Oral Granisetron for the Prevention of Acute Late Onset Nausea and Vomiting in Patients Treated with Moderately Emetogenic Chemotherapy

Oral Granisetron for the Prevention of Acute Late Onset Nausea and Vomiting in Patients Treated with Moderately Emetogenic Chemotherapy CARL J. FRIEDMAN, a HOWARD A. BURRIS III, b KAREN YOCOM, a LINDA M. BLACKBURN, a DAVID GRUBEN c a Clinical Research and Development, and Medical Affairs, North America, SmithKline Beecham Pharmaceuticals, Collegeville, Pennsylvania, USA; b The Sarah Cannon Cancer Center, Nashville, Tennessee, USA; c Biostatistics and Data Sciences, SmithKline Beecham Pharmaceuticals, Collegeville, Pennsylvania, USA Key Words. Oral granisetron Moderately emetogenic chemotherapy Prochlorperazine Antiemetics ABSTRACT Purpose. To demonstrate the efficacy of oral granisetron 1 mg twice daily for the prevention of late onset nausea and vomiting after moderately emetogenic chemotherapy that includes cyclophosphamide, carboplatin, or doxorubicin. Methods. Prior to chemotherapy, patients were stratified by gender and randomized to receive oral granisetron (1 mg tablet twice daily) or prochlorperazine (10 mg sustained release capsule twice daily). Study agents were administered 1 h prior to and 12 h after chemotherapy. Antiemetics were administered for seven consecutive days. Efficacy variables were assessed 48 and 72 h after administration of chemotherapy, and included no emesis, no nausea, no moderate or severe nausea, and no antiemetic rescue. Safety analysis included all patients who received medication. Results. A total of 230 patients were included in the intent-to-treat analysis; 119 patients received granisetron and 111 patients received prochlorperazine. Females, and all patients combined, who received granisetron had significantly higher no-emesis rates at 48 h (p =.010 and p =.016, respectively) than patients who received prochlorperazine. No-nausea rates at 48 h were numerically higher for all patients combined and females who received granisetron rather than prochlorperazine. Response rates for no nausea or mild nausea were also numerically higher in females treated with granisetron, compared to prochlorperazine, at 48 h. Significantly more patients (p <.001) and females (p <.001) in the granisetron group than in the prochlorperazine group did not require rescue antiemetics at 48 h. At 72 h, efficacy results were comparable for granisetron and prochlorperazine. Conclusion. Oral granisetron is well tolerated and more effective than prochlorperazine in preventing nausea and vomiting for up to 48 h following treatment with moderately emetogenic chemotherapy. The Oncologist 2000;5:136-143 INTRODUCTION Acute onset emesis occurs within 24 h of chemotherapy administration, while delayed onset emesis occurs more than 24 h after chemotherapy and may persist for several days [1, 2]. For some chemotherapeutic agents, however, the onset of emesis occurs later in the first 24 h following chemotherapy and may continue into the following days [3-5]. Data support that antiemetic regimens with different mechanisms of action vary in their efficacy profiles in acute and delayed emesis [1, 5-7]. Therefore, delayed and acute emesis may be mediated by different processes and biological mediators. The majority of studies evaluating the mechanisms of acute and delayed emesis have focused on cisplatin chemotherapy. Cisplatin has a well-defined acute emetic phase that occurs approximately 2 to 4 h after administration, and a delayed phase that begins 24 to 120 h postchemotherapy and lasts for two to five days [5, 7]. Other chemotherapies, such as cyclophosphamide and carboplatin, may differ in the time Correspondence: Carl J. Friedman, M.D., SmithKline Beecham Pharmaceuticals, 1250 South Collegeville Road, Collegeville, Pennsylvania 19426, USA. Telephone: 610-917-6469; Fax: 610-917-4100. Received January 17, 2000; accepted for publication March 10, 2000. AlphaMed Press 1083-7159/00/$5.00/0 The Oncologist 2000;5:136-143 www.theoncologist.com

