episodes; major response, one to two emetic episodes; minor response (MR), three to five emetic episodes; and failure, more than five emetic episodes, requirement of rescue antiemetic therapy, or withdrawal from the study. The severity of nausea was assessed by patients, independently from emesis, using a 100-mm visual analog scale (0, no nausea; 100, nausea as bad as it could be). Food intake (unless physically unable to ingest liquids and/or solids) was assessed using the following scale: full meals, light snacks, liquids only, or nothing by mouth. Both assessments were made immediately before the first dose of the study drug and again 24 hours after cisplatin initiation. The change in nausea was calculated by subtracting the baseline score from the posttreatment score. Blood samples, to monitor complete blood cell counts and biochemistry, were obtained 48 hours before the first dose of study drug and at the end of the 24-hour study period. Abnormal values considered related to ondansetron were followed up until they returned to normal or were otherwise explained. StatisticalMethods Treatment groups were compared with regard to the number of emetic episodes experienced during the 24-hour study period using the Wilcoxon rank-sum test. As in previous studies, for this analysis, patients who reported more than five emetic episodes or who were rescued or withdrawn for any reason were assigned the same arbitrarily high value (> five) for number of emetic episodes.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1971
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN Treatments were also compared with respect to the proportion of patients with a CR and the proportion of patients who were considered to have undergone unsuccessful treatment using the Mantel-Haenszel test. Additionally, treatment groups were compared with respect to time to first emetic episode (hours from the start of cisplatin infusion) using the Wilcoxon rank-sum test. Patients with no emetic episodes were assigned the same arbitrary time (> 24 hours). The Wilcoxon rank-sum test also was used to compare treatment groups with respect to severity of nausea. Treatment groups were compared with regard to food intake using the Mantel-Haenszel test.
RESULTS
Seven hundred three patients were enrolled onto this trial from 26 different centers, and 699 patients received active treatment. Three hundred fifty-nine patients received high-dose cisplatin (scheduled dose, > 100 mg/ m2; actual range, 94 to 200 mg/m 2), and 340 received medium-dose cisplatin (scheduled dose, 50 to 70 mg/m 2; actual range, 45 to 77 mg/m 2). Of the 699 patients randomized and treated with study medication, 15 (2%) were ineligible and 66 (9%) were unassessable for treatment efficacy because of protocol deviations. Table 1 lists the reasons for the exclusion of these patients. The most common reasons were receipt of an excluded medication (any agent with potential antiemetic effect) and ondansetron- and cisplatin-dosing errors. Exclusions were distributed evenly to the treatment groups and strata. The efficacy data presented are for the 618 assessable patients; however, an intent-to-treat analysis was also performed on all 699 patients who received active study drug. This resulted in a similar outcome (ie, all results that were statistically significant in the analysis presented below were also significant in the intent-toTable 1. Protocol Deviations Resulting in Patient Exclusion From
Efficacy Evaluation No. of Patients (N = 699)'
Deviation No. of ineligible patients (%) Received antiemetic within 24 hours before the study Vomited within 24 hours before the study
15 (2) 8 4
Previously received chemotherapy Abdominal radiation before or during study Deviations during study (%) Received excluded concomitant medication Ondansetron dosing error Cisplatin dosing error Withdrawn due to administrative error Received alternate antiemetic by mistake Blind broken by study staff Received cyclophosphamide dose > 550 mg/m Missing dosing and efficacy data
2 1
2
*Nineteen patients had > one reason for exclusion.
66 (9) 30 24 16 7 5 3 3 1
Table 2. Patient Characteristics Ondansetron Dose 0.15 mg/kg x 3 No. No. of patients 234 Sex Male 157 Female 77 Age, years Median 61 Range 20-87 Alcohol use* Nonuser or occasional use 162 Moderate use 19 Current or prior heavy use 53 Type of cancer Lung 107 Head and neck 41 Gastrointestinal 23 Genitourinary 20 Gynecologic 14 Bone and soft tissue 8 Other 21 Chemotherapy Cisplatin alone 47 Cisplatin plus other agents 187
32 mg x 1
8mgxl
%
No.
