218 Journal of Pain and Symptom Management Vol. 18 No. 3 September 1999 Original Article Antiemetic Prophylaxis with Ondansetron and Methylprednisolone vs Metoclopramide and Methylprednisolone in Mild and Moderately Emetogenic Chemotherapy Nicholas B. Tsavaris, MD, Christos Koufos, Michael Katsikas, MD, Adonis Dimitrakopoulos, MD, Eleni Athanasiou, MD, and Garyfalia Linardaki, MD Oncology Unit, Pathophysiology Department, University of Athens, School of Medicine, Laiko General Hospital, Athens, Greece Abstract The purpose of the present study was to examine whether its is possible to successfully replace ondansetron (OND) with metoclopramide (MCP) in patients exposed to moderately emetogenic chemotherapy who did not experience severe nausea and vomiting while undergoing OND treatment during their first chemotherapy cycle. After switching to MCP, patients continued with this drug for three cycles, provided that they had adequate control of nausea and vomiting. Otherwise, they were switched back to OND. There were 76 patients, 60 women and 16 men, whose median age was 56 (mean 58) years. Karnofsky performance status score was 100 in18 patients, 90 in 23, and 80 in 11 patients. No patient had previous chemotherapy. Thirty-four patients had breast cancer and received fluorouracil 500 mg/m 2, epirubicin 100 500 mg/m 2, and cyclophosphamide 500 mg/m 2. Twelve patients had small cell lung cancer and received carboplatin 400 mg/m 2 etoposide 120 mg/m 2 3 days. Twenty patients with ovarian cancer received carboplatin 350 mg/m 2 and cyclophosphamide 500 mg/m 2. Ten patients had cancer of unknown primary and received carboplatin 400 mg/m 2, epirubicin 60 mg/m 2, and etoposide 120 mg/m 2 3 days. The OND schedule consisted of methylprednisolone 40 mg intravenous bolus followed by OND 8 mg in a 15-min infusion before chemotherapy, followed by OND 4 mg orally 3 on the same and the next 2 days. Patients who did not experience nausea and vomiting with OND continued with an MCP schedule consisting of methylprednisolone 40 mg bolus followed by MCP 2 mg/kg in a 15-min infusion before chemotherapy, followed by MCP (20 mg 4 on the day of therapy and the next 2 days after). Patients who failed with MCP or OND continued with OND. Considering our results as a whole, the intensity of nausea does not appear to influence the results of Gralla s scale. The results of Gralla s scale do not appear to be affected by the analysis of the antiemetic results and nausea on the next 2 days following chemotherapy administration. Overall, patients received 145 cycles with OND and 159 cycles with MCP. Of the 76 patients receiving OND-based antiemetic regimen during the first cycle, 13 (21%) experienced severe vomiting (Grade 2, 3) and the remaining 63 (79%) had mild or no vomiting (Grade 0, 1). Patients Address reprint requests to: Nicolas B. Tsavaris, MD, Oncology Unit, Pathophysiology Department, University of Athens, School of Medicine, Laiko General Hospital, Athens 115 27, Greece. Accepted for publication: November 24, 1998 U.S. Cancer Pain Relief Committee, 1999 0885-3924/99/$ see front matter Published by Elsevier, New York, New York PII S0885-3924(99)00071-8
Vol. 18 No. 3 September 1999 Antiemetic Prophylaxis in Emetogenic Chemotherapy 219 Clinical and laboratory follow-up was carried out before starting each cycle. Follow up biochemical and hematologic tests were performed daily during the first week of treatment and weekly thereafter; chest radiographs and creatinine clearance were obtained monthly. All patients included in the trial were 70 years old and had no prior chemotherapy or antiemetic therapy. The Karnofsky performance status score was 80%. Each patient had to have a leukocyte count 4000/mm 3, a platelet count 100,000/mm 3, serum creatinine below 1.4 mg/dl, and serum bilirubin below 1.0 mg/ dl to be eligible for the current study. Patients were evaluable if they completed four chemotherapeutic cycles. The 76 patients included 60 women and 16 men, median age 56 years (mean 58 years), performance status (Karnofsky) was 100 in18 patients, 90 in 23 patients, and 80 in 11 patients. No patient had previous chemotherapy (see Table 2). The 34 patients with breast canwith Grade 0, 1 vomiting (63, 83%) continued with MCP in the second cycle. The final number of patients who failed on MCP, after 4 cycles of chemotherapy increased to 33 (43%); 43 (57%) were able to complete chemotherapy with MCP. Headache occurred in 15 (10%) cycles with OND and 8 (5%) with MCP. Flushing was noted in 12 (8%), and constipation occurred in 43 (30%) of OND cycles, and extrapyramidal manifestations occurred in 3 (5%) of patients receiving MCP. Diarrhea was noted in 3 (2%) of cycles with OND and in 28 (18%) with MCP. The cost ratio between MCP and OND was 1:14. If we administered OND only in patients who needed it, the overall cost decreased to 44%. Following the strategy applied in the present study, the cost decreased to 47%. J Pain Symptom Manage 1999;18: 218 222. U.S. Cancer Pain Relief Committee, 1999. Key Words Antiemetic Prophylaxis, nausea-vomiting ondansetron, metoclopramide, methylprednizolone Introduction Intravenous cyclophosphamide, doxorubicin, epirubicin, and carboplatin, used alone or in combination, are considered moderately emetogenic chemotherapeutic drugs. 1 These cytotoxic agents are widely applied in a variety of tumors, such as breast, lung, and lymphomas. They are usually administered in the outpatient setting. Protection from nausea and vomiting after chemotherapy significantly improved by the addition of a corticosteroid to either metoclopramide or ondansetron. 2 5 The aim of the present study was to examine whether its is possible to obtain a similarly successful antiemetic result by substituting metoclopramide for ondansetron in patients exposed to moderately emetogenic chemotherapy. This substitution was tested in patients who did not experience severe nausea and vomiting during the first chemotherapy cycle, during which they received ondansetron. Patients continued on metoclopramide for three cycles, provided that they had adequate control of nausea and vomiting, otherwise they were switched back to ondansetron. Patients and Methods Patients Seventy-six patients with histologically confirmed cancer entered this study; all were evaluable (Table 1). The study was approved by the ethics committee of our institution. Pretreatment evaluation included a complete history and physical examination, hematologic and biochemical tests, and chest radiography. Table 1 Patient Characteristics Number 76 Male/Female 60/16 Age Mean (yr) 58 Median (yr) 56 Performance status 100 38 (Karnofsky) 90 23 80 15 Previous chemotherapy 0 Breast cancer 34 Small cell lung cancer 12 Ovarian cancer 20 Unknown primary origin 10
220 Tsavaris et al. Vol. 18 No. 3 September 1999 cer received fluorouracil 500 mg/m 2, epirubicin 100 500 mg/m 2, cyclophosphamide 500 mg/m 2. Twelve patients with small cell lung cancer received carboplatin 400 mg/m 2 plus etoposide 120 mg/m 2 for 3 days. The 20 ovarian cancer patients received carboplatin 350 mg/m 2 and cyclophosphamide 500 mg/m 2. Ten patients had cancer of unknown primary and received carboplatin 400 mg/m 2, epirubicin 60 mg/m 2, and etoposide 120 mg/m 2 3 days. Procedure We tried to achieve the best result of antiemetic therapy (Grade 0, 1, according to Gralla s scale 6 ), specifically nausea Grade 0, 1 and vomiting control of 0, 1 for the 3 days after treatment. Patients started chemotherapy with ondansetron as the antiemetic therapy during the first chemotherapy cycle. The reason was to separate out those with a tendency for nausea and vomiting. Thereafter, patients who failed antiemetic therapy (Grade 2 4 according to Gralla s scale 6 ) continued with ondansetron, whereas the rest (Grade 0 1 according to Gralla s scale) continued with metoclopramide as the antiemetic during the next cycle. After the second cycle, patients who failed the antiemetic therapy (Grade 2 4 according to Gralla s scale, nausea Grade 2, 3, and vomiting control during 3 days of Grade 2, 3) with metoclopramide and those with extrapyramidal symptoms due to metoclopramide continued with ondansetron. During the third and fourth cycle of chemotherapy, patients who failed metoclopramide also were crossed over to the group with ondansetron as antiemetic therapy. After the fourth cycle, we evaluated the percentage of patients who needed ondansetron, and the total cost of antiemetic therapy. Diagnosis Table 2 Chemotherapeutic Schedules Number of patients Cytostatic combination a Breast cancer 34 FE(*)C Small cell lung cancer 12 JM8 VP Ovarian cancer 20 JM8 CTX Unknown primary origin 10 JM8 E(**) VP a F fluorouracil 600 mg/m 2 ; E epirubicin 100 (*) or 60 (**); C or CTX cyclophosphamide 600 mg/m 2 ; JM8 carboplatin 350 mg/m 2 ; VP etoposide 120 mg/m 2 3 days. The ondansetron schedule consisted of methylprednisolone 40 mg iv bolus