Friedman, Burris, Yocom et al. 137 and duration of onset of acute emesis [3-5]. The mechanisms involved in this later and more prolonged phase of acute emesis are not as well characterized. While serotonin plays a key role in mediating the initial phase of emesis, its role in emesis on the days following chemotherapy is less well-defined [1, 5-7]. The 5-HT 3 -receptor antagonists have become the standard of care for the prevention of acute nausea and vomiting associated with chemotherapy. Granisetron (Kytril, SmithKline Beecham Pharmaceuticals; Philadelphia, PA) is effective in preventing nausea and vomiting induced by emetogenic chemotherapy [8, 9] and is indicated for the prevention of highly emetogenic chemotherapy-induced nausea and vomiting [10]. Previously published results reported that granisetron tablets had greater efficacy compared to sustained-release prochlorperazine capsules in the prevention of nausea and vomiting during the 24-h period after administration of moderately emetogenic chemotherapy [11]. This report provides additional data comparing the safety and efficacy of oral granisetron and prochlorperazine at 48 and 72 h after moderately emetogenic chemotherapy administration. METHODS Study Design This was a multicenter, double-blind, parallel group comparative study performed in the United States from 1992 to 1993. Patients were screened within one week of the start of their scheduled chemotherapy. To ensure that male and female patients were equally distributed between treatment groups, patients were stratified by gender at randomization. Eligible patients were randomized to receive either oral granisetron or oral prochlorperazine on the first day of chemotherapy (day 0 or 24 h). Patient observations were conducted on site after the ingestion of the study drug, during chemotherapy infusion, and for at least 1 h following chemotherapy infusion. Vital signs were recorded within 10 min of discharge. Each patient received at discharge a packet containing study medication for 6.5 days with instructions for use. A diary card was also issued for the patient to record the number of emetic episodes, nausea severity, use of rescue medication or concomitant medication, and adverse experiences. All patients were required to return 5 to 11 days after the last study day for a follow-up evaluation and update of concomitant medications and adverse experiences. Patients who received two doses of study medication daily for seven days, and who returned for the 5- to 11-day follow-up visit, were considered to have completed the study. This study was conducted in accordance with Good Clinical Practices and the Declaration of Helsinki. The Institutional Review Board of each institution approved the protocol and statement of informed consent. Written informed consent was obtained from each participant before study entry. Patients Patients were eligible for this study if they were over 18 years of age and scheduled to receive their first cycle of moderately emetogenic chemotherapy. Males must have been surgically sterilized or agreed to practice adequate contraception. Females were of nonchildbearing potential, or had negative pregnancy tests (serum HCG) and had agreed to practice adequate contraception. Patients were chemotherapy naïve or had received nonemetogenic chemotherapy that did not require antiemetic medication and did not result in nausea or vomiting. Karnofsky performance status scores had to be at least 60%, and the physical examination, vital signs, and laboratory hematology and blood chemistries (renal and hepatic) had to be within acceptable limits. Exclusion criteria included the use of an investigational drug within 30 days or five half-lives (whichever was longer) prior to screening. Patients with unstable endocrine, cardiovascular, hepatic, renal, gastrointestinal, central nervous system (CNS), or pulmonary disease; adrenal insufficiency; or other medical disorder (including mental retardation or psychiatric disorder) were excluded. Chronic or concurrent treatment with agents known to have significant antiemetic effect (including ondansetron, corticosteroids, benzodiazepines, cannabinoids, antipsychotics, metoclopramide, or antihistamines) was not permitted. Other CNS agents were permissible if they were without antiemetic activity and the dose had been stable for one week prior to screening and during the study period. Analgesics (other than opioid agents), midazolam hydrochloride, fentanyl, and other short-acting agents were permitted. Abdominal-field radiotherapy within 24 h of antiemetic dosing or during the study period was prohibited. Patients