All other authors are compensated consultants for Merck and Company, Inc.
Prevention of cisplatin-induced acute and delayed emesis by the selective neurokinin-1 antagonists, L-758,298 and MK-869
A randomized controlled trial
Article first published online: 23 MAY 2002
Copyright © 2002 American Cancer Society
Volume 94, Issue 11, pages 3032–3041, 1 June 2002
How to Cite
Van Belle, S., Lichinitser, M. R., Navari, R. M., Garin, A. M., Decramer, M. L. A., Riviere, A., Thant, M., Brestan, E., Bui, B., Eldridge, K., De Smet, M., Michiels, N., Reinhardt, R. R., Carides, A. D., Evans, J. K. and Gertz, B. J. (2002), Prevention of cisplatin-induced acute and delayed emesis by the selective neurokinin-1 antagonists, L-758,298 and MK-869. Cancer, 94: 3032–3041. doi: 10.1002/cncr.10516
- Issue published online: 23 MAY 2002
- Article first published online: 23 MAY 2002
- Manuscript Accepted: 26 DEC 2001
- Manuscript Revised: 19 DEC 2001
- Manuscript Received: 17 SEP 2001
- Merck and Company, Inc.
- substance P;
- clinical trial
Recent studies have suggested that antiemetic therapy with a triple combination of the neurokinin-1 receptor antagonist MK-869, a serotonin (5-HT3) antagonist, and dexamethasone provides enhanced control of cisplatin-induced emesis compared with standard therapy regimens. The authors compared the antiemetic activity of a dual combination of MK-869 and dexamethasone with that of a standard dual combination of ondansetron and dexamethasone to characterize further the efficacy and tolerability profile of MK-869.
This was a multicenter, double-blind, randomized, active agent-controlled study of 177 cisplatin-naïve patients with malignant disease. On Day 1, MK-869 was given intravenously as its water-soluble prodrug, L-758,298. Patients were randomized to one of three groups as follows. Group I received L-758,298 100 mg intravenously (i.v.), then dexamethasone 20 mg i.v., and cisplatin ≥ 70 mg/m2 on Day 1 followed by 300 mg MK-869 (tablet) orally on Days 2–5; Group II received L-758,298 100 mg i.v., then dexamethasone 20 mg i.v., and cisplatin ≥ 70 mg/m2 on Day 1 followed by placebo on Days 2–5; and Group III received ondansetron 32 mg i.v., then dexamethasone 20 mg i.v., and cisplatin ≥ 70 mg/m2 on Day 1 followed by placebo on Days 2–5. Emesis was recorded over Days 1–5 in a diary. Nausea was assessed every 24 hours by visual analog scale. Additional medication was available for emesis or nausea at any time. The primary efficacy parameters of interest were the proportion of patients without emesis and the proportion without emesis or rescue therapy on Day 1 (acute phase) and on Days 2–5 (delayed phase).
No serious adverse events were attributed to L-758,298 or MK-869. On Day 1, the proportions of patients with no emesis and no use of rescue medication were 44% of patients in Group I, 36% of patients in Group II, 40% of patients in Groups I and II combined, and 83% of patients in Group III (P < 0.001 for Group III vs. the combined Groups I and II). The proportions of patients with no emesis and no use of rescue medication on Days 2–5 were 59% of patients in Group I, 46% of patients in Group II, and 38% of patients in Group III (P < 0.05 for Group I vs. Group III). The proportions of patients who were without emesis on Day 1 were 49% of patients in Group I, 47% of patients in Group II, and 84% of patients in Group III (P < 0.01 for Group I or II vs. Group III). On Days 2-5, however, the proportions of patients who were without emesis on Days 2–5 were 65% of patients in Group I, 61% of patients in Group II, and 41% of patients in Group III (P < 0.05 for Group I or II vs. Group III). Nausea scores in the acute phase were lower for Group III than for Group I, Group II, or Groups I and II combined (P < 0.05), although there was no significant difference among groups either for the delayed phase or overall for Days 1–5.
