|
|
| |||||||||||||||||||||||||||||||||||||||||
Background The localization of substance P in brain-stem regions associated with vomiting, and the results of studies in ferrets, led us to postulate that a neurokinin-1–receptor antagonist would be an antiemetic in patients receiving anticancer chemotherapy.
Methods In a multicenter, double-blind, placebo-controlled trial involving 159 patients who had not previously received cisplatin, we evaluated the prevention of acute emesis (occurring within 24 hours) and delayed emesis (occurring on days 2 to 5) after a single dose of cisplatin therapy (70 mg or more per square meter of body-surface area). Before receiving cisplatin, all the patients received granisetron (10 µg per kilogram of body weight intravenously) and dexamethasone (20 mg orally). The patients were randomly assigned to one of three treatments in addition to granisetron and dexamethasone: 400 mg of an oral trisubstituted morpholine acetal (also known as L-754,030) before cisplatin and 300 mg on days 2 to 5 (group 1), 400 mg of L-754,030 before cisplatin and placebo on days 2 to 5 (group 2), or placebo before cisplatin and placebo on days 2 to 5 (group 3). Additional medication was available at any time to treat occurrences of vomiting or nausea.
Results In the acute-emesis phase, 93 percent of the patients in groups 1 and 2 combined and 67 percent of those in group 3 had no vomiting (P<0.001). In the delayed-emesis phase, 82 percent of the patients in group 1, 78 percent of those in group 2, and 33 percent of those in group 3 had no vomiting (P<0.001 for the comparison between group 1 or 2 and group 3). The median nausea score in the delayed-emesis phase was significantly lower in group 1 than in group 3 (P=0.003). No serious adverse events were attributed to L-754,030.
Conclusions The neurokinin-1–receptor antagonist L-754,030 prevents delayed emesis after treatment with cisplatin. Moreover, combining L-754,030 with granisetron plus dexamethasone improves the prevention of acute emesis.
The severity and pattern of chemotherapy-induced emesis depend on the specific agents used, the dose, and the regimen.4 Cisplatin is most commonly associated with profound nausea and vomiting, which follow a distinct pattern of an acute phase (within 24 hours after chemotherapy) and a delayed phase (on days 2 to 5).5 Severe acute emesis occurs in virtually all patients who receive doses of cisplatin of 
50 mg per square meter of body-surface area without prophylactic antiemetics6; delayed emesis occurs in 57 to 89 percent, with maximal intensity on days 2 and 3 after chemotherapy.7,8,9,10,11
Efforts to prevent chemotherapy-induced emesis have been directed at blocking neurotransmitter receptors in the brain-stem vomiting center. The chemoreceptor trigger zone in the area postrema transmits impulses to the vomiting center after the administration of chemotherapeutic agents.12 Receptors for neurotransmitters such as dopamine, endorphin, serotonin, and substance P are found in these areas. Initial efforts focused on blocking dopamine with phenothiazines, butyrophenones, and metoclopramide (which was later found to bind the serotonin receptor).13 Each compound is partially effective when given as a single agent. Combinations of antiemetics, such as metoclopramide plus dexamethasone, were found to improve the prevention of chemotherapy-induced emesis significantly.14 The next major breakthrough occurred with the introduction of selective serotonin antagonists, such as ondansetron and granisetron, which have peripheral and central sites of action.
Although serotonin antagonists by themselves improve the prevention of acute chemotherapy-induced emesis,15,16,17 better control is achieved with the combination of a serotonin antagonist and dexamethasone, which completely prevents vomiting in 48 to 91 percent of patients.18,19,20 However, the serotonin antagonists are not as effective against delayed emesis.21,22 The most effective current approach to the prevention of delayed emesis is the combination of a serotonin antagonist or metoclopramide with dexamethasone, which completely prevents delayed emesis in 52 to 69 percent of patients.9,23 These regimens require multiple daily doses, and metoclopramide can cause sedation and extrapyramidal side effects. Thus, there is considerable opportunity for improvement in the prevention of delayed emesis.
