Find synonyms Find exact match
INTRODUCTION — Few side effects of cancer treatment are more feared by the patient than nausea and vomiting. Although nausea and emesis (vomiting and/or retching) can result from surgery, opiates, or radiotherapy, chemotherapy-induced nausea and vomiting (CINV) is potentially the most severe and most distressing. Although significant progress has been made, CINV remains an important adverse effect of treatment.
Three distinct types of CINV have been defined, with important implications for both prevention and management:
●Acute emesis, which most commonly begins within one to two hours of chemotherapy and usually peaks in the first four to six hours
●Delayed emesis, occurring more than 24 hours after chemotherapy
●Anticipatory emesis, occurring prior to treatment as a conditioned response in patients who have developed significant nausea and vomiting during previous cycles of chemotherapy
The objective of antiemetic therapy is the complete prevention of CINV, and this should be achievable in the majority of patients receiving chemotherapy, even with highly emetic agents (table 1 and table 2). The three categories of drugs with the highest therapeutic index for the management CINV include type three 5-hydroxytryptamine (5-HT3) receptor antagonists, the neurokinin-1 (NK1) receptor antagonists aprepitant and fosaprepitant, and glucocorticoids (table 3) .
The use of these drugs alone and in combination for antiemetic prophylaxis in patients receiving cancer chemotherapy will be reviewed here. The pathophysiology of CINV is discussed separately. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting" and "Characteristics of antiemetic drugs".)
CHEMOTHERAPY DRUG EMETOGENICITY — The management of CINV has been greatly facilitated by the development of classification schemes that reflect the likelihood of emesis developing following treatment with particular agents. A 1997 classification scheme gained broad acceptance and was utilized as the basis for treatment recommendations by guideline panels .
●Highly emetic — >90 percent risk of emesis
●Moderately emetic — >30 to 90 percent risk of emesis
●Low emetogenicity— 10 to 30 percent risk of emesis
●Minimally emetic — <10 percent risk of emesis
This drug classification schema is utilized in both the updated antiemetic guidelines of the Multinational Association of Supportive Care in Cancer (MASCC) and the American Society of Clinical Oncology (ASCO) [4,5]. For combination regimens, the emetic level is determined by identifying the most emetic agent in the combination and then assessing the relative contribution of the other agents. As an example, cyclophosphamide and doxorubicin are both moderately emetogenic agents, but when given together, the regimen is highly emetic [2,4,5]. In updated antiemetic guidelines from ASCO, combined anthracycline and cyclophosphamide regimens have been reclassified as highly emetic .
ACUTE EMESIS — Extensive clinical trials have evaluated the 5-HT3 receptor antagonists, the NK1 receptor antagonists, and glucocorticoids in patients with acute and delayed CINV. These trials have focused primarily on patients receiving either highly or moderately emetic chemotherapy regimens. Although not all antiemetic regimens have been evaluated with all chemotherapy combinations, it is reasonable to extrapolate data to other chemotherapy regimens of comparable emetogenicity.
5-HT3 receptor antagonists — A key advance in the prevention of CINV was the development of selective type three 5-hydroxytryptamine (5-HT3) receptor antagonists, a drug class that has a high therapeutic index for prevention of CINV .
Randomized trials have shown that single-agent 5-HT3 receptor antagonists are more effective than less specific agents such as high-dose metoclopramide and as effective as the combination of high-dose metoclopramide and dexamethasone. When 5-HT3 antagonists are used in combination with dexamethasone, they are more effective than high-dose metoclopramide plus dexamethasone [7-10]. In addition to increased efficacy, these agents are easier to administer and are associated with significantly fewer serious side effects than the less specific serotonin inhibitor metoclopramide.
Four first-generation 5-HT3 receptor antagonists (dolasetron, granisetron, ondansetron, and tropisetron) and one second-generation agent (palonosetron) are available (table 3). An orally disintegrating formulation of ondansetron also is available that disperses rapidly when placed on the tongue and does not need to be swallowed with water . This formulation may be particularly useful for patients with dysphagia or anorexia. A granisetron transdermal system is also available. (See 'Granisetron transdermal patch' below.)
First generation agents — A large number of randomized trials have clarified the properties of the first-generation 5-HT3 receptor antagonists. Key findings include the following:
●The first-generation 5-HT3 receptor antagonists all appear equally effective at preventing CINV at the recommended doses. A meta-analysis has shown no clear advantage for either ondansetron or granisetron in the prophylaxis of acute or delayed emesis .
●There is a plateau in therapeutic efficacy at a definable dose level for each drug, and further dose escalation does not improve outcome .
●A single dose of a 5-HT3 receptor antagonist prior to chemotherapy is therapeutically equivalent to a multiple dose schedule [14-18].
●The efficacy of 5-HT3 receptor antagonists is significantly improved when they are combined with glucocorticoids. (See 'Glucocorticoids' below.)
●Oral formulations of these agents are as effective as intravenous formulations [13,19,20].
EKG interval changes and cardiac arrhythmias — EKG interval changes are a class effect of the first-generation 5-HT3 antagonists, including ondansetron, granisetron and dolasetron, although they have not been reported with transdermal granisetron . (See 'Granisetron transdermal patch' below.) EKG interval changes appear to be most prominent one to two hours after a dose of these agents, are mostly small and clinically insignificant, and return to baseline within 24 hours [22-24]. However, potentially fatal cardiac arrhythmias, including torsade to pointes, have been reported in association with QTc prolongation [22,24-26]. The following sections describe the warnings/precautions regarding cardiotoxicity of these agents from the US FDA.
Dolasetron — Due to the risk of QTc prolongation from increased drug exposure the injection form of dolasetron is contraindicated for prophylaxis of CINV in both children and adults . The risk of developing an abnormal heart rhythm with oral dolasetron is less than that seen with the injection form. However, there is still a potential risk.
●Potassium and magnesium levels should be assessed, and if abnormal, corrected before initiation of treatment with dolasetron. These electrolytes should be monitored after administration as clinically indicated.
●Use electrocardiographic monitoring in patients with heart failure, a slow heart rate, underlying cardiac disease, the elderly, and in patients with renal impairment.