%
No.
%
100
245
100
220
100
67 33
160 85
65 35
148 72
67 33
62 21-82
62 22-82
69 8 23
160 31 52
65 13 21
158 16 46
72 7 21
46 18 10 9 6 3 9
117 45 34 16 14 2 17
48 18 14 7 6 1 7
108 41 13 21 12 4 21
49 19 6 10 5 2 10
20 80
55 190
22 78
57 163
26 74
NOTE. Nonuse and/or occasional use are defined as < seven drinks per week. Moderate use is defined as one to four drinks per day. Heavy use is defined as five or more drinks per day. *Data were unavailable for two patients in the 8 mg x I group.
treat population, and the P values obtained were very similar). Finally, because 26 institutions participated in this study, the data were also analyzed stratified by institution, using the Cochran-Mantel-Haenszel test, to control for variability amongst institutions. Once again the outcome was unchanged and all statistically significant results are comparable with those presented in later tables. PatientCharacteristics
Patient groups were similar with regard to age and sex distribution. Patient characteristics for both dose levels of cisplatin are listed in Table 2. Approximately two thirds of the patients in each treatment group were male; the ratio of male:female patients was slightly higher in the high-dose cisplatin stratum (70% male) compared with the medium-dose stratum (60% male). The percentage of patients with previous or current heavy alcohol use was also slightly higher in the highdose cisplatin stratum (26% v 17%). Lung cancer was the most common malignancy in both strata (50%), followed by head and neck cancer (18%) and gastrointes-
tinal malignancies (10%). Twenty-three percent (23%)
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1972
BECK ET AL
of patients received cisplatin alone, whereas 77% received cisplatin in combination with other chemotherapeutic agents. The agents most frequently used in combination were etoposide (32%) and fluorouracil (26%). Combination chemotherapy regimens were similar for both treatment groups.
32 mg x 1 v8 mg x 1 32 mg x 1 v0.15 mg/kg x 3 8mg x 1 vO.15 mg/kg x 3
CR
EE
FR
0.048* 0.315 0.349
0.015* 0.095 0.590
0.018* 0.009* 0.749
Abbreviation: FR, failure (> 5 EE, rescue, or withdrawal). *P < .05.
Antiemetic Efficacy High-dose cisplatin stratum (> 100 mg/m2). Efficacy data for the 317 assessable patients who received highdose cisplatin are listed in Table 3. The primary efficacy variables listed in Table 3 include CR with no emetic episodes, major response with one to two emetic episodes, MR with three to five emetic episodes, and failure with more than five emetic episodes, withdrawal, or rescue. Statistical test results are listed in Table 4. Patients who received 32 mg of ondansetron as a single dose experienced significantly fewer emetic episodes than those who received 8 mg (P = .015). Additionally, patients who received the 32-mg dose experienced a higher CR rate (no emetic episodes; P = .048) as well as a lower failure rate than those who received 8 mg (P = .018).
With regard to secondary efficacy variables, the time to the first emetic episode was significantly longer (P = .015) in the 32-mg versus 8-mg group of patients. The 32-mg single-dose group also experienced less nausea and a lower failure rate than those patients who received standard doses (0.15 mg/kg times three) of ondansetron (P = .036 and .009, respectively). In no instance among the efficacy criteria evaluated was a single 32-mg dose of ondansetron inferior to the standard 0.15-mg/kg three-dose regimen of ondansetron in those patients who received high-dose cisplatin. No Table 3. Antiemetic Efficacy in the High-Dose Cisplatin Stratum (> 100 mg/m2): Primary Efficacy Variables
No.
%
100
100
8 mgxl No.
%
Safety Assessments All 699 patients who received active study medication were included in the evaluation of safety. Safety results are shown in Table 7 for the two strata combined. The most common events reported were headache, diarrhea,
Ondansetron Dose
32mg xl No.