Although the L-758,298 and dexamethasone combination reduced acute (Day 1) emesis compared with historic rates, dual therapy with ondansetron and dexamethasone was superior in controlling acute emesis. Continued dosing with MK-869 may enhance control of other measures of delayed emesis, such as the use of rescue medication, although confirmation is required before a definitive conclusion may be drawn. MK-869 given as dual therapy with dexamethasone was superior to ondansetron with dexamethasone for the control of delayed emesis (Days 2–5) and control of the need for rescue medication on Days 2–5. Cancer 2002;94:3032–41. © 2002 American Cancer Society.
Patients with cancer consistently report that chemotherapy-induced emesis and associated nausea are among the most unpleasant and distressing aspects of treatment; in fact, patients may even delay or refuse potentially curative therapy because of severe chemotherapy-induced emesis. 1, 2 It has been shown that the specific chemotherapy, dose, and regimen of administration influence the severity and pattern of chemotherapy-induced emesis,3 with cisplatin most commonly associated with profound nausea and emesis. These symptoms tend to occur in a distinct, biphasic pattern. The acute phase generally is considered the first 24 hours (or Day 1) after the administration of chemotherapy, and the delayed phase is considered to be Days 2–5 postchemotherapy.4 Virtually all patients who receive cisplatin in doses ≥ 50 mg/m2 without prophylactic antiemetics experience severe acute-phase emesis,5 whereas 57–89% of patients experience delayed emesis,6–8 which occurs with maximal intensity on Days 2 and 3 after cisplatin chemotherapy.9, 10
The best control of acute emesis has been achieved with the combination of a serotonin (5-HT3) antagonist and dexamethasone, a regimen on which 48–91% of patients report no acute emesis. 11–13 Although this therapy represents a significant advance in the control of acute nausea and emesis, room for improvement still exists. Furthermore, the 5-HT3 antagonists are not highly effective in preventing delayed emesis,14, 15 for which the most effective therapy currently available is the combination of a 5-HT3 antagonist or metoclopramide with dexamethasone (52–69% of patients report no delayed emesis).8, 16 This combination of medications is disadvantageous in that it requires multiple doses per day and carries a risk of sedation and extrapyramidal side effects associated with metoclopramide.17 Therefore, a need remains for improved therapy for the prevention of both acute and delayed emesis.
The neurokinin 1 (NK1) receptor antagonists are a new class of antiemetic compounds. L-758,298 is the intravenous (i.v.) prodrug of MK-869 (previously known as L-754,030), a potent, selective, orally active, nonpeptide NK1 receptor antagonist. Previous publications have reported the efficacy of MK-869 in preclinical models as well as in patients when given as monotherapy and as part of triple-combination therapy with dexamethasone and a 5HT3 antagonist. 17–19 Monotherapy on Day 1 with L-758,298 was less effective numerically than monotherapy with ondansetron for the prevention of acute emesis, but there was a significant reduction in delayed emesis after the initial treatment with L-758,298 on Day 1.19 The triple combination was superior to the current standard dual therapy with a 5HT3 antagonist and dexamethasone for the prevention of both acute emesis and delayed emesis;18 in addition, continued dosing with the NK1 antagonist had a favorable impact on certain measures of delayed emesis, such as nausea ratings. However, although triple-combination therapy was shown to be more effective than standard dual therapy, it remained unclear whether the NK1 antagonist should be combined with both dexamethasone and the 5HT3 antagonist to achieve maximum reduction of acute emesis, or whether a dual combination of the NK1 antagonist and dexamethasone would suffice. This dual combination (using an oral formulation of the NK1 antagonist) was assessed preliminarily in the context of an additional triple-therapy study, the results of which suggested that the NK1 antagonist plus dexamethasone combination was superior to standard dual therapy in the prevention of delayed emesis but not acute emesis.20 The dual therapy study described here was designed to confirm those earlier results by assessing the tolerability and benefits of an NK1 antagonist (in i.v. formulation) in combination with dexamethasone for the prevention of acute and delayed cisplatin-induced emesis, as well as to evaluate the benefit of continued dosing with the oral formulation.