Ondansetron — FDA has issued a warning about QTc prolongation and potentially fatal cardiac arrhythmias in patients treated with ondansetron . QT prolongation occurs in a dose-dependent manner and specifically at a single IV dose of 32 mg. Revised labeling includes a recommendation to limit single IV doses to no more than 16 mg, avoid use of ondansetron in patients with congenital long-QT syndrome, and to use ECG monitoring in certain patients, including those with hypokalemia or hypomagnesemia, heart failure, bradyarrhythmias, and in patients taking other medications that increase the risk of QTc prolongation (table 5).
Palonosetron — The second-generation agent palonosetron has a 30- to 100-fold higher affinity for the 5-HT3 receptor and a significantly longer half-life (40 hours) compared to first generation 5-HT3 receptor antagonists (table 3). In contrast to first-generation 5-HT3 antagonists, QTc prolongation has not been described with palonosetron .
As a single agent, palonosetron is more effective than ondansetron or dolasetron at preventing emesis due to moderately emetic chemotherapy [30-32]. This was illustrated by a multicenter trial in 592 patients, the majority of whom received doxorubicin and cyclophosphamide for breast cancer. Subjects were randomly assigned to a single IV dose of palonosetron at one of two dose levels (0.25 or 0.75 mg IV) or dolasetron (100 mg) . More patients treated with palonosetron (0.25 mg) had complete control of both acute (63 versus 53 percent) and delayed emesis (54 versus 39) compared to dolasetron. A dose of 0.75 mg was not significantly superior compared to 0.25 mg.
When used in combination with glucocorticoids, palonosetron provides superior control of delayed emesis compared to first-generation 5-HT3 receptor antagonists combined with glucocorticoids:
●In a phase III double-blind, double-dummy trial, 1143 patients receiving cisplatin or an anthracycline/cyclophosphamide combination were randomly assigned to dexamethasone plus either palonosetron or granisetron on day 1 prior to chemotherapy followed by dexamethasone alone on days 2 and 3 . During the acute phase, the rate of complete control of CINV was similar (75 versus 73 percent with palonosetron and granisetron, respectively), but during the delayed phase (24 to 120 hours), complete responses occurred in significantly more patients receiving palonosetron (57 versus 45 percent). An unresolved question arising from this study design is whether the efficacy differences noted would have persisted with the addition of aprepitant, which all evidence-based guidelines recommend in this setting.
●In a second phase III trial, in which 667 patients receiving cisplatin-based chemotherapy were randomly assigned to palonosetron (0.25 mg), palonosetron (0.75 mg) or ondansetron (32 mg), no significant differences in antiemetic control were noted between palonosetron and ondansetron . Approximately two-thirds of patients received concomitant dexamethasone. In this subset of patients, complete response rates were numerically higher in both palonosetron arms compared to ondansetron during the first 24 hours. During the delayed (24 to 120 hours) phase, complete response was significantly higher on the 0.25 mg palonosetron arm compared to the ondansetron arm (42 versus 29 percent, p = 0.021).
Updated antiemetic guidelines from the American Society of Clinical Oncology (ASCO) recommend palonosetron as the preferred 5-HT3 antagonist for patients who receive moderately emetic chemotherapy .
A non-inferiority trial documented similarity between the oral and IV formulations and validated the correct dose (0.5 mg oral) (table 3) . However, an oral formulation of palonosetron, which was approved by the FDA in 2008, was discontinued by the manufacturer and is not available anywhere.
Neurokinin-1 receptor antagonists — The introduction of the NK1 receptor (NK1R) antagonists aprepitant and fosaprepitant (a parenteral water-soluble prodrug of aprepitant that is effective as a one-day treatment (see 'One versus three-day administration' below)) have significantly improved the ability to prevent both acute and delayed CINV in patients receiving highly and moderately emetic chemotherapy (table 1 and table 2). Casopitant is another NK1R antagonist that can be given as a single day-one oral dose or in a mixed intravenous plus oral three day schedule in conjunction with dexamethasone and ondansetron; it is not yet commercially available in any country.
Efficacy — The benefit of combining an NK1R antagonist (aprepitant, fosaprepitant, or casopitant) with an 5-HT3 receptor antagonist plus a glucocorticoid for the prevention of acute CINV was addressed in a meta-analysis of 17 trials, totaling 8740 patients who were receiving highly or moderately emetogenic chemotherapy . The addition of a NK1R antagonist to standard antiemetic therapy significantly improved the rate of complete response (CR, absence of emesis and no need for rescue antiemetics) in both the overall phase (during the first 120 hours of chemotherapy, 72 versus 54 percent, OR 0.51, 95% CI 0.46-0.57) and in the acute (first 24 hours, OR 0.56, 95% CI 0.48-0.65) as well as delayed phase (OR 0.48, 95% CI 0.42-0.56). For other secondary outcomes (rate of emesis, no nausea, no substantial nausea in all phases), the addition of an NK1R antagonist was also superior to the control arm. (See 'NK1 receptor antagonists' below.)
In subgroup analyses, benefit was seen for both highly emetogenic (CR, 73 versus 54 percent, OR 0.46, 95% CI 0.40-0.53) and moderately emetogenic chemotherapy (CR 71 versus 54 percent, OR 0.59, 95% CI 0.61-0.67). There appeared to be no differences in treatment efficacy for aprepitant/fosaprepitant and casopitant. The use of an NK1R antagonist did not increase the risk of diarrhea, although rates of hiccups and fatigue/asthenia were significantly higher. There was a suggestion that use of an NK1R antagonist increased the risk of a severe infection (6 versus 2 percent in a pooled analysis of three trials); however, this was not associated with an increased rate of neutropenia or febrile neutropenia.
Need for a 5-HT3 agent — Aprepitant and fosaprepitant improve control of CINV when combined with a 5-HT3 receptor antagonist and dexamethasone. Aprepitant plus dexamethasone alone is not as effective as the three-drug combination regimen. A 5-HT3 receptor antagonist remains necessary, at least in patients receiving cisplatin-based chemotherapy.