115 100 102
0.15mg/kgx3
%
100
41
41
40
35
49
48
14 5
14 5
16 9
14 8
14 11
14 11
4
4
11
10
8
8
36
36
39
34
20
20
Abbreviation: EE, emetic episode. *Number of patients assessable for antiemetic efficacy.
significant differences in food intake were observed between the three treatment groups. Medium-dose cisplatinstratum (50 to 70 mg/m 2). The antiemetic efficacy data for the 301 assessable patients who received medium doses of cisplatin are presented in Table 5. Statistical test results are summarized in Table 6. In this group of patients, the 32-mg single-dose of ondansetron showed even greater statistical superiority than in the group of patients who received higher doses of cisplatin. The 32-mg dose was significantly superior to the 8-mg single dose for all primary and secondary efficacy variables evaluated. Additionally, the 32-mg single dose was significantly superior to the 0.15-mg/kg times three dose regimen with regard to both total number of emetic episodes and the number of patients who underwent unsuccessful treatment (> five emetic episodes, rescue, or withdrawal). Once again, the 32-mg single dose was not inferior to the standard 0.15-mg/kg times three dose regimen in any respect. Food intake with the 32-mg dose was superior to that with both the standard regimen and the 8-mg single dose.
Table 5. Antiemetic Efficacy in the Medium-Dose Cisplatin Stratum (50 to 70 mg/m2): Primary Efficacy Variables
Ondansetron Dose 0.15 mg/kg x 3
No. of patients* Complete response 0EE Major response 1 EE 2 EE Minor response 3-5 EE Failure > 5 EE or withdrawn/rescued
Table 4. Antiemetic Efficacy in the High-Dose Cisplatin Stratum 2 (>100 mg/m ): Statistical Test Results
No. of patients* Complete response 0 EE Major response 1 EE 2 EE Minor response 3-5 EE Failure > 5 EE or withdrawn/rescued
8mgxl
32mgxl
No.
No.
%
93
100
No.
%
101
100
62
61
54
50 68
73
7 4
7 4
12 8
11 7
10 4
11 4
6
6
8
7
3
3
22
22
25
23
8
9
*Number of patients assessable for antiemetic efficacy.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
%
107 100
1973
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN Table 6. Antiemetic Efficacy in the Medium-Dose Cisplatin Stratum (50 to 70 mg/m2): Statistical Test Results
32 mg x 1 v8mg x 1 32 mg x 1 v 0.15 mg/kg x 3 8 mg x 1 v 0.015 mg/kg x 3
CR
EE
FR
0.001" 0.083 0.114
0.001" 0.033* 0.201
0.005* 0.011* 0.786
Abbreviation: CR, complete response. *P < .05.
and fever. The incidence of headaches was similar in each stratum; 21% of patients in the high-dose cisplatin stratum and 20% of patients in the medium-dose stratum experienced headaches. Within the medium-dose cisplatin stratum, significantly more patients who received the 32-mg single dose (26%) experienced headaches than those who received the 8-mg single dose (15%; P = .046). No other significant differences were observed in the incidence of headache between the treatment groups. Headaches were generally mild or moderate in nature and treatable with nonnarcotic analgesics. Five patients (two medium-dose stratum and three high-dose stratum) had severe headaches, which lasted from 1 to 30 hours. The incidence of diarrhea was greater in the high-dose cisplatin stratum (11%) than in the medium-dose cisplatin stratum (6%). Within the medium-dose stratum, significantly more patients who received the 0.15-mg/kg times three doses (9%) experienced diarrhea than those who received the 8-mg single dose (3%; P = .046). No other significant differences were observed in the incidence of diarrhea between the treatment groups. The incidence of fever was the same Table 7. Safety Results
DISCUSSION
The efficacy, safety, and simplicity of ondansetron as a single, effective antiemetic agent has been demonstrated in numerous clinical trials. 3,7,9,11 During the clinical development of ondansetron, a variety of multidose schedules were evaluated. The schedules studied were based on traditional methods of administering antiemetics, the pharmacokinetics of ondansetron (half-life of 3 to 4 hours), and what was believed to be the minimum effective plasma concentration necessary to block 5-HT3 receptors. The phase I and II studies evaluated three IV doses of ondansetron ranging from 0.01 mg/kg to 0.48 mg/kg per dose, with antiemetic activity observed even at the lowest dose tested. 13 ,14 These studies demon-
Ondansetron Dose 0.15 mg/kg x 3
in each stratum (9%), and no significant differences were observed between treatment groups. No significant differences were observed between the three treatment groups with respect to laboratory indices of safety, which included transaminase elevations. However, there was an approximate 10-fold increase in the incidence of clinically significant transaminase elevations between the two strata. In the high-dose cisplatin stratum, 6.5% of patients who had normal or belownormal baseline values of aspartate aminotransferase (AST) had increases to at least twice the upper limit of the normal range during the study compared with 0.7% of patients in the medium-dose cisplatin stratum. Additionally, 5% of patients in the high-dose cisplatin stratum who had normal or below normal baseline values of ALT had increases to at least twice the upper limit of the normal range during the study compared with 0.3% of patients in the medium-dose cisplatin stratum.