MATERIALS AND METHODS
This multicenter, double-blind, randomized, active agent-controlled (ondansetron plus dexamethasone) study was conducted in cisplatin-naïve male and female patients with cancer. The local ethical committee in each of the participating centers approved the study protocol. Patients, all of whom signed an informed consent form, were assigned to one of three treatment groups according to a computer-generated, randomized allocation schedule. Randomization was stratified both for gender and for treatment with additional highly emetogenic therapy based on a published categorization. 21
It is important to note that when this study was designed and performed (with the first patient in the study in June, 1997 and the last patient out of the study in March, 1998), consensus guidelines had not yet been established regarding the use of prophylaxis for chemotherapy-induced nausea and emesis. It is acknowledged that at the time these data are published, such guidelines have been incorporated into the conduct of oncology research 23 and include recommendations either for a 5HT3 antagonist and dexamethasone or for metoclopramide and dexamethasone in the delayed phase after patients receive high-dose cisplatin.
Patients ≥ 18 years old who were scheduled to receive their first course of cisplatin-based chemotherapy (≥ 70 mg/m2) were enrolled in the study. Female patients of reproductive potential demonstrated a serum β human chorionic gonadotropin level negative for pregnancy at the prestudy visit. Primary criteria for exclusion included a Karnofsky score < 60, allergy or intolerance to dexamethasone or ondansetron, current abuse of alcohol or use of illicit drugs, use of another antiemetic agent within 72 hours of Day 1 of the study (serotonin antagonists, phenothiazines, butyrophenones, cannabinoids, metoclopramide, corticosteroids, or lorazepam), an episode of emesis or retching within 24 hours prior to the start of the cisplatin infusion on Day 1 of the study, severe concurrent illness other than neoplasia, gastrointestinal obstruction or an active peptic ulcer, radiation therapy to the abdomen or pelvis within 1 week before or after Day 1 of the study, or one of the following laboratory measurements: hemoglobin < 8.5 g/dL; white blood cell (WBC) count < 3500/μL; platelets < 100,000/μL; alanine aminotransferase, aspartate aminotransferase, bilirubin, or alkaline phosphatase levels > 2 × the upper limit of normal; albumin < 3 g/dL; or serum creatinine > 2.0 mg/dL.
The schedule of treatments for each group is outlined in Table 1. On Day 1, patients in Groups I and II received L-758,298 (100 mg i.v.) infused over 15 minutes beginning 1 hour prior to cisplatin infusion. Patients in Group III received ondansetron (32 mg i.v.) infused over 15 minutes beginning 1 hour prior to cisplatin infusion. All patients received dexamethasone 20 mg i.v. 30 minutes prior to cisplatin (≥ 70 mg/m2).
|Group||Day 1||Days 2–5|
|One hour prechemotherapy||Thirty minutes prechemotherapy||Chemotherapy|
|I||L-758,298 (100 mg i.v.a)||Dexamethasone (20 mg i.v.)||Cisplatin (≥ 70 mg/m2 i.v.b)||MK-869 (300 mg orally q.d.)|
|II||L-758,298 (100 mg i.v.a)||Dexamethasone (20 mg i.v.)||Cisplatin (≥ 70 mg/m2 i.v.b)||Placebo (orally q.d.)|
|III||Ondansetron (32 mg i.v.a)||Dexamethasone (20 mg i.v.)||Cisplatin (≥ 70 mg/m2 i.v.b)||Placebo (orally q.d.)|
An oral formulation of MK-869 (for which L-758,298 is an i.v. prodrug) was used in the delayed phase, when patients were taking study medication on their own. On Days 2–5, patients in Group I received oral doses of MK-869 (300 mg tablet) once daily. On Day 2, the dose of MK-869 was taken approximately 24 hours after the initiation of the cisplatin infusion, and on subsequent days, the dose of MK-869 was taken between 8:00 AM and 10:00 AM.
Patients in Groups II and III received placebo on Days 2–5. Rescue therapy was available on an as-needed basis at any time for all patients.
Beginning immediately after infusion of chemotherapy and continuing for a total of 5 days, episodes of emesis or retching (date, time, and number of episodes) were recorded by the patients on diary cards. An emetic episode was defined as a single vomit or retch or any number of continuous vomits or retches. Distinct episodes were defined as those separated by at least 1 minute.