followed by ondansetron 8 mg in a 15-min infusion before chemotherapy, followed by ondansetron 4 mg orally three times on the same day and the next 2 days. The metoclopramide schedule consisted of methylprednisolone 40 mg iv bolus followed by metoclopramide 2 mg/kg in a 15-min infusion before chemotherapy, followed by metoclopramide tablets 20 mg 4 times day of therapy and each of the next 2 days after. Patients who failed with metoclopramide or ondansetron continued with ondansetron. Patients completed a diary card daily on the day of chemotherapy and the 2 days after. Vomits, retches, nausea, and side effects from drugs were evaluated. The patients and the accompanying person were trained in evaluating nausea, retches, and vomiting. Patients were evaluated for the presence of vomits, retches, nausea, and side effects before each dose of antiemetic medications, at the end of the 96-hour observation period, and at least every 3 hours during the first 24 hours. The antiemetic response during the first 24 hours and for the following 2 days after chemotherapy was defined according to Gralla s classification, 6 i.e., complete response (no emetic episode), major response (1 2 emetic episodes), minor response (3 5 emetic episodes) or failure ( 5 emetic episodes or rescue antiemetic therapy required per 24 hours). One emetic episode signifies any episode of vomiting or retching. A vomiting scale was also employed in the evaluation of retches. In the same way, the next 2 days after chemotherapy were evaluated. Nausea was evaluated as 0: no nausea; 1: transient, of less than 24 hours duration; 2: continuous or periodic, at least 24 hours requiring bedrest intermittenly, lifestyle is affected; 3: needs bedrest for at least 1 day and is unable to stand up or walk and feels relief in bed. Side effects of treatment were also directly observed and recorded, such as diarrhea, headache, hypotension, constipation, or other symptoms. These were evaluated according to the World Health Organization (WHO) scales. 7 Statistical Analysis To test for differences between the examined groups, the nonparametric Wilcoxon rank sum
Vol. 18 No. 3 September 1999 Antiemetic Prophylaxis in Emetogenic Chemotherapy 221 Fig. 1. Results according to Gralla s scale. (*) Three patients (one Grade 1, two Grade 2) presented extrapyramidal symptoms and returned to ondansetron therapy. test was used. To examine group differences more closely, the chi-square tests and, in one case, Fisher s exact test were used. The two treatments were compared for their efficacy over the first 24 hours following chemotherapy (acute emesis) and over the following 3 days (delayed emesis). 8 11 Results There were no differences in reported nausea, and only the vomiting results are depicted in Fig. 1. There were 145 cycles with ondansetron as antiemetic treatment and 159 with metoclopramide. Of the 76 patients receiving ondansetron during the first cycle, 13 (21%) Table 3 Side Effects a Ondansetron Metoclopramide P Headache 15 (10%) 8 (5%) NSS Flash 12 (8%) 0 Constipation 43 (30%) 0 Diarrhea 3 (2%) 28 (18%) Extrapyramidal manifestations 0 3 (5%) a According to the number of chemotherapeutic courses except extrapyramidal manifestation which was evaluated according to the number of patients. experienced intense vomiting (Grade 2, 3) and the remaining 60 (79%) low (Grade 0, 1) vomiting activity. Patients with low vomiting activity (Grade 0, 1 63%, 83%) continued with metoclopramide therapy in the second cycle. Of these patients, 8 failed with vomiting Grade 2, 3, and 3 patients (5%) (two Grade 2 and one Grade 3) presented extrapyramidal manifestations. All these patients continued with ondansetron therapy. In the third cycle, 52 patients received metoclopramide and 5 experienced intense vomiting activity and continued with ondansetron. Finally 47 patients entered the fourth cycle. Of these, 4 failed metoclopramide therapy. Thus, the total number of patients who failed metoclopramide was 33 (43%), whereas the number who were able to complete chemotherapy with metoclopramide was 43 (57%). Headache was encountered in 15 (10%) of cycles with ondansetron and 8 (5%) in metoclopramide; flushing was noted only in 12 (8%) of ondansetron cycles and constipation occurred in 43 (30%). Extrapyramidal manifestations occurred in 3 (5%) of patients receiving metoclopramide. Diarrhea was noted in 3 (2%) of cycles with ondansetron and in 28 (18%) with metoclopramide (Table 3).