Substance P is one of four mammalian tachykinins found in neurons, including vagal afferent fibers innervating the brain-stem nucleus tractus solitarii and area postrema. Exogenous substance P applied to cells in the nucleus tractus solitarii causes vomiting.24 The biologic actions of substance P are mediated through the neurokinin-1 receptor, a G-protein receptor coupled to the inositol phosphate signal-transduction pathway.25
L-754,030, a trisubstituted morpholine acetal, is a potent and selective nonpeptide neurokinin-1–receptor antagonist. Neurokinin-1–receptor antagonists inhibit emesis induced by cisplatin in ferrets, and central nervous system penetration is essential for the antiemetic activity of the drug.26,27,28 In ferrets, L-754,030 resulted in a dose-dependent inhibition of both acute and delayed cisplatin-induced vomiting.29 Moreover, additive efficacy was achieved by combining L-754,030 with either dexamethasone or ondansetron (Tattersall FD: personal communication). Preliminary observations suggested that similar benefits might be obtained with a neurokinin-1–receptor antagonist in patients.30
We undertook this study to evaluate a neurokinin-1–receptor antagonist, L-754,030, for the prevention of delayed emesis due to cisplatin; to determine whether L-754,030 enhances the acute antiemetic effect of the combination of a serotonin antagonist and dexamethasone; and to assess the safety of L-754,030.
Methods
Study Design
The subjects of this double-blind, multicenter, placebo-controlled study were men and women with cancer who had not previously been treated with cisplatin. The patients were assigned to one of three treatment groups according to a computer-generated randomization schedule that incorporated stratification according to sex and according to whether or not the patient was receiving additional highly emetogenic therapy, as previously defined.31 All the patients provided written informed consent.
In all three groups, the patients received an oral dose of dexamethasone (20 mg) and an intravenous dose of granisetron (10 µg per kilogram of body weight) 30 minutes before receiving a single dose of cisplatin (day 1). On day 1, the patients also received 400 mg of oral L-754,030 (groups 1 and 2) or placebo (group 3). The patients in group 1 also received 300 mg of L-754,030 once daily on days 2 through 5, whereas the patients in groups 2 and 3 received placebo on days 2 through 5. Rescue therapy for day 1 (20 to 30 mg of metoclopramide orally four times daily or 1 to 2 mg per kilogram intravenously four times daily) and days 2 to 5 (8 mg of dexamethasone orally twice daily with or without metoclopramide) was permitted at any time but was not given prophylactically. L-754,030 or placebo was given orally 2 hours before cisplatin infusion on day 1, approximately 24 hours after initiation of the cisplatin infusion on day 2, and on subsequent days between 8 and 10 a.m.
Eligibility
Patients were enrolled who were at least 18 years of age and who were scheduled to receive a first course of cisplatin at a dose of at least 70 mg per square meter. Women of child-bearing age had to have a negative test for the ß subunit of human chorionic gonadotropin in serum. The primary exclusion criteria included a Karnofsky score of less than 60; allergy to or intolerance of metoclopramide, dexamethasone, or granisetron; therapy with another antiemetic drug (serotonin antagonists, phenothiazines, butyrophenones, cannabinoids, metoclopramide, or glucocorticoids) within 72 hours before day 1; an episode of vomiting or retching within 24 hours before the start of the cisplatin infusion; treatment for or history of a seizure within the previous two years; severe concurrent illness other than cancer; gastrointestinal obstruction or active peptic ulcer; radiation therapy to the abdomen or pelvis within one week before or after day 1; or any of the following laboratory measurements: hemoglobin level below 8.5 g per deciliter, white-cell count below 3500 per cubic millimeter, platelet count below 100,000 per cubic millimeter, serum aspartate aminotransferase level more than twice the upper limit of normal, serum alanine aminotransferase level more than twice the upper limit of normal, serum bilirubin level more than twice the upper limit of normal, serum alkaline phosphatase level more than twice the upper limit of normal, serum albumin level below 3 g per deciliter, and serum creatinine level of more than 2 mg per deciliter (180 µmol per liter). Five patients who were scheduled to receive paclitaxel in addition to cisplatin were permitted to receive additional glucocorticoids before day 1.
Assessments
Episodes of vomiting or retching were recorded by the patients on diary cards. An emetic episode was defined as a single instance of vomiting or retching, or any number of continuous instances of vomiting or retching; distinct episodes were separated by at least one minute. The primary measure of efficacy was the proportion of patients without emesis in the delayed-emesis phase.