Nausea was assessed every 24 hours (from Day 1 to Day 5) by the patients and was recorded as a vertical mark in the diary on a 100-mm horizontal visual analogue scale (VAS) headed as follows: How much nausea have you had over the last 24 hours?, with 0 mm labeled on this scale as no nausea and 100 mm labeled nausea as bad as it could be.
Using a different 100-mm VAS, patients reported global satisfaction with their antiemetic treatment on the morning of Day 2 and the morning of Day 6 postcisplatin. The scale for Day 2 was headed as follows: How satisfied are you with your antiemetic treatment over the past 24 hours? The scale for Day 6 was headed as follows: How satisfied are you with your antiemetic treatment over the entire study period? On each scale, 0 mm was labeled not at all satisfied, and 100 mm was labeled completely satisfied. The VAS for nausea and the VAS for global satisfaction with antiemetic treatment were translated into the native languages of the participating centers and were validated before use.
Safety was evaluated by physical examinations, electrocardiograms, and laboratory safety studies (hematology, serum chemistry, and urinalysis) performed for each patient prior to chemotherapy, once during the interval between Day 6 and Day 8, and once again during the interval between Day 17 and Day 29. Adverse events other than episodes of emesis and nausea were recorded by the patients on diary cards from Day 1 to Day 5. Patients also recorded any adverse events, including episodes of emesis and nausea, from Day 6 to their poststudy visit (Days 17–29).
The analysis approach was intent to treat. All randomized patients who received at least one dose of study drug and had at least one post-treatment assessment were included in the efficacy analyses. For comparisons involving the incidence of emesis, the Fisher exact test was used, and frequencies were rounded to the nearest percent. A nominal 95%, exact, two-sided confidence interval was calculated for each of the relevant comparisons.
For comparison of the effect of the NK1 antagonist and dexamethasone combination with the ondansetron and dexamethasone combination on delayed emesis, a response rate of 35% for ondansetron and dexamethasone followed by 4 days of placebo was assumed, with sample sizes projected at 50 patients receiving L-758,298 on Day 1 followed by 4 days of MK-869, 50 patients receiving L-758,298 on Day 1 followed by 4 days of placebo, and 50 patients receiving ondansetron on Day 1 followed by 4 days of placebo. The study had 80% power to detect a 28 percentage point increase or a 24 percentage point decrease for either of the NK1 antagonist groups compared with the standard dual-therapy group in the percentage of patients with no emesis in the delayed phase. Specifically, the efficacy analyses focused on the following end points: the proportions of patients with no emesis and no use of rescue medication and the proportions of patients with no emesis in both the acute phase and the delayed phase; the proportions of patients with one or two emetic episodes in the delayed phase; and patients with three or more emetic episodes in the delayed phase.
In addition, patient self-assessment of nausea was evaluated as a secondary efficacy parameter. Nonparametric analyses were performed on the ranked scores for nausea ratings due to a non-normal distribution of values in each treatment group. An analysis for Day 1 used the VAS value recorded on Day 1, and a similar analysis for Day 2 used the VAS value recorded for that day. For analysis of the delayed phase (Days 2–5) and the overall cumulative interval of Days 1–5, an average score was computed for each patient using the VAS values recorded over those intervals. The Kruskal–Wallis chi-square test was used to compare the distributions of these average scores among the treatment groups for the two intervals, and the median values for these distributions are reported. The Wilcoxon test was used to perform pairwise comparisons. The Fisher exact test was used to make pairwise comparisons of the proportions of patients with no nausea or minimal nausea (defined post hoc as a VAS rating that averaged ≤ 5 mm over the entire given interval) for Day 2, for the delayed phase (Days 2–5), and for the overall cumulative interval of Days 1–5.
Patient global satisfaction with antiemetic therapy was evaluated as an exploratory parameter. Due to a non-normal distribution of values, nonparametric analyses were performed for Hour 24 and Day 6 using the ranked VAS global satisfaction scores for the corresponding time points. The distributions were compared among the three treatment groups using the Kruskal–Wallis chi-square test. In addition, pairwise comparisons were performed using the Wilcoxon test.
Safety analyses were performed by examination of laboratory parameters outside the normal range and by examination of between-group differences in the incidence of clinical adverse events. Summary statistics (the mean and standard deviation for each treatment group by time point) were calculated for relevant laboratory analytes, vital signs, and PR interval and QTc, and an analysis of variance was used to compare mean changes in laboratory values between groups.