This was illustrated by a randomized trial in which patients receiving cisplatin chemotherapy were randomly assigned to the combination of aprepitant plus granisetron, granisetron, or aprepitant (on one of two schedules) . All patients also received dexamethasone (20 mg orally) before cisplatin. While the three-drug combination blocked emesis in 80 percent of patients, dexamethasone plus either granisetron or aprepitant was effective in only 57 percent and 43 to 46 percent of cases, respectively.
One versus three-day administration — In the United States, both aprepitant and fosaprepitant are approved for use in three-day schedules. However, a single-day dosing schedule for fosaprepitant was approved by the FDA based upon the results of a phase III trial involving 2247 patients receiving single-day cisplatin (>70 mg/m2) based chemotherapy . The control group received aprepitant administered in the standard three-day schedule along with ondansetron plus dexamethasone; this was compared to a single 150 mg dose of fosaprepitant combined with ondansetron on day 1 plus dexamethasone on days 2, 3, and 4. Complete antiemetic response rates were nearly identical between the aprepitant and fosaprepitant arms (72.3 versus 71.9 percent).
Issues related to inhibition of CYP3A4 — NK1 receptor antagonists such as aprepitant and fosaprepitant are moderate inhibitors of the cytochrome P450 enzyme CYP3A4, which is particularly important in drug metabolism .
CYP3A4 is responsible for the metabolism of glucocorticoids, and thus the dose of dexamethasone was reduced in clinical trials from 20 mg to 12 mg on day 1 and from 8 mg twice daily to 8 mg daily on days 2 and 3 when given concurrently with aprepitant [40-42]. This dose reduction applies only when glucocorticoids are used as antiemetics in conjunction with NK1 receptor antagonists, not when given as an antitumor component of a chemotherapy regimen .
Theoretically, aprepitant could decrease the clearance of drugs metabolized by CYP3A4 (cyclophosphamide, docetaxel, etoposide, irinotecan, vinca alkaloids), resulting in prolonged exposure and increased toxicity. However, there is no clinical evidence that this actually occurs [42,43].
Glucocorticoids — Short courses of glucocorticoids are widely used both as single agents for regimens with low risk of causing CINV and in combination with 5-HT3 receptor inhibitors and/or NK1 receptor antagonists for more emetic chemotherapy regimens. When used in this fashion, glucocorticoids have a high therapeutic index. Although the various glucocorticoids are probably equally effective when used at an appropriate dose, dexamethasone has been the most extensively evaluated and is the most widely used.
Single agent — Single agent dexamethasone has been compared to either placebo or no treatment in a number of randomized trials. A meta-analysis of 32 randomized trials evaluated 5613 patients who received moderately or highly emetogenic chemotherapy . Dexamethasone was superior to placebo or no treatment for complete protection from both acute emesis (risk ratio [RR] 1.30) and delayed emesis (RR 1.30). However, dexamethasone as a single agent is insufficient to control CINV in most of these patients .
Combination with a 5-HT3 antagonist — Glucocorticoids alone represent insufficient first-line therapy for patients receiving either moderate or highly emetic chemotherapy agents. However, the antiemetic efficacy of the 5-HT3 receptor antagonists is significantly enhanced by the addition of a glucocorticoid [7,45-49].
Benefit for combined therapy was shown in a meta-analysis of 3791 patients enrolled in 22 randomized trials in which a 5-HT3 receptor antagonist plus dexamethasone was compared with a 5-HT3 antagonist plus placebo or no treatment in patients receiving moderate or highly emetic chemotherapy . The pooled risk ratio for emesis protection was 1.25, indicating that the addition of dexamethasone increased the chance of no acute vomiting by 25 percent.
Dose — The impact of glucocorticoid dose was explored in a double-blind trial that randomly assigned 531 patients receiving cisplatin ≥50 mg/m2 to one of four intravenous doses of dexamethasone administered by a 15-minute infusion prior to cisplatin administration . All patients received 8 mg of ondansetron as well. At doses of 20, 12, 8, and 4 mg, complete protection from vomiting was achieved in 83, 79, 69, and 69 percent of patients, respectively, and nausea was prevented in 71, 67, 61, and 61 percent.
The optimal dose of dexamethasone for highly to moderately emetic chemotherapy not containing cisplatin was evaluated by the Italian Group for Antiemetic Research . In this trial, all patients received IV ondansetron and were randomized to one of three schedules of dexamethasone following chemotherapy (either 8 or 24 mg IV prior to chemotherapy, or 8 mg IV before treatment followed by 4 mg every six hours). Rates of complete protection from acute or delayed emesis were similar among the groups, and the authors concluded that a single 8 mg IV dose prior to chemotherapy represented the appropriate dexamethasone regimen.
Other agents — Other agents that have been used in the treatment or prevention of CINV include phenothiazines (eg, prochlorperazine) metoclopramide, butyrophenones, and cannabinoids. These agents have a lower therapeutic index than the 5-HT3 receptor antagonists, aprepitant, and glucocorticoids for highly or moderately emetogenic chemotherapy regimens. Their use should be restricted to patients who are intolerant of or refractory to these first line agents. The benefits of synthetic oral cannabinoids in this setting remain controversial given the lack of evidence on their safety and efficacy [52-56]. Phenothiazines could be used as an alternative to single agent dexamethasone for those receiving chemotherapy with a low risk of emesis, if a glucocorticoid is contraindicated . (See "Characteristics of antiemetic drugs" and 'Poor emesis control' below.)
DELAYED EMESIS — Delayed emesis is defined by its occurrence more than 24 hours after chemotherapy. Although it is most common following high-dose cisplatin [57-59], delayed emesis may occur with other agents as well .
Regimens with a high risk of delayed emesis — The risk of delayed emesis after cisplatin (doses >70 mg/m2) ranges between 60 and 90 percent in the absence of effective prophylaxis. The risk of delayed emesis without any prophylaxis is estimated to be between 20 and 30 percent in patients receiving chemotherapy with an anthracycline plus cyclophosphamide .