could not have primary or secondary brain neoplasm with signs and symptoms of intracranial pressure, or brain metastases requiring treatment within 30 days of study initiation. Patients were excluded if they had known hypersensitivities to any 5-HT 3 -receptor antagonist or to phenothiazines, were unable or unwilling to comply with study protocol, or had nausea, vomiting, or retching within 24 h prior to receiving study medication. Dosage and Administration of Study Drugs Granisetron was provided as a white, film-coated triangular tablet containing 1 mg of granisetron free base. The 1-mg tablet was administered 1 h prior to the start of chemotherapy and then 12 h following chemotherapy on day 0. Prochlorperazine maleate was supplied in a sustained release capsule containing the equivalent of 10 mg of

138 Oral Granisetron for Prevention of Acute Late Onset Nausea and Vomiting prochlorperazine. The 10-mg capsule was administered 1 h prior to chemotherapy and then 12 h following chemotherapy on day 0. For both agents the remaining doses were administered at 12-h intervals, for days 1 through 6, to provide seven consecutive days of therapy. Identical placebo tablets and capsules were provided to all patients to ensure the integrity of the blinding. Patients requiring antiemetic rescue received medication of the investigator s choosing (except prochlorperazine), indicated and given specifically for nausea and vomiting. Chemotherapeutic Regimens At least one of the following moderately emetogenic chemotherapeutic agents was to be administered on day 0 and not beyond: cyclophosphamide 500 mg/m 2 to be given with other chemotherapeutic agents; cisplatin 20 mg/m 2 to 50 mg/m 2 ; carboplatin 300 mg/m 2 ; dacarbazine 350 mg/m 2 to 500 mg/m 2 ; doxorubicin >40 mg/m 2 given as a single agent or doxorubicin 25 mg/m 2 given in combination with other chemotherapeutic agents. Efficacy Assessments Efficacy parameters for nausea and vomiting assessed at 48 and 72 h after administration of moderately emetogenic chemotherapy included the proportion of patients with no emesis, no nausea, no moderate or severe nausea, and no antiemetic rescue. No emesis was defined as no vomiting or retching, and no nausea was the absence of nausea of any severity. The proportion of patients experiencing a successful outcome was determined for each end point. Response rates were reported for the percentage of patients experiencing a favorable outcome during a specified time interval for each parameter. Patients were categorized as a treatment failure if they experienced nausea and vomiting on day 1 or had insufficient data for the efficacy analysis. Safety and Tolerance Safety analysis included all patients who received study medication, and results of the safety analysis included the seven days of therapy and the follow-up period. A physical examination was conducted at screening and follow-up. The pretreatment history of medication usage, plus significant medical and surgical conditions, was recorded. Vital signs, including sitting blood pressure, pulse rate, and temperature, were recorded at screening, prior to chemotherapy, 1 h after chemotherapy, and at follow-up. Laboratory abnormalities were assessed by evaluation of blood samples collected before screening and at the 15-day follow-up visit. Adverse experiences observed by the investigator, solicited from the patient, or volunteered by the patient were also recorded. Time to onset, intensity, investigator action taken on study medication dosage, and event relationship to study medication were primary safety evaluation considerations. STATISTICAL EVALUATION Assuming a response rate of 60% with granisetron therapy, a sample of 200 patients (100 per regimen) was necessary to provide 95% power and to detect a 25% difference in response between treatment arms. All patients who received the study drug were included in the intent-to-treat analysis. The comparability of the treatment groups was assessed with respect to gender, age, and emetogenic stimulus. Proportions of patients were calculated for each day of the study and analyses were stratified by gender. Gender-specific p values were based on the Mantel-Haenszel test. Combined gender strata p values were based on the Cochran-Mantel-Haenszel test, controlling for gender. A p value less than.034 (adjusted for interim analysis) was considered statistically significant. Descriptive statistics were used for the safety evaluations. RESULTS A total of 230 patients were included in the intent-to-treat analysis; 119 patients received granisetron and 111 patients received prochlorperazine. Demographic characteristics of the patient population are shown in Table 1. A majority of patients were female, 78.2% (93/119) in the granisetron group and 83.8% (93/111) in the prochlorperazine group. Breast cancer, the most common primary disease, occurred in 55.5% of patients treated with granisetron and in 62.2% of patients treated with prochlorperazine. Eight patients treated with granisetron and 12 patients treated with prochlorperazine were withdrawn from the study for protocol violations, adverse events, or lack of efficacy. Nine patients received a single chemotherapeutic agent (three in the granisetron group and six in the prochlorperazine group). All other patients received a combination of agents. The chemotherapeutic agents as administered by gender are shown in Table 2. A total of 85 patients received rescue antiemetic medication: 28.6% (34/119) of patients treated with granisetron, and 45.9% (51/111) of patients treated with prochlorperazine. EFFICACY ASSESSMENT BY VARIABLE FOR 48 HOURS (DAY 1) AND 72 HOURS (DAY 2) No Emesis At 48 h the proportion of patients with no emesis (Table 3) and no rescue (Table 4) was greater in the group that received granisetron. Granisetron provided significantly greater emetic control than prochlorperazine in females (80.4% and 63.3%, respectively, p =.010) and in the combined gender strata (82.2% and 67.6%, respectively, p =.016). Overall, males demonstrated higher

Friedman, Burris, Yocom et al. 139 Table 1. Demographic characteristics Granisetron Prochlorperazine 1 mg BID (n = 119) 10 mg BID (n = 111) p value n (%) n (%) Gender.278 Male 26 (21.8) 18 (16.2) Female 93 (78.2) 93 (83.8) Race.692 Caucasian 94 (79.0) 90 (81.1) Black 15 (13.0) 13 (12.0) Other 10 (8.4) 8 (7.2) Age (years).630 Mean ± SD 58.4 ± 14.1 59.3 ± 13.5 Range 21-84 25-93 Alcohol use (units per week)*.265 Mean ± SD 4.0 ± 13.2 2.4 ± 8.8 Range 0-86.4 0-84.0 p value for gender and race is from a chi-square test of independence of the two treatment groups. p value for age and alcohol composite is from a one-way ANOVA F-test. *A unit of alcohol composite is equivalent to 150 ml of wine, 0.25 l of beer, or 50 ml of spirits. Table 2. Moderately emetogenic chemotherapeutic agents most frequently reported Granisetron Prochlorperazine (n = 119) (n = 111) Chemotherapeutic agent n (%) n (%) Female 93 (100.0) 93 (100.0) Cyclophosphamide 85 (91.4) 83 (89.2) Fluorouracil 59 (63.4) 58 (62.4) Methotrexate 37 (39.8) 36 (38.7) Doxorubicin 32 (34.4) 37 (39.8) Carboplatin 13 (14.0) 13 (14.0) Male 26 (100.0) 18 (100.0) Carboplatin 12 (46.2) 9 (50.0) Doxorubicin 11 (42.3) 6 (33.3) Cyclophosphamide 10 (38.5) 7 (38.9) Etoposide 9 (34.6) 5 (27.8) Vincristine 9 (34.6) 8 (44.4) Fluorouracil 5 (19.2) 2 (11.1) Combined 119 (100.0) 111 (100.0) Cyclophosphamide 95 (79.8) 90 (81.1) Fluorouracil 64 (53.8) 60 (54.1) Doxorubicin 43 (36.1) 43 (38.7) Methotrexate 37 (31.1) 36 (32.4) Carboplatin 25 (21.0) 22 (19.8) Vincristine 12 (10.1) 12 (10.8) response rates than females to antiemetic therapy. There were no significant differences, however, at 48 or 72 h between males who received granisetron and those who received prochlorperazine. At 72 h, no-emesis rates were similar for all patients who received granisetron or prochlorperazine, and more than 90% of patients in both treatment groups were emesis-free (Table 3). No Nausea Overall, proportionately more patients experienced nausea than emesis, as indicated by notably smaller no-nausea rates (Table 5). No-nausea rates at 48 h were higher for patients treated with granisetron (53.4%) than for patients treated with prochlorperazine (48.6%, p =.629). At 48 h the no-nausea rates were also higher for females treated with granisetron than for females treated with prochlorperazine (47.8% and 42.9%, respectively, p =.501). The differences in the 48-hour no-nausea rates were not statistically significant. At 72 h no-nausea rates for female patients treated with granisetron (51.1%) were significantly lower than those treated with prochlorperazine (65.6%, p =.050); however, the differences in rates at 72 h for all patients did not reach statistical significance (Table 5). No More than Mild Nausea The response rates for severity of nausea almost exclusively reflect observations in the female study population. Only one male in each treatment group reported moderate or severe nausea on any study day. While not statistically significant, higher response rates at 48 h were observed in females treated with granisetron (73.9%) rather than with prochlorperazine (64.8%). No nausea or mild nausea rates (85.6%) were equivalent at 72 h for females in both treatment groups. Only one male reported this symptom after 72 h (Table 6). Rescue Medication At least one dose of rescue medication was administered to 37% of patients during the seven-day study period. Overall,

140 Oral Granisetron for Prevention of Acute Late Onset Nausea and Vomiting Table 3. Patients with no emesis at 48 h and at 72 h Granisetron Prochlorperazine p value n/n* (%) n/n (%) Female 48 h 74/92 80.4 57/90 63.3.010 72 h 80/90 88.9 82/89 92.1.460 Male 48 h 23/26 88.5 16/18 88.9.965 72 h 25/26 96.2 18/18 100.0.405 Combined 48 h 97/118 82.2 73/108 67.6.016 72 h 105/116 90.5 100/107 93.5.358 Gender-specific p values are based on the Mantel-Haenszel test. The p value for combined strata is based on the Cochran-Mantel-Haenszel test, controlling for gender. *One patient in the granisetron group did not provide information for analysis. Table 4. Number and percent of patients not receiving antiemetic rescue medication Granisetron Prochlorperazine p value n/n* (%) n/n (%) Female 48 h 76/93 81.7 51/93 54.8 <.001 72 h 77/93 82.8 70/93 75.3 <.209 Male 48 h 24/26 92.3 13/18 72.2 <.077 72 h 23/26 88.5 16/18 88.9 <.965 Combined 48 h 100/119 84.0 64/111 57.7 <.001 72 h 100/119 84.0 86/111 77.5 <.245 Gender-specific p values are based on the Mantel-Haenszel test. The p value for combined strata is based on the Cochran-Mantel-Haenszel test, controlling for gender. Table 5. Patients with no nausea at 48 h and at 72 h Granisetron Prochlorperazine p value n/n* (%) n/n (%) Female 48 h 44/92 47.8 39/91 42.9.501 72 h 46/90 51.1 59/90 65.6.050 Male 48 h 19/26 73.1 14/18 77.8.726 72 h 21/26 80.8 15/18 83.3.830 Combined 48 h 63/118 53.4 53/109 48.6.629 72 h 67/116 57.8 74/108 68.5.057 Gender-specific p values are based on the Mantel-Haenszel test. The p value for combined strata is based on the Cochran-Mantel-Haenszel test, controlling for gender. *One patient in the granisetron group did not provide data for analysis. at 48 h, 84% of patients who received granisetron did not require rescue antiemetic medication versus 57.7% of patients who received prochlorperazine (p <.001). Significantly more females treated with granisetron did not require the use of rescue antiemetic medication at 48 h than females treated with prochlorperazine (81.7% versus 54.8%, p <.001). Although

Friedman, Burris, Yocom et al. 141 Table 6. Patients with no nausea or mild nausea at 48 h and at 72 h Granisetron Prochlorperazine p value n/n* (%) n/n (%) Female 48 h 68/92 73.9 59/91 64.8.184 72 h 77/90 85.6 77/90 85.6 1.000 Male 48 h 25/26 96.2 17/18 94.4.791 72 h 25/26 96.2 18/18 100.0.405 Combined 48 h 93/118 78.8 76/109 69.7.176 72 h 102/116 87.9 95/108 88.0.866 Gender-specific p values are based on the Mantel-Haenszel test. The p value for combined strata is based on the Cochran-Mantel-Haenszel test, controlling for gender. *One patient in the granisetron group did not provide data for analysis. not statistically significant, more patients overall (84%) and female patients (82.8%) treated with granisetron, did not require the use of rescue antiemetic medication at 72 h than respective groups for prochlorperazine (77.5% and 75.3%) (Table 4). Overall, gender subanalysis found that a higher proportion of males, compared to females, did not require rescue medication. SAFETY AND TOLERANCE EVALUATION The evaluation of safety and tolerance includes all intentto-treat patients (n = 230) who received study medication. At least one adverse event was reported by 81.7% of patients during the study; 84% treated with granisetron and 79.3% treated with prochlorperazine. Overall, the most commonly reported adverse events were headache, constipation, diarrhea, and fatigue. In patients receiving granisetron headache (36.1%) and