In addition, at the end of the study, the numbers and proportions of patients in each treatment group with laboratory value changes from baseline outside predefined limits were tabulated. National Cancer Institute toxicity criteria guidelines 23 were used to tabulate the frequency with which laboratory values of interest fell into each toxicity grade per treatment group, and the numbers and proportions of patients in each group at each time point were reported. No formal statistical tests were performed on either of these tabulations.
Patient Characteristics and Inclusion in the Analysis
A total of 177 patients received an allocation number, and all patients received either L-758,298 with dexamethasone or ondansetron with dexamethasone. The baseline characteristics of the patients were similar across treatment groups (Table 2). Two patients were excluded from both the acute phase and the delayed phase efficacy analyses. One of these patients (from Group I) expired and the diary was lost, and the other patient (from Group I) discontinued due to an adverse laboratory experience that occurred after the patient received L-758,298 and dexamethasone; this patient did not receive cisplatin and did not complete the study. Two additional patients (one each in Groups I and II) were excluded from the delayed-phase analysis because they discontinued after the acute phase due to lack of efficacy, and incomplete data were collected for them. All 177 patients were included in the safety analysis.
|Characteristic||Group I (n = 61 patients)||Group II (n = 58 patients)||Group III (n = 58 patients)|
|Mean ± SD age (yrs)||59 ± 13||56 ± 13||59 ± 15|
|Alcohol drinks per week (%)|
|Mean ± SD cisplatin dose (mg/m2)||90 ± 15||87 ± 15||86 ± 16|
|Additional emetogenic chemotherapy (%)b||26||28||28|
|Type of cancer (%)|
|Head and neck||20||21||21|
Acute phase (Day 1)
The proportion of patients without emesis or use of rescue medication in Group III (83%) was higher than in either Group I (44%), Group II (36%), or the combined group (I and II), which received L-758,298 and dexamethasone (40%) on Day 1 (P < 0.001 for III vs. the combined group) (Fig. 1). The difference between Group III and the combined group was 43 percentage points (95% exact confidence interval, 28–57%). Furthermore, more patients in the combined group (38%) took rescue medication in the acute phase compared with Group III (9%) (P < 0.001).
Likewise, the proportion of patients without emesis in the acute phase (regardless of use of rescue medication) was significantly higher in Group III (84%) compared with Group I (49%), with Group II (47%), and with Groups I and II combined (48%; P < 0.001 for Group III vs. the combined group) (Table 3). The difference between Group III and the combined group was 36 percentage points (95% exact confidence interval, 22–52%).
|Treatment||Group I||Group II||Group III|
Delayed phase (Days 2–5)
Compared with the acute phase, better control during the delayed phase was achieved in those patients who received an NK1 receptor antagonist (Groups I and II) (Tables 3, 4; Fig. 1). Differences in the proportion of patients with neither emesis nor use of rescue medication favored the NK1 antagonist (59% in Group I, 46% in Group II, and 38% in Group III); the difference between Groups I and III was significantly different (P < 0.05). The difference was 21 percentage points between Groups I and III (95% exact confidence interval, 1–39%) and 8 percentage points between Groups II and III (95% exact confidence interval, −11–27%). Thus, Group I (which received the NK1 receptor antagonist on Day 1 and daily thereafter) had the highest proportion of patients with neither emesis nor use of rescue medication, although this was not statistically significantly higher compared with Group II, which received the NK1 antagonist on Day 1 only.
|Treatment||Group I (n = 58 patients)||Group II (n = 57 patients)||Group III (n = 58 patients)|
|Without emesis (%)||65b||61b||41|
|One to two emetic episodes (%)c||19||17||17|
|Three or more emetic episodes (%)c||15||21||41|
|Without emesis or use of rescue medication (%)||59b||46||38|
Similarly, the proportions of patients with no delayed emesis (regardless of use of rescue medication) in Group I (65%) and in Group II (61%) were significantly superior compared with the proportion of patients with no delayed emesis in Group III (41%; P < 0.05 for Group I or II vs. Group III) (Table 3). The difference in the proportion of patients without emesis between Groups I and III was 24 percentage points (95% exact confidence interval, 5–42%), and the difference between Groups II and III was 20 percentage points (95% exact confidence interval, 1–39%). Moreover, numerically fewer patients took rescue medication in Group I (26%) compared with Group III (36%), and the percentage of patients with 0–2 emetic episodes was significantly lower in Group I (84%) and in Group II (79%) compared with Group III (59%; P < 0.03 for Group I or II vs. Group III).