222 Tsavaris et al. Vol. 18 No. 3 September 1999 The cost ratio between methylprednisolone metoclopramide and methylprednisolone ondansetron was 1:14. Based on the final percentages of our study, if we administered ondansetron only in patients who needed it, the total cost of therapy would decrease to 44% of the total cost if ondansetron was administered to all. Following the strategy of the present study, the cost was decreased to 47%. Discussion In mild or moderately emetogenic chemotherapy, there is established antiemetic superiority of ondansetron over metoclopramide. 1 3 In the present study, an attempt was made to switch patients to metoclopramide after a successful first course of ondansetron. Almost half of the patients (51%) could complete their full chemotherapy program with metoclopramide. Failure of antiemetic therapy after switching to metoclopramide was successfully controlled with reinstitution of ondansetron. Extrapyramidal manifestations due to metoclopramide administration, a rare occurrence in the present study (5%), most probably due to the age of the patients and to the moderate metoclopramide doses, also can be overcome by switching antiemetic therapy to ondansetron. The less than ideal protection of acute emesis (compared to ondansetron), however, might adversely affect anticipatory nausea and vomiting during subsequent chemotherapy cycles. Thus, using a less effective antiemetic for these patients could make anticipatory emesis more problematic and this may well balance out the cost savings achieved by using metoclopramide. This possibility requires additional study. If the concern about anticipatory emesis is not borne out, the approach evaluated in this study provides a method for reducing the cost of antiemetic therapy by almost half that required by the use of ondansetron alone. This may be highly advantageous in some settings when chemotherapy involves mildly or moderately emetogenic drugs. References 1. Del Favero A, Roila F, Tonato M. Reducing chemotherapy-induced nausea and vomiting: current perspectives and future possibilities. Drug Safety 1993;9:410 428. 2. Roila F, Tonato M, Ballatori E, et al. Comparative studies of various antiemetic regimens. Support Care Cancer 1996;4:270 280. 3. Italian Group for Antiemetic Research. Ondansetron dexamethasone versus breast metoclopramide dexamethasone diphenhydramine in prevention of cisplatin-induced emesis. Lancet 1992; 340:96 99. 4. Tsavaris N, Tsaroucha-Noutsou E, Bacoyiannis C, Mylonakis N, Valilis P, Kozatsani-Halividi D, Tsoutsos H, Droufakou S, Kosmidis P. Antiemetic efficacy of high dose metoclopramide and dexamethasone in patients receiving cisplatin chemotherapy: a randomized study. Oncology 1992;49:479 483. 5. Tsavaris N, Mylonakis N, Bacoyiannis C, Katsikas M, Lioni A, Kosmidis P. Comparison of ondansetron (GR 38032F) vs ondansetron plus methylprednisolone as antiemetic prophylaxis during cisplatin containing chemotherapy. J Pain Symptom Manage 1994;9:254 257. 6. Gralla R. Approaches to management of emesis. Clinician 1988;6:26 39. 7. Miller A, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981; 47:207 217. 8. Van Elteren PH. On the contribution of independent two sample test of Wilcoxon. Bulletin Int Statist Inst 1960;12:351 361. 9. Machin D, Campbell MJ. Statistical tables for the design of clinical trials. Oxford: Blackwell, 1987: 79 83. 10. Campbell MJ. Statistical tables for the design of clinical trials. Oxford: Blackwell, 1987:79 83. 11. Dixon WJ, Massey FJ Jr. Introduction to statistical analysis, 4th ed. New York: McGraw-Hill, 1983.