Nausea was assessed on a 100-mm horizontal visual-analogue scale in the patient diary with the heading "How much nausea have you had over the past 24 hours?" The left-hand end of the scale (0 mm) was labeled "no nausea," and the right-hand end (100 mm) was labeled "nausea as bad as it could be." Every 24 hours the patient indicated the degree of nausea during the previous 24 hours by placing a vertical mark on the scale line.
Global satisfaction with the antiemetic treatment was assessed on a 100-mm visual-analogue scale on the mornings of days 2 and 6 after cisplatin treatment. The scale for day 2 was headed "How satisfied are you with your anti-emetic treatment over the past 24 hours?" The scale for day 6 was headed "How satisfied are you with your anti-emetic treatment over the entire study period?" On both scales, 0 mm was labeled "not at all satisfied," and 100 mm was labeled "completely satisfied."
Adverse events were recorded up to the post-study visit, which occurred between day 17 and day 29. The patients also underwent laboratory safety studies (hematologic, serum chemical, and urine analyses), electrocardiography, and physical examinations between days 6 and 8 and between days 17 and 29.
Statistical Analysis
The statistical analysis of efficacy was performed on an intention-to-treat basis (all patients with data on emesis after the administration of L-754,030 were included). The incidence of vomiting in the acute-emesis and delayed-emesis phases and the use of rescue medication in both phases were evaluated. Fisher's exact test was used for pairwise comparison of the incidence of emesis (and no emesis or use of rescue therapy) between treatment groups. The proportions of patients with one or two and with three or more emetic episodes in the delayed-emesis phase were also summarized. A Bonferroni adjustment for multiplicity was used in testing for significance only in the analysis of emesis, since emesis was the primary end point. Nominal 95 percent confidence intervals are reported. A sample size of 43 patients per group provided the study with 80 percent power to detect a 30 percentage point increase or a 25 percentage point decrease as compared with placebo in the proportion of patients without vomiting in the delayed-emesis phase (two-sided test, alpha=0.05).
A secondary measure of efficacy was the patient's self-assessment of nausea. In the analysis for days 1 to 5 and 2 to 5, an average score was calculated for each patient from the values on the visual-analogue scale over the given interval; the analysis for day 2 used the rating recorded on day 2. Because the distribution of values was not normal, nonparametric analyses were performed on the ranked scores. For days 1 to 5 and days 2 to 5, the distributions of these average scores were compared among the treatment groups with use of the Kruskal–Wallis chi-square test, and median values for these distributions are reported. Pairwise comparisons were also made with use of the Wilcoxon test. In addition, Fisher's exact test was used to make pairwise comparisons of the proportions of patients who had no nausea or minimal nausea (defined post hoc as a rating on the visual-analogue scale that averaged less than 5 mm over the entire period) on days 1 to 5, days 2 to 5, and day 2 only.
An exploratory measure was the patient's global degree of satisfaction with the antiemetic therapy. The groups were compared with use of a general linear model and a nonparametric analysis of the ranked scores. The results of the two analyses were similar, and the medians are reported.
Results
Of the 159 patients who enrolled in this study, 1 patient in group 1 was excluded from both phases of the trial, because the patient vomited before the cisplatin infusion and no data on emesis were collected. Three others in group 1 were excluded from the analysis of the delayed-emesis phase; all three withdrew from the study and provided no data on emesis for the delayed-emesis phase, one because of inability to swallow the pills and two because of lack of efficacy in the acute phase. Two patients in group 3 received glucocorticoids during the study for reasons other than rescue medication but were included in the efficacy analysis. The base-line characteristics of the patients assigned to the three treatment groups were similar (Table 1).
|
During the acute-emesis period, the treatment that best prevented vomiting was triple therapy with granisetron, dexamethasone, and L-754,030 (groups 1 and 2). The proportion of patients who did not vomit was significantly higher in group 1 (93 percent), group 2 (94 percent), and in groups 1 and 2 combined (93 percent) than in group 3, which received the standard therapy of granisetron and dexamethasone (67 percent, P<0.001 for the comparison between groups 1 and 2 combined and group 3) (Figure 1). The difference in the proportions of patients with no emesis between groups 1 and 2 combined and group 3 was 26 percentage points (95 percent confidence interval, 12 to 43).