To assess whether effects in the delayed phase were related to efficacy in the acute phase, a post hoc analysis was performed, which showed that the percentage of patients who were without emesis in the delayed phase despite having had emesis in the acute phase was greater in the L-758,298 groups than in the ondansetron group (13 patients [22%] in Group I, 15 patients [26%] in Group II, and 0 patients in Group III). To check the consistency of this finding, the proportions of patients who reported no acute emesis but then experienced emesis in the delayed phase were also tabulated: Patients who had emesis only in the delayed phase were least common in Group I (the daily MK-869 group; 4 patients), whereas 7 such patients were noted in Group II (the single-dose L-758,298 group), and 25 such patients were noted in Group III (the ondansetron group).
For the acute phase, the distribution of nausea scores was lower in Group III (acute phase median VAS 1) compared with either Group I, Group II, or Groups I and II combined (acute phase median VAS 11; P < 0.05). However, for both the delayed phase and the overall cumulative interval, no significant difference in the distribution of nausea scores was observed among the three treatment groups (delayed-phase median VAS = 5 for Group I, 4 for Group II, and 1 for Group III; overall median VAS for Days 1–5 = 5 for Group I, 6 for Group II, and 1 for Group III). In considering patients who reported no nausea or minimal nausea (VAS average rating ≤ 5 mm) in the delayed phase, the relative frequencies across treatment groups were 38% in Group I, 46% in Group II, and 50% in Group III, with no significant differences observed among groups. Exploratory analysis of the overall cumulative interval (Days 1–5) showed frequencies of no nausea or minimal nausea of 33% in Group I, 37% in Group II, and 46% in Group III, with no significant difference seen among groups.
The median global satisfaction ratings 24 hours after cisplatin were 91 mm for Groups I and II combined and 99 mm for Group III (P < 0.05 for Group III vs. Groups I and II combined). Day 6 global satisfaction rating medians were 92 mm for Group I, 88 mm for Group II, and 96 mm for Group III, with no significant differences noted among the distribution of the scores in the 3 groups.
All 177 patients who received study medication were included in the tolerability analysis. A total of 30 serious clinical adverse events were reported, of which none was considered by any investigator to be related to study drug. Table 5 lists the most common adverse events (occurring in > 10% of patients in at least 1 treatment group) reported through Day 7, including various gastrointestinal symptoms, headache, dizziness, hiccups, abdominal pain, and asthenia. No significant differences were seen in the incidence of these adverse events among the treatment groups except for a higher incidence of diarrhea in Groups I and II (which did not receive ondansetron on Day 1).
|Adverse eventb||Group I (n = 61 patients)||Group II (n = 58 patients)||Group III (n = 58 patients)|
|Anorexia||6 (10)||7 (12)||5 (9)|
|Constipation||5 (8)||4 (7)||8 (14)|
|Diarrhea||14 (23)||13 (23)||3 (5)|
|Nausea||7 (11)||11 (19)||3 (5)|
|Dizziness||5 (8)||6 (11)||3 (5)|
|Headache||8 (13)||11 (19)||7 (12)|
|Hiccups||5 (8)||6 (11)||2 (4)|
|Asthenia||10 (16)||11 (19)||7 (12)|
|Pain, abdominal||5 (8)||4 (7)||6 (11)|
No significant differences in laboratory measures were observed among the treatment groups. Among patients who had normal or above-normal baseline values, fewer than 10% had transient decreases in WBC count (defined as WBC < 1.9 × 109/L and neutrophils/bands < 0.9 × 109/L; National Cancer Institute Grade 3 or 4 24). Similarly, fewer than 10% of patients had transient increases in transaminases (defined as at least 2.6 times the upper limit of the normal range in patients who had normal or below-normal baseline values; National Cancer Institute toxicity Grade 2, 3, or 423).