Although the risk of delayed emesis has been best studied with high-dose cisplatin and the combination of doxorubicin plus cyclophosphamide, other moderately emetogenic agents are also associated with delayed emesis. These include doxorubicin ≥40 mg/m2 as a single agent or ≥25 mg/m2 in combination with other chemotherapeutic agents (especially cyclophosphamide), epirubicin ≥75 mg/m2 as a single agent or ≥50 mg/m2 when given in combination with other agents, combinations of cyclophosphamide ≥600 mg/m2 in combination with other drugs, carboplatin ≥300 mg/m2, oxaliplatin (as used in the FOLFOX [oxaliplatin plus short-term infusional fluorouracil and leucovorin] regimen for advanced colorectal cancer), and cisplatin at doses between 20 and 50 mg/m2 [18,61-64]. One study found that among 68 patients treated with one of these regimens who had no post-chemotherapy vomiting in the 24 hours after administration of prechemotherapy ondansetron and dexamethasone, 28 (41 percent) vomited in the next four days when no further antiemetics were given . This frequency was reduced to 15 of 75 (20 percent) when ondansetron was continued.
Management — Many of the regimens associated with delayed emesis are classified as high-emetic risk, and guidelines from ASCO recommend the use of an NK1 receptor antagonist (either aprepitant on days 1 to 3 or fosaprepitant on day 1 only), plus a glucocorticoid on days 1 to 4, and a 5-HT3 receptor antagonist on day 1 (table 6) . This regimen is effective against both acute and delayed emesis. The data supporting the individual components of this regimen are reviewed below.
Glucocorticoids — The value of maintenance therapy with oral glucocorticoids following treatment of acute CINV was shown in patients treated with cisplatin-based chemotherapy regimens in at least two randomized trials [57,59]. However, glucocorticoids alone are often not sufficient to prevent delayed emesis in patients receiving cisplatin.
NK1 receptor antagonists — The benefit of adding an NK1R antagonist to dexamethasone and a 5-HT3 antagonist for prevention of delayed emesis in patients receiving highly or moderately emetogenic chemotherapy was shown in a meta-analysis of 17 trials and described above . Importantly, the addition of NK1R antagonists increased control rates in the overall phase (ie, 120 hours after chemotherapy) independently of whether ondansetron was used in the control arm beyond day 1 or not. However, patients who did not use ondansetron after day 1 seemed to have a greater benefit from the addition of the NK1R antagonist. (See 'Neurokinin-1 receptor antagonists' above.)
Versus glucocorticoids in patients receiving AC — As noted above, the doxorubicin plus cyclophosphamide (AC) regimen used in patients with breast cancer is considered to have a high risk of both acute and delayed emesis. As such, ASCO  and NCCN guidelines recommend aprepitant plus dexamethasone for the delayed period (dexamethasone alone if fosaprepitant is used on day 1). These recommendations represent an extrapolation from data from cisplatin trials, in which aprepitant plus dexamethasone in the delayed phase was better than dexamethasone alone . The role of dexamethasone only in the delayed phase was directly addressed in a trial in which patients receiving AC for breast cancer were randomly assigned to oral dexamethasone (4 mg twice daily) or aprepitant (80 mg once daily) on days 2 and 3; all patients received IV palonosetron 0.25 mg, dexamethasone 8 mg, and aprepitant 125 mg on day 1 . From days 2 to 5, the primary end point, complete response rates (defined as no vomiting and no rescue treatment) were the same with both antiemetic regimens (79.5 percent), as were secondary endpoints (which included no nausea). The incidence rates of insomnia (2.9 versus 0.4 percent) and heartburn (8.1 versus 3.6 percent) were significantly greater with dexamethasone on days 2 to 5.
These data suggest that for patients receiving aprepitant, a 5-HT3 antagonist, and dexamethasone therapy on day 1, dexamethasone is as effective in the delayed period as is aprepitant, and it could be considered in such patients, particularly in resource-limited settings. What is not clear is whether the combination of delayed phase aprepitant plus dexamethasone, or fosaprepitant on day 1 followed by delayed phase dexamethasone is even better than aprepitant on day 1 followed by dexamethasone alone for the delayed period.
5-HT3 antagonists — Conflicting results have been described with the use of first generation 5-HT3 receptor antagonists as single agents for protection against delayed emesis [57,66-70]. Although some benefit has been seen when these agents are used as monotherapy, the benefit was not as great as that seen with corticosteroids. Furthermore, continuing a 5-HT3 receptor antagonist beyond 24 hours along with glucocorticoids did not confer additional benefit compared to corticosteroids alone. Thus the use of the 5-HT3 receptor antagonists as a sole maneuver to prevent delayed emesis in patients receiving cisplatin is not recommended.
●In a phase III trial of patients receiving cisplatin-based chemotherapy, palonosetron (at one of two doses, either 0.25 mg or 0.75 mg) yielded higher rates of emetic control compared to ondansetron in preventing delayed emesis (complete response rates of 45 and 48 versus 39 percent, respectively), although the results were not statistically significant . However, in a subset analysis, palonosetron 0.25 mg was superior to ondansetron for control of delayed and overall emesis in patients receiving concomitant dexamethasone.
●The superiority of palonosetron for prevention of delayed emesis was also shown in another phase III trial in which 1114 patients receiving cisplatin or an anthracycline/cyclophosphamide (AC) combination were randomly assigned to a single dose of palonosetron or granisetron 30 minutes prior to chemotherapy with all patients receiving dexamethasone for three days . Significantly better control of delayed emesis was achieved in both the cisplatin and AC subgroups on the palonosetron arm (complete response 57 versus 45 percent with granisetron).
Although these results suggest the superiority of palonosetron over other 5-HT3 antagonists for the prevention of delayed emesis, it is unclear whether this difference would persist with the addition of an NK1 receptor antagonist. To date, there are no randomized trials in which the combination of palonosetron and an NK1 receptor antagonist has been compared to a first generation 5-HT3 receptor antagonist plus an NK1 receptor antagonist.