8 mg x 1
32 mg x 1
No.
%
No.
%
No.
%
No. of patients Most common events Headache Fever Diarrhea Transaminase elevations* High-dose cisplatin 2 (>100 mg/m )
234
100
245
100
220
100
43 26 25
18 11 11
44 19 16
18 8 7
55 16 18
25 7 8
ASTt ALTt Medium-dose cisplatin 2 (50-70 mg/m ) ASTt ALTt
8/106 4/105
8 4
6/112 5/110
5 7/105 5 7/102
7 7
1/97 1/100
1 1
1/98 0/99
1 0/91 0 0/91
0 0
*Transient increases to at least twice the upper limit of normal ranges in patients who had normal or below-normal baseline values. tSGOT. tSGPT.
strated that ondansetron had a wide therapeutic index with minimal toxicity. Efficacy performance was similar over the dose range of 0.10 mg/kg to 0.18 mg/kg. Although slightly greater antiemetic activity was observed at the higher doses tested, a traditional doseresponse relationship was not clearly established, and it seemed that no additional efficacy was obtained with doses higher than 0.18 mg/kg. Therefore, a midrange dose of 0.15 mg/kg (total daily dose, 32 mg in a 70-kg person) was selected for further clinical development. A variety of dosing schedules were then tested and similar response rates were obtained with intermittent dosing at 2-, 4-, 6-, or 8-hour intervals. 9 " With the dosing schedule of every 2 hours for three doses, the last dose of ondansetron was given only 3.5 hours after the initiation of treatment with cisplatin. That this regimen had similar response rates to the other regimens studied, including continuous infusion, raised questions with
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1974
BECK ET AL
regard to the need for repeated dosing during the 24 hours after cisplatin administration. The results of the current study demonstrate that the acute nausea and vomiting induced by cisplatin, in both high and medium doses, can be prevented effectively with a single, 32-mg dose of ondansetron. In this study, a single 32-mg IV dose was clearly superior to a single 8-mg dose, and was at least as effective as, if not superior to, three 0.15 mg/kg doses of ondansetron given every 4 hours. Unexpectedly, there was no difference between the standard schedule of 0.15 mg/kg every 4 hours for three doses and a single 8-mg IV dose. These data suggest that a sustained circulating plasma level of ondansetron may not be necessary to control emesis in the first 24 hours after administration of cisplatin. This leads us to speculate on what may be happening at the receptor level to explain the effectiveness of a single dose of ondansetron. The release of 5-HT from enterochromaffin cells is thought to be a crucial factor in the initiation and maintenance of the vomiting reflex. 3 Measurements of 5-hydroxyindole acetic acid in the urine 3 suggest a large amount of 5-HT is released during the first 2 to 6 hours after administration of cisplatin. This observation might explain why patients who received cisplatin without antiemetics experienced the most intense emesis during the first 6 hours after treatment. 15 Also, in most cases, lack of response to antiemetic therapy with metoclopramide was apparent within the first 6 to 8 hours, 7,8,12 in sharp contrast to patients who received ondansetron, for which median time to first emesis has consistently been longer than 20 hours. Thus although recent speculation on the mechanism of antiemetic action of metoclopramide has expanded from dopaminergic blockade to also include 5-HT 3 receptor blockade, these observations would suggest that the schedules and/or doses of metoclopramide used to date have been inadequate to achieve optimal 5-HT 3 receptor blockade. One may speculate that after a large initial release of 5-HT, the rate of release then decreases to one that corresponds to the synthesis of new 5-HT. In this case, provided that there is adequate blockade of the 5-HT 3 receptors during these first few hours after the emetic stimulus, a major or complete suppression of emesis may be achieved for up to 24 hours. Systemic exposure (area-under-the-curve [AUC]) of ondansetron during the first 4 hours after administration has been correlated positively with antiemetic efficacy. 16,17 The single 32-mg dose may offer greater protection against emesis than the other two dosing regimens