|
During the delayed-emesis period, the prevention of emesis was best achieved in the patients who received L-754,030 (groups 1 and 2). The prevention of delayed emesis in groups 1 and 2 was significantly superior to that in group 3, in which placebo was given during both the acute- and delayed-emesis phases (the proportions of patients without delayed emesis in groups 1, 2, and 3 were 82 percent, 78 percent, and 33 percent, respectively; P<0.001 for the comparison between group 1 or 2 and group 3) (Figure 1 and Table 2). The difference between groups 1 and 3 in the proportion of patients who did not vomit was 49 percentage points (95 percent confidence interval, 30 to 67); between groups 2 and 3, the difference was 45 percentage points (95 percent confidence interval, 26 to 62).
|
Assessment of Nausea
Figure 2 shows the median scores on the visual-analogue scale for nausea over time, and Table 3 gives these scores for days 1 to 5, days 2 to 5, and day 2. Over days 1 to 5 and days 2 to 5, the distribution of nausea scores was significantly lower in group 1 than in group 3 (P=0.003 for both comparisons). In the analysis for day 2, the nausea scores in groups 1 and 2 were significantly lower than those in group 3 (P= 0.002 for group 1 vs. group 3, and P=0.005 for group 2 vs. group 3). The proportions of patients with no nausea or minimal nausea in groups 1, 2, and 3 on days 1 to 5 were 49 percent, 48 percent, and 25 percent, respectively (P=0.02 for group 1 vs. group 3, and P=0.03 for group 2 vs. group 3); on days 2 to 5, the proportions were 51 percent, 48 percent, and 24 percent, respectively (P=0.007 for group 1 vs. group 3, and P=0.01 for group 2 vs. group 3); and on day 2 the proportions were 65 percent, 61 percent, and 41 percent, respectively (P=0.03 for group 1 vs. group 3).
|
|
There was no difference among the treatment groups in the global-satisfaction rating on day 2 (the medians for groups 1 and 2 combined and for group 3 were both 100 mm). The median ratings on day 6 for global satisfaction in groups 1, 2, and 3 were 100, 98, and 82 mm, respectively. Groups 1 and 2 had significantly better ratings than group 3 (P=0.001 for group 1 vs. group 3, and P=0.03 for group 2 vs. group 3).
Safety
All 159 patients who received study medication were included in the analysis of safety. Table 4 lists the most common adverse events through day 7. There were no significant differences in the incidence of these events among the three groups, nor were significant differences observed with respect to laboratory indexes of safety. The same pattern in clinical and laboratory adverse events emerged when the entire study period (day 1 through the last study visit between days 17 and 29) was considered, and there were no significant differences among the treatment groups.
|
The use of serotonin antagonists in combination with dexamethasone has greatly reduced the frequency of acute chemotherapy-induced emesis but has had less effect on delayed nausea and vomiting caused by highly emetogenic agents such as cisplatin.18,19,20,21,22,23 The results of our trial demonstrate that the standard therapy (granisetron plus dexamethasone) with the addition of the selective neurokinin-1 antagonist L-754,030 is superior to the standard therapy alone in preventing acute emesis, and that L-754,030 significantly reduces the frequency of vomiting and nausea in the delayed-emesis phase.
During the acute-emesis phase, the addition of L-754,030 to granisetron plus dexamethasone increased the proportion of patients who did not vomit by 26 percentage points. In addition, significantly fewer patients used rescue medication in the triple-therapy group. In the standard-therapy group, 67 percent of the patients did not vomit, a result similar to those of previous trials. In this group, 57 percent of the patients did not vomit and did not require rescue therapy, a result that may reflect the high expectations patients now have for control of nausea and vomiting, resulting in a lower threshold for rescue therapy.33,34,35
The proportion of patients with no delayed emesis in the groups given a single dose of L-754,030 on day 1 (group 2) or daily doses of L-754,030 (group 1) was approximately 50 percentage points greater than that in the group given placebo (group 3) and was up to 30 percentage points greater than the proportions reported with the most successful complex dual-therapy regimens (52 to 69 percent of patients given serotonin antagonists or metoclopramide with dexamethasone have no emesis).9,23 In addition, significantly fewer patients used rescue medication when receiving multiple or single doses of L-754,030.