Olanzapine — Conventional antiemetics are more successful at preventing emesis than in preventing nausea, particularly delayed nausea. The superiority of the antipsychotic olanzapine over aprepitant for the prevention of delayed nausea was suggested in a phase III trial conducted in 247 patients receiving cisplatin or doxorubicin plus cyclophosphamide . Patients were randomly assigned to olanzapine (10 mg orally on the day of chemotherapy, and then 10 mg once daily on days 2 through 4), or aprepitant (125 mg orally prior to chemotherapy, followed by 80 mg orally on days 2 and 3), both in combination with palonosetron (0.25 IV on the day of chemotherapy) plus dexamethasone. Dexamethasone (20 mg) was only given on the day of chemotherapy, while the aprepitant group received dexamethasone 12 mg on the day of chemotherapy followed by dexamethasone 4 mg twice daily on days 2 and 3.
Following cycle 1, rates of prevention of acute nausea (87 percent in both the olanzapine and aprepitant groups), and of complete control of acute (97 versus 87 percent) and delayed vomiting (77 versus 73 percent) were similar. However, patients treated with olanzapine had a significantly higher rate of nausea control in the delayed period (69 versus 38 percent). The results were maintained during cycles 2 to 4.
Moderate risk agents — The incidence of delayed emesis following treatment with moderate risk agents other than the combination of cyclophosphamide plus an anthracycline is not well characterized. Other cyclophosphamide or doxorubicin-based regimens, as well as carboplatin and oxaliplatin, can cause delayed emesis. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Delayed emesis'.)
Glucocorticoids are also consistently useful agents. The value of maintenance dexamethasone was demonstrated in a randomized placebo controlled trial of patients undergoing cyclophosphamide-based chemotherapy . In this trial, 98 patients were given granisetron and dexamethasone before chemotherapy and randomized to receive either oral dexamethasone 4 mg twice daily as maintenance or no maintenance. Maintenance dexamethasone was associated with a higher rate of compete (57 versus 33 percent) and major control (33 versus 15 percent) of delayed emesis.
The 5-HT3 receptor antagonists also have activity as single agents for delayed emesis with cyclophosphamide-based chemotherapy . However, there is no evidence that they are superior to dexamethasone alone or that combination therapy with dexamethasone is superior to dexamethasone alone [60,61].
A randomized trial of 708 patients receiving moderately emetogenic chemotherapy and concurrent antiemetics illustrates the relative roles of dexamethasone and 5-HT3 receptor antagonists in the management of delayed emesis :
●Patients without acute nausea or vomiting (the low-risk group) were randomized to receive dexamethasone (4 mg orally twice daily on days 2 through 5) plus ondansetron (8 mg orally twice daily on days 2 through 5), dexamethasone alone on the same schedule, or a placebo. Among these 618 patients, there was a complete absence of delayed nausea and vomiting in 92, 87, and 77 percent of patients in the combined therapy, dexamethasone, and placebo groups, respectively. Protection with dexamethasone alone or with dexamethasone plus ondansetron was better than that with placebo; however, the combination was not statistically superior to dexamethasone alone.
●The key factor in preventing delayed emesis was the control of acute symptoms following chemotherapy. Patients who had either vomiting or moderate to severe nausea in the 24 hours following chemotherapy constituted a high-risk group. These 87 patients were randomly assigned to oral dexamethasone alone or in combination with ondansetron, at the same doses and schedules as in the low-risk group. Despite treatment, complete protection from delayed emesis or moderate to severe was achieved in only 41 and 23 percent of patients treated with the combination and dexamethasone, respectively.
The relative roles of palonosetron and aprepitant in controlling delayed nausea was studied in a randomized trial of 944 evaluable patients receiving primarily (95 percent) moderately emetogenic chemotherapy . Palonosetron did not provide superior control of delayed nausea compared with granisetron, when both were provided on day 1 with dexamethasone, and prochlorperazine was administered on days 2 and 3. In addition, aprepitant was not more effective than prochlorperazine when both were combined with dexamethasone on days 2 and 3.
ANTICIPATORY EMESIS — Anticipatory emesis is a conditioned response that occurs in patients who experienced severe nausea and vomiting during prior cycles of chemotherapy . It appears to be induced by sensory cues and cognitive anticipation of subsequent chemotherapy. Anticipatory nausea has also been described among patients who have a high expectation of developing nausea despite never having received chemotherapy .
The most effective means to prevent anticipatory nausea or emesis is to ensure good control of acute and delayed emesis, starting from the initial chemotherapy cycle (table 6). Once anticipatory emesis has been established, nonpharmacologic methods (eg, hypnosis, behavioral therapy with systemic desensitization) may be effective [75-77]. (See "Complementary and alternative therapies for cancer".)
Although few formal trials have been carried out, benzodiazepines before and during chemotherapy may be useful [78-80]. In one double-blind trial of 57 women undergoing adjuvant chemotherapy for primary breast cancer for example, the addition of low-dose alprazolam (0.5 mg to 2 mg/day) to a psychologic support program including progressive relaxation training was associated with a significantly reduced rate of anticipatory nausea compared to placebo (0 versus 18 percent) .
Consecutive day therapy with highly emetogenic agents — When moderately or highly emetogenic chemotherapy agents (eg, cisplatin, dacarbazine) are administered on several consecutive days, prophylaxis is more difficult. This may be due to anticipatory emesis on the subsequent days of therapy or to the compounding of acute and delayed effects of treatment. Updated ASCO antiemetic guidelines suggest that antiemetics appropriate for the emetogenic risk class of the chemotherapy be administered for each day of the chemotherapy and for two days after, if appropriate .
Trials conducted before the availability of NK1R antagonists suggested that repetitive daily dosing with a 5-HT3 receptor antagonist combined with dexamethasone was the best approach [81,82]. However, more recently, the benefit of adding aprepitant to a 5-HT3 antagonist and dexamethasone was shown in a small trial of 69 patients receiving a five-day cisplatin-containing chemotherapy regimen for germ cell cancer . All patients received a 5-HT3 antagonist (other than palonosetron) once daily on days 1 through 5, plus dexamethasone 20 mg once daily on days 1 and 2, and were randomly assigned to aprepitant (125 mg on day 3, 80 mg on days 4 and 5) or no aprepitant. The group receiving aprepitant also received dexamethasone 4 mg twice daily on days 6, 7, and 8, while the placebo group received dexamethasone 8 mg twice daily on days 6 and 7 and 4 mg twice daily on day 8. A complete response (no emetic episodes and no use of rescue medication) was noted in significantly more patients receiving aprepitant (42 versus 13 percent), and the visual analog scale (VAS) score for nausea was numerically lower for aprepitant, although the difference compared to placebo was not statistically significant.