constipation (31.1%) occurred most frequently. Patients treated with prochlorperazine reported headache (28.8%) and diarrhea (19.8%) as the most common adverse events. Approximately 75% of patients in both treatment groups reported mild adverse events, and approximately 36% of patients reported moderate adverse events. There was a similar incidence of severe adverse events for granisetron and prochlorperazine (12.6% versus 13.5%). Serious adverse events occurred in 20 patients (9 granisetron and 11 prochlorperazine) of which none were considered to be related to study agents. Adverse events considered to be related to a study agent were reported in 1.7% of patients who received granisetron and in no patients who received prochlorperazine. Of the three deaths reported (one granisetron patient and two prochlorperazine patients), two were considered to be unrelated to therapy and due to the progression of primary disease; one death (congestive cardiomyopathy) was determined to be probably unrelated to prochlorperazine administration. The clinical chemistry results changed minimally from screening baseline values and follow-up values at 5 to 11 days after the last day of the chemotherapy cycle. DISCUSSION There is evidence that the pathogenesis of the different temporal phases of emesis, as well as the phases themselves, may differ according to the chemotherapy agent. Cisplatin is the most widely studied agent and has a distinct biphasic pattern of emesis differentiated as acute and delayed emesis [5]. While 5-HT 3 -receptor antagonist efficacy data and 5-hydroxyindoleacetic acid urinary excretion studies indicate that serotonin mediates cisplatin-induced acute emesis [8, 12] cisplatin-induced delayed emesis may be primarily caused by nonserotonin-related mechanisms [12]. Approximately 80% of patients in our study, however, received cyclophosphamide. Emesis induced by cyclophosphamide appears to occur in a monophasic pattern, with a later onset than is seen with cisplatin, that continues into the following days [4, 5]. The use of 5-HT 3 -receptor antagonists has been shown to achieve good control of acute late-onset emesis induced by cyclophosphamide-based regimens [5, 13]. Therefore, serotonin release may play a more significant role in the later phases of cyclophosphamide-induced emesis than in cisplatin-induced delayed emesis. In a recent study of a cyclophosphamide-based regimen, patients received an initial dose of i.v. granisetron 3 mg on the day of chemotherapy, followed by randomization to oral granisetron or placebo for two days [14]. Oral granisetron (2 mg daily) provided significantly better total control than placebo (p <.05 and p <.01, days 1 and 2, respectively) for two days after the initial 24 h postcyclophosphamide administration. In another study i.v. granisetron (3-mg single dose) was administered prior to chemotherapy in 43 patients who received fluorouracil, epirubicin, and cyclophosphamide. On day 2, 84% of patients had experienced zero to two emetic episodes, and

142 Oral Granisetron for Prevention of Acute Late Onset Nausea and Vomiting this proportion increased to greater than 90% after day 4 of therapy [13]. Oral ondansetron (8 mg three times daily) has also been shown to provide complete control (zero emetic episodes) in 92.3% of patients on the second day following chemotherapy in a cyclophosphamide-based regimen [15]. A more recent investigation with cyclophosphamidedoxorubicin chemotherapy randomized patients to oral ondansetron plus i.v. dexamethasone, or i.v. dexamethasone plus i.m. orphenadrine (control group) on the first day of chemotherapy. To control delayed emesis, patients initially randomized to ondansetron received oral ondansetron (8 mg three times daily), and control patients received oral metoclopramide (0.5 mg/kg four times daily) plus i.m. dexamethasone (8 mg twice daily) on days 2 and 3. In the worst-day analysis, ondansetron provided greater complete and major protection (zero to two emetic episodes) from delayed emesis (days 2 to 5) than metoclopramide plus dexamethasone (91% versus 66%) [16]. Although other neurotransmitters may be involved [17], these data suggest that serotonin plays a role in the acute late onset of cyclophosphamide-induced emesis and provides a rationale for the use of 5-HT 3 -receptor antagonists with