examined in this study because there is greater systemic exposure (ie, greater AUC) in the first 4 hours after the 32-mg single dose than the other two schedules, which is when the emetic stimulus is thought to be the greatest. The standard dosing regimen and a single 8-mg dose of ondansetron may have produced similar efficacy results because systemic exposure of ondansetron during the first 4 hours would be similar for both regimens. Clearly, this is an area that requires further investigation. Ondansetron was safe and well tolerated in all patients in this study. The most frequent adverse effect was headache, which is in line with previously reported studies. 8,9,12 Significantly more patients who received the 32-mg single dose in the medium-dose cisplatin stratum experienced headaches than those who received the 8 mg dose; however, no difference in the incidence of headache was observed in the high-dose stratum. The second most common adverse event reported was diarrhea, the incidence of which was greater in the high-dose cisplatin stratum. This is again consistent with earlier observations 9 and suggests that diarrhea is related to cisplatin administration. However, the greatest difference between the two cisplatin strata in terms of safety assessments was the increased incidence of clinically significant transaminase elevations in the high-dose cisplatin stratum. Transient transaminase elevations have been observed frequently in patients who received ondansetron and cisplatin. However, no ondansetron dose-related effect was observed in this study, nor was an ondansetron dose-related effect observed in a previous ondansetron dose comparison study9 in regard to either the proportion of patients with elevations or the extent of the increase. Furthermore, cisplatin previously has been shown to have a cumulative-dose effect on transaminase values.'8 This study seems to confirm a cisplatin dose-related effect on transaminase elevations. However, transaminase elevations are transient and produce no apparent clinical sequelae. The demonstration of efficacy and safety of ondansetron when given as a single prophylactic dose before an emetogenic stimulus has significant economic implications. Pharmacy and nursing costs can be reduced. Increased consideration can be given to the administration of highly emetogenic drugs on an outpatient basis. The results of this study establish that a single 32-mg prophylactic dose of ondansetron is more effective than a single 8-mg dose and at least as effective as intermittent dosing for antiemetic protection against emesis induced by moderately high to highly emetogenic chemo-
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1975
SINGLE-DOSE IV ONDANSETRON FOR CISPLATIN therapy regimens. This suggests that sustained plasma levels of ondansetron producing continuous blockade of 5-HT 3 receptors are not necessary to maximize efficacy.
We gratefully acknowledge the following investigators for patient recruitment: William Graydon Harker, MD, Walter H.
Harvey, MD, Gerald J. Kallas, MD, Basil S. Kasimis, MD, Ali Khojasteh, MD, Saeeda Kirmani, MD, Montague Lane, MD, Ronald D. Lawson, MD, Greg G. Monaghan, MD, George A. Omura, MD, Alan L. Rosenblum, MD, Tariq Siddiqui, MD, Martin R. York, MD. We also gratefully acknowledge the help of Susan E. Patrick, PhD and Susan G. Dickerson in conducting this trial and Elaine M. Anderson, PhD, in the preparation of the manuscript.