Although the authors provided a rationale for starting the NK1R on day 3 rather than on day 1, the optimal schedule for NK1R antagonists for patients receiving consecutive day therapy with highly emetogenic chemotherapy is unknown given the lack of comparative trials addressing this question. For regimens containing five days of cisplatin (eg, for testicular germ cell cancer), we suggest a daily dose of an oral 5-HT3 receptor antagonist (or granisetron transdermal patch, see below) plus dexamethasone, with the addition of aprepitant or fosaprepitant starting on day 1.
Granisetron transdermal patch — A transdermal preparation of granisetron is available (Sancuso®) that contains 34.3 mg of granisetron and is designed to deliver 3.1 mg of the drug every 24 hours for up to seven days. The efficacy of the granisetron patch relative to daily oral administration of granisetron was shown in a multinational randomized, double-blind, double-dummy controlled trial in which 641 patients receiving the first cycle of a multiday regimen of either moderately or highly emetogenic chemotherapy were randomly assigned to the patch (applied 24 to 48 hours before the first day of chemotherapy) plus a placebo capsule or to oral granisetron (2 mg daily, one hour prior to chemotherapy on each day of chemotherapy administration) plus a placebo patch . Concurrent glucocorticoids, which were administered at the investigator's discretion, were given to about 70 percent of the patients in each arm.
The percentage of patients who had complete control of nausea and vomiting until 24 hours after the last chemotherapy dose was not worse with the transdermal patch (60 versus 65 percent with oral granisetron). Fewer than 1 percent of the patches became detached during treatment. In both groups, the most commonly reported toxicities were constipation and headache.
Based upon this trial, the granisetron transdermal system was approved in the US for the prevention of CINV in patients receiving moderately or highly emetogenic chemotherapy for up to five days. It is recommended that the patch be applied to the upper outer arm a minimum of 24 hours before chemotherapy and removed 24 hours or more after the last chemotherapy dose is administered. It can be worn for up to seven days, depending on the duration of the chemotherapy regimen.
Induction therapy for acute leukemia — High-dose cytarabine regimens (in which cytarabine is administered daily for five or seven days, often with an anthracycline) are the cornerstone of treatment for acute myeloid leukemia (AML). Few studies have addressed the issue of CINV and optimal prophylaxis in this setting [30,85,86]. Although data are lacking, a daily dose of a 5-HT3 receptor antagonist (eg, ondansetron 16 mg) with or without dexamethasone appears to be a reasonable option in this setting.
High-dose chemotherapy regimens — The use of high dose chemotherapy in association with a bone marrow or peripheral blood stem cell transplant presents a special challenge to achieving good antiemetic control. The chemotherapy agents employed are often of moderate to high emetogenic risk.
In addition, there are a number of potential factors that can contribute to an increased incidence and severity of CINV in this setting:
●Higher doses of chemotherapy
●Consecutive day administration
●Prior treatment with chemotherapy
●Inclusion of radiation therapy (especially total body irradiation), which has high emetogenic risk
●Associated other medical conditions or medications that may cause emesis
There are few randomized trials specifically studying the issue of emesis in the high dose setting [87-90]. Most studies have focused on the combination of a 5-HT3 receptor antagonist and dexamethasone [91-93]. However, a role for aprepitant in this setting was suggested in a phase III trial in which 181 patients undergoing a preparative regimen for hematopoietic stem cell transplantation were randomly assigned to ondansetron plus dexamethasone and either aprepitant or placebo . Significantly better emetic control was noted in patients who received aprepitant (no emesis for the entire study period in 73 versus 23 percent of the placebo group), although there were no differences between the groups in use of rescue antiemetics.
Poor control of emesis — Despite the use of appropriate antiemetic prophylaxis, many patients experience clinically significant CINV. Before considering any change in the antiemetic regimen, it is important to exclude other disease- and medication-related causes for emesis. Examples include the following:
●The use of opiate analgesics
●Certain antibiotics (eg, erythromycin)
●Central nervous system metastases
Assuming these factors are excluded, it is important to ensure that the patient is receiving the antiemetic appropriate for the drug(s) being given (table 1 and table 2) and the correct dose (table 3). If the patient was receiving chemotherapy with low emetic risk and is experiencing poor emesis control, it is possible to adjust the regimen to that typically used for a higher risk group. The addition of a benzodiazepine may help to counter increased patient anxiety and possible anticipatory emesis.
The majority of patients who have breakthrough emesis have derived some benefit from the original antiemetic regimen employed. One or two episodes of emesis with cisplatin is less than ideal but still reflects substantial antiemetic efficacy. Thus, the original antiemetic regimens should usually be retained.
Additional agents can be added, including lorazepam or alprazolam, olanzapine , a dopaminergic antagonist (eg, prochlorperazine, thiethylperazine, haloperidol), or substituting high-dose intravenous metoclopramide for the 5-HT3 antagonist . Superiority for olanzapine (10 mg orally daily for three days) over metoclopramide (10 mg orally three times daily for three days) for treatment of breakthrough nausea and vomiting was shown in a double-blind randomized trial involving 80 patients receiving highly emetogenic chemotherapy . During the 72-hour observation period, significantly more patients receiving olanzapine had no recurrent emesis (70 versus 31 percent) and no nausea (68 versus 23 percent).
Another alternative is to switch to a different serotonin antagonist, since there may be incomplete cross-resistance between agents [97-99]. This approach was tested in a double-blind trial, in which patients who failed ondansetron plus dexamethasone in the first 24 hours following highly emetogenic chemotherapy were randomly assigned to continue the same treatment or switch to granisetron plus dexamethasone. There was a significantly higher rate of complete protection from emesis in the patients who switched to granisetron (47 versus 5 percent) .
In other cases, chemotherapy can be altered. Alterations could include single day instead of multiple day therapy, lengthening infusion time, or substituting a less emetogenic agent if such maneuvers do not compromise antitumor activity.
