Eligible criteria included histologically or cytologically confirmed malignant diseases, age ≥18 years, either chemotherapy naive or non-naive, no history of nausea or emesis from chemotherapy with other agents, and a Karnofsky performance status ≥50%. Female patients who were not sterile or postmenopausal were required to use a prescribed form of birth control and achieve a negative pregnancy test at a prestudy visit. Patients were to receive cisplatin-based chemotherapy (cisplatin ≥60 mg/m²); however, patients were excluded if they were to receive carboplatin (>1.0 g/m²), cyclophosphamide (>1.0 g/m²), nitrogen mustards (>1.5 g/m²), dacarbazine (>1.5 g/m²), or ifosphamide (>1.5 g/m²) during the 24 hours after the cisplatin infusion. Exclusion criteria included preexisting nausea and/or vomiting from brain metastasis or gastrointestinal obstruction, symptoms of hepatic failure, congestive heart failure, cardiomyopathy, first-degree heart block, preexisting complete bundle branch block, or the use of anti-arrhythmic medication. Other exclusion criteria were the evidence of a seizure disorder requiring anticonvulsants and vomiting or Southwest Oncology Group (SWOG) Common Toxicity Criteria grade 2 or 3 nausea in the 24 hours preceding chemotherapy. Non-narcotic analgesics were always allowed to be given to the patients, but narcotic analgesics were allowed to be given in the trial period only if the patient was not nauseated over 1 week of use. The administration of any drug with antiemetic activity was not allowed during 24 hours prior to or during the study. Written informed consent was obtained from each patient before trial enrollment.

This single-centered, phase IV, randomized, and open-labeled study was conducted from the 27^th of February 2002 to the 4^th of October 2002. On Day 1, eligible patients were randomized to receive a single i.v. dose of dolasetron 100 mg or ondansetron 8 mg infused over 15min, administered 30 min before cisplatin-based chemotherapy to control acute emesis. In the ondansetron group, two additional doses of ondansetron 8 mg after chemotherapy were given at intervals of 2 to 4 hours. To prevent delayed emesis, dolasetron 200 mg p.o. daily or ondansetron 8 mg p.o. bid was administered from the 2^nd day to a maximum of 5 days after chemotherapy Adjuvant antiemetics, except serotonin antagonists, were given only to patients with severe emesis All patients gave written informed consent.

The primary end point of the study was the proportion of patients considered to have achieved a complete response (CR; defined as no emetic episode and no use of rescue medication) during the first 24 hours (acute period) and the delayed time period (from Day 2 to Day 5±2 days) after chemotherapy administration. Secondary end points included the following: the number of emetic episodes and severity of nausea during the 24 hours after the initiation of chemotherapy; the number of emetic episodes and severity of nausea during the delayed time period (from Day 2 to Day 5±2 days); and the time to administration and need for rescue medication.

Baseline procedures were documented at a prestudy screening visit within 7 days preceding Day 1. Patients were admitted to hospital during at least the first 24 hours after chemotherapy administration (Day 1) and were assessed by clinic staffs. On Day 2 and Day 5±2 days, if patients were discharged, emetic episodes were to be recorded in diaries by the patients. Patient diaries were organized to record the following: emetic episodes; the use of rescue medication; patient global satisfaction; and the severity of nausea, which was evaluated daily until Day 5±2 days.

Safety was assessed by the following: an adverse event (AE) reporting for a period of 15 days (30 days for serious AEs); vital sign measurements; laboratory tests (hematology, blood chemistry, urinalysis); a physical examination; and electrocardiogram (ECG) recordings performed 24 hours and 1 week after drug administration. A subset of patients had an additional ECG evaluation 15 minutes after study drug administration.

The primary efficacy hypothesis of the study was that at least one dose of dolasetron was non-inferior to the ondansetron dose using a maximum delta of 25% for a CR at 24 hours. The number of patients to be included in the study was estimated based on the assumption of a responder rate of 50% in the dolasetron and ondansetron groups (18) and a difference of no more than 25% in the CR rate.

Cohorts for the analyses included an intent-to-treat (ITT) cohort, a per protocol (PP) cohort, and a safety cohort. The primary analysis was performed on the ITT cohort, which included all randomized patients who received chemotherapy and study medication. The PP cohort included all patients who completed the study at least until Day 1 and who were complaint with the study protocol. The PP analysis was performed for the primary efficacy parameter, demographic data, and baseline characteristics. The safety cohort included all treated patients who had at least one safety assessment after treatment with study drugs.

Patients were evaluated between April 2002 and October 2002. A total of 114 patients were randomized to receive one of two treatments, although two of these patients did not receive treatment. Treatment groups received either dolasetron 100 mg (n=56), or ondansetron 8 mg t.i.d. (n=58), respectively. Of the 112 patients treated, 7 were excluded from the ITT analysis because they had received chemotherapy with unacceptably low emetogenic potential. Therefore, 105 patients were included in the ITT cohort analysis.

Demographic data and baseline characteristics for patients in the ITT cohort are presented in Table 1. As a result of stratification, the distribution of patients by age, gender, chemotherapeutic history, and types of malignant disease was similar among the two treatment groups. The previous regimen and duration of chemotherapy were not considered in stratification. The most common types of malignant disease included lung cancer (approximately 26%) and gastric cancer (approximately 11%). The majority of patients (65%) were chemotherapy naive. The two treatment groups were comparable regarding the type and dose of chemotherapy administered. No patient received prophylactic corticosteroids. There were no relevant differences between the treatment groups with respect to comorbid medical conditions or the Karnofsky index.

The proportion of patients in the ITT cohort achieving a CR during the first 24 hours after the administration of cisplatin-based chemotherapy is presented in Table 2. The non-inferiority of dolasetron compared with ondansetron was demonstrated, as the lower bound of the 95% CI of the difference with ondansetron (-22.7%) was greater than the preset threshold of the 25% difference.

For the delayed (24~120 hours) time periods, the proportion of patients achieving a CR was similar for ondansetron compared with dolasetron (Table 2). Dolasetron was as effective as ondansetron at all time points.

No statistically significant difference was observed for dolasetron compared with ondansetron during the acute and delayed periods (Table 3). Patients treated with dolasetron 100 mg had similar emetic episodes compared with those treated with ondansetron 8 mg t.i.d. during the acute (p=0.3377) and delayed (p=0.1917) periods. More severe nausea and a greater need for rescue medication were observed in the dolasetron treatment group compared with the ondansetron treatment group during the acute and delayed periods, but these differences did not reach statistical significance (Table 3).

A total of 112 patients were evaluable for safety. Dolasetron was well tolerated and no AE-related withdrawals were reported during the study. There were no clinically relevant differences between the treatments with respect to the overall incidence of AEs (Fisher's exact test, p=0.267). Table 4 provides a list of treatment-emergent, drug-related AEs. Six AEs occurred in five patients of the dolasetron group (5/56 patients, 8.9%) and three AEs in two patients of the ondansetron group (2/58 patients, 3.4%). No clinically relevant differences were found between the treatment groups with respect to laboratory test results, vital sign changes, and ECG findings.