The introduction of NK1 antagonists as a novel class of antiemetic has shown promising initial results, especially in the prevention of delayed emesis. A comparison of L-758,298 and ondansetron given as monotherapies prior to cisplatin showed that L-758,298 was active against cisplatin-induced acute emesis and was superior to the 5HT3 antagonist in the prevention of delayed emesis. 19 Two previous trials also demonstrated the superiority of the triple combination of the NK1 antagonist with a 5HT3 antagonist and dexamethasone,17, 20 but it was not clear from those trials whether the dual combination of L-758,298 and dexamethasone required the addition of a 5HT3 antagonist for maximum effect. The present dual-therapy trial not only evaluated the NK1 antagonist versus a 5HT3 antagonist when each was given in combination with dexamethasone, but it also assessed the potential benefit of continued dosing with an NK1 antagonist beyond Day 1.
The results of the dual-therapy trial reported here reinforce the findings of previous studies 19, 20 by showing that patients who received L-758,298/L754,030 and dexamethasone had significantly better reduction of delayed emesis (in terms of complete response, which was defined as no emesis and no use of rescue therapy, and in terms of no emesis) compared with patients who received ondansetron and dexamethasone on Day 1. Consistent with previously observed trends,17, 19, 20 L-758,298 and dexamethasone also appeared to be active against acute emesis, although not to the same extent as the 5-HT3 antagonist and dexamethasone; the percentage of patients protected from acute emesis in the ondansetron and dexamethasone group (84%) was comparable to that found in previous trials24–26 and was superior to that observed in Groups I and II. In the delayed phase, however, L-758,298 followed by once-daily oral doses of MK-869 improved the proportion of patients who achieved complete response (59%) by 21 percentage points compared with the patients who received parallel treatment with ondansetron and dexamethasone followed by placebo (38%). In addition, L-758,298 followed by daily MK-869 improved the proportion of patients without emesis (65%) by 24 percentage points compared with patients who received ondansetron and dexamethasone followed by placebo (41%), results similar to the best reported results for delayed emesis with a complex dual-therapy regimen (5-HT3 antagonists or metoclopramide with dexamethasone, 52–69%).8, 16
Given the comparatively modest efficacy of L-758,298 and dexamethasone in preventing acute emesis, its more pronounced efficacy in preventing delayed emesis is notable and cannot be attributed merely to carry-over effects from the acute phase. Moreover, it has been shown that the plasma half-life of MK-869 is in the range of 10–17 hours (data on file), suggesting that the delayed-phase efficacy findings cannot be explained on a pharmacokinetic basis. Post hoc analyses lend additional support to these assertions by revealing a numeric trend toward better control of delayed emesis despite less control of acute emesis in patients who received MK-869 compared with patients who received ondansetron. In addition, analysis of the proportions of patients who reported no acute emesis but then experienced emesis in the delayed phase found the fewest of such patients in the daily MK-869 group (4 patients), whereas 7 such patients were noted in the single-dose L-758,298 group, and 25 such patients were noted in the ondansetron group. These findings, combined with the evidence of the superiority of the NK1 antagonist in the delayed phase despite lesser efficacy in the acute phase, suggest that control of acute emesis and control of delayed emesis do not necessarily represent a straightforward predictive relationship; this, in turn, is suggestive of different pathophysiologic mechanisms for acute and delayed emesis.