Cannabinoids and medical marijuana — The potential antiemetic utility of cannabinoids was first observed in scattered reports of improved emetic control in patients using marijuana during chemotherapy . Two synthetic cannabinoids are available (dronabinol and nabilone), but antiemetic efficacy is modest at best, rigorous comparisons of either drug with the most effective antiemetic therapies are lacking, and adverse effects tend to be more intense and more frequent than with other rescue agents such as neuroleptics . (See "Characteristics of antiemetic drugs", section on 'Cannabinoids'.)
The modest antiemetic activity of this class of agents combined with their relatively unfavorable side effect profile (vertigo, xerostomia, hypotension, dysphoria), especially in older patients, has limited their clinical utility. Nevertheless, guidelines from the National Comprehensive Cancer Network (NCCN), ASCO , and the MASCC  state that cannabinoids can be considered for refractory nausea and vomiting and as a rescue antiemetic.
The use of medical marijuana for refractory CINV is very controversial. Medical use of marijuana is legal in several countries, including the Netherlands and Canada. Despite legalization by several states, marijuana use is still illegal in the United States at the federal level (which considers marijuana a schedule I controlled substance), and individuals prescribing or using marijuana for medical use are at risk for prosecution .
While an early prospective uncontrolled pilot study from 1988 found that inhaled cannabis was effective in 78 percent of 56 patients who had inadequate control of nausea and vomiting with the conventional antiemetics that were available at that time , there have been no other clinical reports of efficacy of inhaled marijuana and there are no controlled clinical trials comparing marijuana versus other rescue strategies in patients who are refractory to modern antiemetics . Furthermore, the use of marijuana is associated with adverse effects on the cardiovascular, respiratory, and central nervous systems, and concerns about increased risk of malignancy. (See "Cannabis use disorder: Treatment, prognosis, and long-term medical effects", section on 'Long term medical effects of cannabis use'.)
Because of medical and legal concerns, the use of medical marijuana is not recommended for management of CINV, and is not included in the most recent guidelines for CINV from the NCCN, ASCO, or MASCC .
Ginger — Conventional antiemetics are more successful at preventing emesis than in preventing nausea. The benefit of ginger (Zingiber officinale) as an aid to reduce nausea during chemotherapy was addressed in a randomized trial in which 744 patients who experienced nausea following any chemotherapy cycle were randomly assigned to placebo, or supplemental ginger (at doses of 0.5, 1.0 or 1.5 g twice daily) for six days, starting three days prior to the first day of the next two chemotherapy cycles . All patients received a 5-HT3 receptor antagonist on day 1 of all cycles. Two-thirds of the enrolled patients were receiving chemotherapy for breast cancer.
All doses of ginger were associated with a significant reduction in acute nausea throughout day 1 of the chemotherapy cycles, although the largest reduction was seen with the 0.5 and 1.0 g doses. The authors concluded that ginger supplementation significantly aids in reduction of day 1 nausea during chemotherapy.
A benefit for ginger root in conjunction with ondansetron and dexamethasone was also shown in a placebo-controlled randomized trial of 57 children and young adults receiving cisplatin/doxorubicin chemotherapy for bone sarcoma . Compared to conventional antiemetics, the addition of ginger (334 or 800 mg depending on the size of the patient) one hour before, and three and eight hours after the start of chemotherapy significantly reduced the incidence of both acute and delayed emesis.
However, in contrast to these results, a benefit for ginger in addition to conventional antiemetics (5-HT3 antagonists and/or aprepitant) could not be shown in a randomized, placebo-controlled phase II trial of 162 cancer patients who had suffered CINV during at least one prior cycle of chemotherapy . In fact, patients who took ginger plus aprepitant had more severe acute nausea than did those who took only aprepitant.
Acupuncture and related therapies — Several techniques have been used to stimulate the pericardium 6 (P6 or neiguan) site, which is commonly thought to be useful in the management of chemotherapy-induced nausea and vomiting. These include manual stimulation with the insertion of fine needles (acupuncture), electrostimulation, and noninvasive pressure on the skin over the P6 pressure point (ie, acupressure). Unfortunately, interpretation of the results of randomized trials is hampered by a high risk of bias in most studies, and lack of standardization of treatment methods and comparison groups. A year 2013 systematic review of acupuncture in cancer care included 11 randomized trials in which nausea and vomiting were assessed; eight were considered to have a high risk of bias . Only one trial  had a low risk of bias and was positive for short term benefit of electroacupuncture after chemotherapy for breast cancer. The authors concluded that based upon the single positive study with low risk of bias that acupuncture could be considered an appropriate adjunctive treatment for chemotherapy-induced nausea and vomiting but that additional studies were needed. (See "Complementary and alternative therapies for cancer", section on 'Nausea and vomiting'.)
Other nonpharmacologic strategies — Findings from randomized controlled trials of reasonable quality provide limited support for several nonpharmacologic methods to reduce CINV, including cognitive distraction (eg, playing video games during treatment), systematic desensitization (a cognitive approach using visualization and learned relaxation techniques), exercise, hypnosis, acupuncture, and transcutaneous electrical nerve stimulation [110,111]. (See "Complementary and alternative therapies for cancer" and "Acupuncture".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)
●Basics topics (see "Patient information: Nausea and vomiting with cancer treatment (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Chemotherapy-induced nausea and vomiting (CINV) remains an important adverse effect of treatment, despite recent advances in prevention. CINV is classified as acute, delayed, and anticipatory, and these distinctions have important implications for patient management. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Types of emesis'.)
●The most important factor determining the likelihood of acute or delayed emesis developing is the intrinsic emetogenicity of a particular chemotherapy agent (table 1 and table 2). (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Chemotherapy agent'.)
●The three categories of drugs with the highest therapeutic index for the management of CINV are the type three 5-hydroxytryptamine (5-HT3) receptor antagonists, the neurokinin-1 (NK1) receptor antagonists (aprepitant and fosaprepitant), and the glucocorticoids (especially dexamethasone) (table 3). (See '5-HT3 receptor antagonists' above and 'Neurokinin-1 receptor antagonists' above and 'Glucocorticoids' above.)