this chemotherapy regimen. Owing to the different protocols used in these studies, additional investigations are required to further define the pathogenesis of emesis following cyclophosphamide and carboplatin chemotherapy regimens. Effective control of nausea and vomiting during the initial 24 h postchemotherapy provides some protection against the onset of delayed nausea and vomiting (first episodes occurring more than 24 h after chemotherapy) [6, 18-20]. If granisetron and other 5-HT 3 -receptor antagonists provide effective control in the acute emetic phases, then they may also decrease the incidence of the delayed emetic phases. In a recent medication-use evaluation that reviewed 200 inpatient charts, fewer patients who received granisetron, as compared to ondansetron, required antiemetic rescue (24% versus 37%; p =.05, χ 2 test) for breakthrough nausea and vomiting after high-dose cisplatin therapy (>75 mg/m 2 ) [21]. An indirect indicator of efficacy in our investigation was the use of rescue antiemetic medication after emesis or for alleviating significant nausea because such use can be viewed as a response to insufficient prophylaxis. Patients treated with granisetron required less use of rescue antiemetic medication, supporting the efficacy of granisetron beyond the 24-h period after chemotherapy administration. These results are consistent with those from the patient chart review mentioned previously. The majority of clinical studies have examined the antiemetic efficacy of granisetron for the 24-h period immediately following chemotherapy administration. The 24-h results previously reported for this investigation used total control (no emesis, no retching, no nausea of any severity, and no use of rescue medication) and complete response (no emesis, no retching, no greater than mild nausea, and no use of antiemetic rescue medication) as the primary efficacy parameters [11]. Total control for all patients treated with granisetron and prochlorperazine was 58% and 33% (p <.001), respectively. Complete response rates for all patients treated with granisetron and prochlorperazine were 74% and 41% (p <.001), respectively. A higher percentage of males than females experienced greater total and complete responses for both study agents [11]. This is not unexpected, since numerous studies have demonstrated that male patients tend to respond more favorably to antiemetic therapy. Data for 48 h of antiemetic therapy also show that oral granisetron has improved efficacy compared to oral, sustained release prochlorperazine. Although total control and complete response were not measured at 48 and 72 h, data are available for secondary parameters. At 24 h 82% of granisetron patients and 48% of prochlorperazine patients (p <.001) had no emesis [11]. The difference in the proportion of patients with no emesis remained significant (p =.016) up to 48 h (Table 3). Overall, patients who received granisetron showed consistently better responses for emesis and use of rescue antiemetics up to 48 h. The differences seen in the no-nausea rates did not reach statistical significance. At the time of this study, oral 5-HT 3 -receptor antagonists had not yet become the standard of care and the use of prochlorperazine 10-mg sustained-release capsules twice daily represented effective antiemetic therapy. This prochlorperazine regimen was also easier to administer than the standard formulation of prochlorperazine tablets, which should be administered more frequently. The results for granisetron and prochlorperazine at 48 and 72 h have been presented to build on the previously published 24-h results [11], and to present new information on the duration of effect of granisetron tablets. These additional data indicate that granisetron is effective in preventing emesis occurring after 24 h in patients receiving moderately emetogenic chemotherapy (primarily cyclophosphamide). Although the incidence of chemotherapy-induced nausea and vomiting was less in male patients, granisetron was more effective overall than prochlorperazine. These data may be used to design comparative studies for 5-HT 3 -receptor antagonists versus current prochlorperazine regimens. The prevention of nausea and vomiting during the early emetic phases after chemotherapy administration may allow more patients to remain emesis-free and reduce the need for antiemetic rescue agents. The results reported here, and previously [11], demonstrate that granisetron, compared to prochlorperazine, is well tolerated and effective in