REFERENCES 1. Von Hoff D, Schilsky R, Reicher CM, et al: Toxic effects of cis-Dichlorodiammineplatinum (II) in man. Cancer Treat Rep 63:1527-1531, 1979 2. Gralla RJ, Itri LM, Pisko SE, et al: Antiemetic efficacy of high dose metoclopramide; randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. N Engl J Med 305:905-909, 1981 3. Cubeddu LX, Hoffmann IS, Fuenmayor NT, et al: Efficacy of ondansetron (GR38032F) and the role of serotonin in cisplatininduced nausea and vomiting. N Engl J Med 322:810-816, 1990 4. Fozard JR: 5-HT 3 receptors and cytotoxic drug-induced vomiting. Trends Pharmacol Sci 8:44-45, 1987 5. Stables R, Andrews PL, Bailey HE, et al: Antiemetic properties of the 5HT 3-receptor antagonist, GR38032F. Cancer Treat Rev 14:333-336, 1987 6. Costall B, Domeney AM, Gunning SJ, et al: GR38032F: A potent and novel inhibitor of cisplatin-induced emesis in the ferret. Br J Pharmacol 90:90, 1987 (suppl; abstr) 7. Hainsworth J, Harvey W, Pendergrass K, et al: A single-blind comparison of intravenous ondansetron, a selective serotonin antagonist, with intravenous metoclopramide in the prevention of nausea and vomiting associated with high-dose cisplatin chemotherapy. J Clin Oncol 9:721-728, 1991 8. Marty M, Pouillart P, Scholl S, et al: Comparison of the 5-hydroxytryptamine 3 (serotonin) antagonist ondansetron (GR38032F) with high-dose metoclopramide in the control of cisplatin-induced emesis. N Engl J Med 322:816-821, 1990 9. Lane M, Grunberg S, Lester EP, et al: A double-blind comparison of three dose levels of IV ondansetron (OND) in the prevention of cisplatin (DDP)-induced nausea and vomiting. Proc Am Soc Clin Oncol 9:329, 1990 (abstr)
10. Kris MG, Gralla RJ, Clark RA, et al: Phase II trials of the serotonin antagonist GR38032F for the control of vomiting caused by cisplatin. JNCI 81:42-46, 1989 11. Hesketh PJ, Murphy WK, Lester EP, et al: GR38032F (GR-C507/75): A novel compound effective in the prevention of acute cisplatin-induced emesis. J Clin Oncol 7:700-705, 1989 12. DeMulder PHM, Seynaeve C, Vermorken JB, et al: Ondansetron compared with high-dose metoclopramide in prophylaxis of acute and delayed cisplatin-induced nausea and vomiting. A multicenter, randomized, double-blind, crossover study. Ann Intern Med 113:834-840, 1990 13. Kris MG, Gralla RJ, Clark RA, et al: Dose-ranging evaluation of the serotonin antagonist GRC507/75 (GR38032F) when used as an antiemetic in patients receiving anticancer chemotherapy. J Clin Oncol 6:659-662, 1988 14. Grunberg SM, Stevenson LL, Russell CA et al: Dose ranging phase I study of the serotonin antagonist GR38032F for prevention of cisplatin-induced nausea and vomiting. J Clin Oncol 7:1137-1141 15. VonHoff DD, Schilsky R, Reichert CM, et al: Toxic Effects of cis-Dichlorodiammineplatinum (II) in Man. Cancer Treat Rep 63:1527-1531, 1979 16. Pritchard JF, Wells CD: Relationship between ondansetron systemic exposure and antiemetic efficacy and safety in cancer patients receiving cisplatin. Pharmacology (in press) 17. Grunberg SM, Groshens, Robinson DC et al: Correlation of anti-emetic efficacy and plasma levels of ondansetron. Eur J Cancer 26:879-882, 1990 18. Hesketh PJ, Twaddell T, Finn A: A possible role for cisplatin (DDP) in the transient hepatic enzyme elevations noted after ondansetron administration. Proc Am Soc Clin Oncol 9:323, 1990 (abstr)
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 22, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