These agents are used alone (glucocorticoids) and in combinations depending on the specific chemotherapy regimen being administered, as recommended in the ASCO and MASCC/ESMO guidelines [4,5]. (See 'Acute emesis' above.)
●All of the first-generation 5-HT3 receptor antagonists appear equally effective when used at the recommended dose. An orally disintegrating formulation of ondansetron, one that disperses rapidly when placed on the tongue and does not need to be swallowed with water, may be particularly useful for patients with dysphagia or anorexia.
EKG interval changes appear to be a class effect of the first generation 5-HT3 antagonists. Their use should be avoided in patients with congenital long-QT syndrome. Potassium and magnesium levels should be assessed, and if abnormal, corrected before initiation of treatment. ECG monitoring is recommended in patients with underlying cardiac disease including heart failure and bradyarrhythmias, and in patients taking other medications that increase the risk of QTc prolongation (table 5). (See 'EKG interval changes and cardiac arrhythmias' above.)
The second-generation agent palonosetron may be more effective than first generation 5-HT3 antagonists, particularly for prevention of delayed emesis, and QTc prolongation has not been described with palonosetron. In settings when an NK1R antagonist is not indicated, available evidence would support palonosetron as the preferred 5-HT3 receptor antagonist. (See '5-HT3 receptor antagonists' above.)
Recommendations — Antiemetic recommendations from ASCO according to the emetogenicity of the chemotherapy regimen are outlined in the table (table 6) . In general, our recommendations parallel these guidelines.
Cisplatin and other high risk agents
For patients receiving cisplatin-based chemotherapy or another agent with high emetogenic risk (table 1 and table 2), we recommend antiemetic therapy with a combination of a 5-HT3 receptor antagonist, dexamethasone, and a neurokinin-1 receptor (NK1R) antagonist (aprepitant or fosaprepitant) (table 3) (Grade 1A). (See 'Efficacy' above.)
•If aprepitant is used on day 1, we recommend maintenance therapy with a combination of aprepitant on days 2 and 3 and dexamethasone on days 2 through 4 (table 3) (Grade 1A). (See 'Management' above.)
•A single dose schedule of fosaprepitant (dose 150 mg IV 30 minutes prior to chemotherapy in conjunction with a 5-HT3 receptor antagonist on day 1 and dexamethasone on days 1 through 4) is an acceptable alternative. (See 'One versus three-day administration' above.)
Anthracycline plus cyclophosphamide
●For patients receiving chemotherapy with cyclophosphamide plus an anthracycline, we recommend antiemetic therapy with a combination of a 5-HT3 receptor antagonist, dexamethasone, and an NK1R antagonist (Grade 1A). (See 'Efficacy' above.)
•If aprepitant is used on day 1, we recommend maintenance therapy with a combination of aprepitant on days 2 and 3 and dexamethasone on days 2 through 4 (table 3) (Grade 1B). (See 'Management' above.) An alternative regimen is dexamethasone alone.
•A single dose schedule of fosaprepitant (dose 150 mg IV 30 minutes prior to chemotherapy in conjunction with a 5-HT3 receptor antagonist on day 1 and dexamethasone on days 1 through 4) is an acceptable alternative. (See 'One versus three-day administration' above.)
Moderate risk agents and regimens
●For patients receiving moderately emetogenic chemotherapy (table 1 and table 2), we recommend the combination of palonosetron plus dexamethasone on day 1 (Grade 1A). If palonosetron is not available, clinicians may substitute a first-generation 5-HT3 receptor antagonist, such as granisetron or ondansetron. (See 'Combination with a 5-HT3 antagonist' above.)
●To prevent delayed emesis in this population, we suggest single agent treatment with dexamethasone on days 2 and 3 (Grade 2B). If a first generation 5-HT3 receptor antagonist is used rather than palonosetron on day 1, then treatment with a first generation 5-HT3 receptor antagonist alone on days 2 and 3 is a reasonable alternative. (See 'Moderate risk agents' above.)
●For patients receiving low emetic risk agents (table 1 and table 2), we suggest treatment with dexamethasone (8 mg) as a single agent (Grade 2C). An alternative approach for patients in whom glucocorticoid use is contraindicated or undesirable (such as with the use of long-term weekly chemotherapy) is a single dose of a drug such as prochlorperazine . This patient population generally does not require prophylaxis against delayed emesis. (See 'Glucocorticoids' above.)
●For most patients receiving chemotherapy agents with minimal risk of causing emesis (table 1 and table 2), we suggest that antiemetic therapy not be routinely administered to prevent either acute or delayed CINV (Grade 2B). Prophylactic antiemetics (dexamethasone 8 mg, prochlorperazine, or metoclopramide) may be administered to patients who have had emesis with prior low-risk regimens, or on as "as needed" basis. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Chemotherapy agent'.)
●The primary approach to the prevention of anticipatory emesis is the prevention of CINV beginning with the initial cycles of chemotherapy. For patients who do develop anticipatory emesis, we suggest behavioral therapy or benzodiazepines (Grade 2B). (See 'Anticipatory emesis' above.)
●For patients receiving high-dose chemotherapy, we recommend a combination of dexamethasone, a 5-HT3 receptor antagonist, and an NK1R antagonist (Grade 1B). (See 'High-dose chemotherapy regimens' above.)
●For patients receiving multiday (three or more) regimens that are moderately or highly emetogenic, we suggest the use of a daily dose of an oral 5-HT3 receptor antagonist or granisetron transdermal patch plus daily dexamethasone, with the addition of aprepitant (days 1, 2, and 3) or fosaprepitant (day 1) for highly emetogenic regimens (eg, five days of cisplatin in regimens for testicular or ovarian germ cell cancer) (Grade 2C). (See 'Consecutive day therapy with highly emetogenic agents' above.)
Poor emesis control
●For patients who do not achieve adequate control of CINV with their initial antiemetic regimen, the patient's management should be reviewed to ensure that there are no other factors responsible for continued emesis and that adequate antiemetic therapy actually was administered for the given chemotherapy regimen. If CINV remains an issue, the addition of a second line agent (such as olanzapine) or changing from one 5-HT3 receptor antagonist to another may be useful. (See 'Poor control of emesis' above and 'Other agents' above.)
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.