Friedman, Burris, Yocom et al. 143 preventing emesis for up to 48 h following treatment with moderately emetogenic chemotherapy regimens that include cyclophosphamide, carboplatin, and doxorubicin. ACKNOWLEDGMENT Supported by an unrestricted educational grant from SmithKline Beecham Pharmaceuticals. REFERENCES 1 Ettinger DS. Preventing chemotherapy-induced nausea and vomiting: an update and a review of emesis. Semin Oncol 1995;22:6-18. 2 Morrow GR, Hickok JT, Burish TG et al. Frequency and clinical implications of delayed nausea and delayed emesis. Am J Clin Oncol 1996;19:199-203. 3 Martin M, Diaz-Rubio E, Sanchez A et al. The natural course of emesis after carboplatin treatment. Acta Oncol 1990;29:593-595. 4 Fetting JH, Grochow LB, Folstein MF et al. The course of nausea and vomiting after high-dose cyclophosphamide. Cancer Treat Rep 1982;66:1487-1493. 5 Martin M. The severity and pattern of emesis following different cytotoxic agents. Oncology 1996;53(suppl 1):26-31. 6 Mantovani G, Maccio A, Curreli L et al. Comparison of oral 5- HT 3 -receptor antagonists and low-dose oral metoclopramide plus i.m. dexamethasone for the prevention of delayed emesis in head and neck cancer patients receiving high-dose cisplatin. Oncol Rep 1998;5:273-280. 7 Cubeddu LX. Serotonin mechanisms in chemotherapyinduced emesis in cancer patients. Oncology 1996;53(suppl 1):18-25. 8 Gralla RJ, Navari RM, Hesketh PJ et al. Single-dose oral granisetron has equivalent antiemetic efficacy to intravenous ondansetron for highly rmetogenic cisplatin-based chemotherapy. J Clin Oncol 1998;16:1568-1573. 9 Perez EA, Hesketh P, Sandbach J et al. Comparison of single-dose oral granisetron versus intravenous ondansetron in the prevention of nausea and vomiting induced by moderately emetogenic chemotherapy: a multicenter, double-blind, randomized parallel study. J Clin Oncol 1998;16:754-760. 10 Kytril (granisetron HCl) Tablets. Prescribing Information. SmithKline Beecham Pharmaceuticals, Philadelphia, PA, 1998. 11 Burris H, Hesketh P, Cohn J et al. Efficacy and safety of oral granisetron versus oral prochlorperazine in preventing nausea and emesis in patients receiving moderately emetogenic chemotherapy. Cancer J Sci Am 1996;2:85-90. 12 Cubeddu LX, Hoffmann IS. Participation of serotonin on early and delayed emesis induced by initial and subsequent cycles of cisplatinum-based chemotherapy: effects of antiemetics. J Clin Pharmacol 1993;33:691-697. 13 Tan EH, Ang PT, Khoo KS. Control of emesis by intravenous granisetron in breast cancer patients treated with 5-FU, epirubicin and cyclophosphamide. Support Care Cancer 1994;2:197-200. 14 Guillem V, Carrato A, Rifá J et al. High efficacy of oral granisetron in the total control of cyclophosphamide-induced prolonged emesis. Proc Am Soc Clin Oncol 1998;17:46a. 15 Rosso R, Campora E, Cetto G et al. Oral ondansetron (GR38032F) for the control of acute and delayed cyclophosphamide-induced emesis. Anticancer Res 1991;11:937-939. 16 Campora E, Giudici S, Merlini L et al. Ondansetron and dexamethasone versus standard combination antiemetic therapy. A randomized trial for the prevention of acute and delayed emesis induced by cyclophosphamide-doxorubicin chemotherapy and maintenance of antiemetic effect at subsequent courses. Am J Clin Oncol 1994;17:522-526. 17 Kris MG, Pisters KMW, Hinkley L. Delayed emesis following anticancer chemotherapy. Support Care Cancer 1994;2:297-300. 18 Bilgrami S, Fallon BG. Chemotherapy-induced nausea and vomiting: easing patients fear and discomfort with effective antiemetic regimens. Postgrad Med 1993;94:55-64. 19 Italian Group for Antiemetic Research. Ondansetron versus granisetron, both combined with dexamethasone, in the prevention of cisplatin-induced emesis. Ann Oncol 1995;6:805-810. 20 Tavorath R, Hesketh PJ. Drug Treatment of chemotherapyinduced delayed emesis. Drugs 1996;52:639-648. 21 Nolte MJ, Berkery R, Pizzo B et al. Assuring the optimal use of serotonin antagonist antiemetics: the process for development and implementation of institutional antiemetic guidelines at Memorial Sloan-Kettering Cancer Center. J Clin Oncol 1998;16:771-778.