In the acute-emesis phase, triple therapy did not significantly reduce the frequency of nausea as compared with standard therapy. However, the group receiving daily L-754,030 had significantly less nausea in the delayed-emesis phase and over the entire study period than the placebo group. In addition, the global satisfaction of patients with their antiemetic therapy was significantly greater with L-754,030. Use of L-754,030 should improve patients' quality of life, because nausea now outranks vomiting as the most unpleasant aspect of chemotherapy.1
During the delayed-emesis phase, the groups receiving single-dose therapy (group 2) and daily therapy (group 1) with L-754,030 did not differ significantly from each other in the proportion of patients who had no emesis. However, as compared with group 2, group 1 consistently had more patients who had no emesis and did not require rescue therapy, lower nausea scores, and higher global-satisfaction scores. Only a larger trial can determine whether continuation of L-754,030 treatment beyond day 1 yields an improvement that is both clinically and statistically significant. An argument could be made for testing the continuation of L-754,030 therapy at least through day 3, encompassing the period during which the incidence and severity of delayed nausea and vomiting are greatest.
L-754,030 (in either single or multiple doses) was generally well tolerated, with an incidence of clinical and laboratory adverse events that was similar to that with placebo.
In summary, the addition of L-754,030 to granisetron and dexamethasone treatment before chemotherapy provided better protection against vomiting in the acute-emesis phase than the combination of dexamethasone and a serotonin antagonist alone. Moreover, single and multiple doses of L-754,030 prevented delayed emesis after high-dose cisplatin therapy. A neurokinin-1–receptor antagonist such as L-754,030 represents a new type of agent acting through a novel mechanism to prevent vomiting in patients receiving highly emetogenic chemotherapy for cancer.
Presented in part at the 34th Annual Meeting of the American Society of Clinical Oncology, Los Angeles, May 16–19, 1998.
Source Information
From the Simon–Williamson Clinic, Birmingham, Ala. (R.M.N.); Merck Research Laboratories, Rahway, N.J. (R.R.R., A.D.C., B.J.G.); Ochsner Medical Center, New Orleans (R.J.G.); Memorial Sloan-Kettering Cancer Center, New York (M.G.K.); St. Elizabeth's Medical Center, Boston (P.J.H.); Capital Comprehensive Cancer Care Clinic, Jefferson City, Mo. (A.K.); Roswell Park Cancer Institute, Buffalo, N.Y. (H.K.); Salem Research Group, Winston-Salem, N.C. (T.H.G.); Research Medical Center, Kansas City, Mo. (K.P.); and Fletcher Allen Health Center, Burlington, Vt. (S.M.G.).
Address reprint requests to Dr. Gertz at Clinical Pharmacology, Merck Research Laboratories, RY33-600, Rahway, NJ 07065, or at barry_ gertz{at}merck.com.
References
The other members of the L-754,030 Antiemetic Trials Group were F.A. Bailey and J. Hankins, SORRA Research Center, Birmingham, Ala.; F.P. Arena and H. Gerstein, Great Neck, N.Y.; S. Luedke, St. Louis Center for Clinical Research, St. Louis; S.J. Yee, Arcadia, Calif.; M. Modiano, Arizona Clinical Research, Tucson; D.F. Roychowdhury, University of Cincinnati, Cincinnati; R.C. Shepard, Medical West Oncology, Chicopee, Mass.; D.P. Gray, Columbus Regional Hospital Cancer Care, Columbus, Ind.; J.A. Reeves, Leecoast Research Center, Fort Myers, Fla.; I. Royston, Sidney Kimmel Cancer Center, San Diego, Calif.; J. Scott, Georgia Cancer Specialists, Decatur; C.J. Badolato, Associated Medical Research, Melbourne, Fla.; and W.R. Edwards, Rockford Clinic, Rockford, Ill.
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Prevention of Cisplatin-Induced Emesis by a Neurokinin-1–Receptor Antagonist
Roila F., Ballatori E., Del Favero A., Navari R. M., Carides A. D., Gertz B. J.
Extract |
Full Text
N Engl J Med 1999;
340:1926-1928, Jun 17, 1999.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