If, in fact, the mechanisms that mediate acute-phase and delayed-phase emesis are different, and, because an initial dose of the NK1 antagonist appears particularly useful in the prevention of delayed emesis, it follows that continued dosing with the NK1 antagonist into the delayed phase may provide even greater efficacy. In support of this reasoning, data from the first triple-therapy study have shown that patients who received an initial dose of MK-869 400 mg orally followed by daily doses of MK-869 300 mg orally in the delayed phase appeared to experience greater relief of delayed emesis compared with patients who received an initial dose of MK-869 400 mg orally followed by placebo in the delayed phase. 18 Likewise, in the present dual-therapy study, the proportion of patients in Group I (initial L-758,298 100 mg i.v. followed by daily MK-869 300 mg orally in the delayed phase) who reported no emesis in the delayed phase was numerically greater than the proportion in Group II (initial L-758,298 100 mg i.v. followed by daily placebo in the delayed phase). Considered in the context of a serial comparison of the three treatment groups, these data reveal a trend toward further improvement of delayed emesis with continued dosing with MK-869: The lowest proportion of patients with no delayed emesis was observed in Group III (initial ondansetron and daily placebo); the proportion in Group II (initial L-758, 298 and daily placebo) was significantly higher, and the percentage in Group I (initial L-758,298 and daily MK-869) was higher still. In both studies, this trend may have been even stronger if patients had not been given such a high initial dose of L-758,298 on Day 1. The possibility that the true magnitude of this enhancing effect may have been somewhat obscured in this study by the high initial dose suggests that the potential benefit of continued dosing may be more pronounced in studies using lower initial oral doses of MK-869.
A 13 percentage point numerical advantage also was observed in this study for the daily MK-869 group compared with the single-dose L-758,298 group in terms of the proportion of patients with complete response in the delayed phase; moreover, in terms of the number of patients who took rescue medication, the superiority of the daily MK-869 group approached statistical significance (26% in Group I vs. 42% in Group II; P < 0.1). Although the results of this study are not definitive, the trends in the data are supportive of the results of the triple-therapy trial 18 regarding the benefit of continued dosing with an NK1 antagonist in the delayed phase, a regimen that will be incorporated into future trials with MK-869.
Although emesis was the primary end point of this study, control of nausea also was examined. Because it is widely believed that response to treatment in the delayed period is determined to some extent by response in the acute period, patients in Groups I and II might be expected to have had worse nausea in the delayed phase than patients in Group III. However, although patients in Group III started out in the first 24 hours with better control of nausea compared with patients in Groups I and II, no such difference in treatment groups was observed in the delayed phase. This similarity of nausea ratings in the delayed phase across all groups indicates that the NK1 antagonist reduced nausea in the delayed phase despite its lesser effect on nausea in the acute phase. The use of MK-869, therefore, may result in an improvement in patient reported outcomes associated with improved control of delayed nausea, because nausea has superseded emesis as the most unpleasant and distressing aspect of chemotherapy. 1
MK-869 administered as an oral prodrug was generally well tolerated: Except for a higher incidence of diarrhea, the adverse events seen in Group III were similar to those seen in the groups that did not receive ondansetron (Groups I and II). This result likely reflects the loss of the constipating influence of a 5HT3 antagonist in patients who did not receive ondansetron; in the absence of antiemetics, it has been shown that cisplatin causes diarrhea in up to 60% of patients, 27 whereas prophylactic administration of a 5-HT3 antagonist has been associated with a lower incidence of cisplatin-related diarrhea.28 In normal volunteer studies, administration of MK-869 has not resulted in diarrhea (data on file; Merck Research Laboratories, Blue Bell, PA).
The findings of the dual-therapy study presented here support the use of MK-869 as a particularly effective agent for the control of delayed emesis in patients receiving highly emetogenic chemotherapy. These results also confirm that MK-869 combined solely with dexamethasone exerts suboptimal efficacy in the acute phase compared with standard therapy. This finding is consistent with and emphasizes the importance of previously published results showing that the triple combination of MK-869, a 5-HT3 receptor antagonist, and dexamethasone was more effective than dual combinations consisting of dexamethasone and one other agent. 17, 20 The results of the current study, combined with the results obtained from other studies of MK-869,17, 19, 20 suggest that the likely different underlying mechanisms of acute and delayed emesis may be addressed most effectively by administering triple therapy (NK1 antagonist, 5-HT3 antagonist, and dexamethasone) prior to initiation of chemotherapy for control of acute emesis, with control of delayed symptoms most likely optimized by continued dosing with an NK1 antagonist after Day 1.
- 23National Cancer Institute. Cancer therapy evaluation program. Common toxicity criteria, version 2.0. Bethesda, MD: National Cancer Institute, National Institutes of Health, Department of Health and Human Services, 1998.
- 28Medical Economics Company. Physician's desk reference. 52nd ed. Montvale: Medical Economics Company, 1998: 2836–2838.