Rolapitant in the Treatment of Nausea and Vomiting Associated With Chemotherapy

Abstract

Rolapitant is a long-acting neurokinin₁ receptor antagonist (NK1 RA) which is metabolized by CYP3A4 but does not inhibit or induce it, thereby reducing potential drug interactions. Four initial randomized studies compared rolapitant with placebo when added to a 5-hydroxytryptamine₃ receptor antagonist (5HT3 RA) and dexamethasone for chemotherapy-induced emesis. Three were with highly emetic chemotherapy (HEC) and one with moderately emetic chemotherapy (MEC) and anthracycline/cyclophosphamide (AC). The rolapitant groups receiving HEC yielded statistically significantly improved responses in all phases of emesis (except the acute and overall phases in study 2) when compared with active controls. Overall (0-120 h), the response was better in the rolapitant arm for AC (P = .0332) and non-AC MEC (including carboplatin) (P = .0003). The efficacy of rolapitant is sustained over multiple cycles. Rolapitant 180 mg orally is well tolerated with similar side effects between the rolapitant and active control arms. It is effective in younger and older adults. Further studies will define its place among other NK1 RAs.

Introduction

Significant impacts on postchemotherapy nausea and vomiting have been made by the introduction of 2 classes of drugs. The first was 5-hydroxytryptamine₃ receptor antagonists (5HT3 RAs), the first of which was ondansetron, which when combined with dexamethasone improved the management of acute vomiting that occurs in the first 24 h after chemotherapy. ¹ Then followed the neurokinin₁ receptor antagonists (NK1 RAs) commencing with aprepitant, which when added to 5HT3 RAs and dexamethasone not only improved the control of acute emesis but also improved the control of delayed emesis over a week, particularly after chemotherapy with high emetic potential. ²

Since then, advances have included the introduction of a long-acting 5HT3 antagonist, palonosetron, and NEPA, a fixed dosage combination of palonosetron with the NK1 RA netupitant. ³ Both classes of drugs are available in oral and intravenous formulations, with the introduction of the intravenous NK1 RA fosaprepitant. ⁴

Despite these advances, there are still areas where improvement is required. Nausea is not as well controlled as vomiting in both the acute and delayed phases. ⁵ Poorly controlled nausea and vomiting initially can lead to anticipatory nausea and vomiting in subsequent cycles and beyond, as a learned response to the anticipation of chemotherapy or as a trigger associated with having chemotherapy, such as smell or sight which may continue to elicit nausea and vomiting, years later. ⁵ Some patients are refractory to the current combination antiemetic therapies, particularly those receiving very high doses prior to bone marrow stem cell transplant, and multiple-day chemotherapy does not have well-defined antiemetic protocols. ⁶ The management of radiation-induced emesis still needs to be optimized, and the control of emesis in pediatric patients has not been well researched as in adults.^7,8

Now a new long-acting NK1 RA, rolapitant, has become available and has been added to the drug choices for controlling chemotherapy-induced nausea and vomiting. ⁹ Its place in the control of the emesis associated with chemotherapy of moderate emetic potential (MEC) and chemotherapy of high emetic potential (HEC) is being defined.

Rolapitant Pharmacodynamics and Pharmacokinetics

Rolapitant is an orally active selective NK1 RA which penetrates the central nervous system and has a half-life of around 169 to 183 h compared with 9 to 13 h for aprepitant.^10,11 It binds to the neurokinin receptors with high affinity (K_i, .66 nmol/L), and positron emission tomographic brain scans have shown a mean 73% receptors in the striatum occupied at 120 h after a single 180 mg dose. ¹²

Rolapitant is metabolized by CYP3A4, but it does not inhibit or induce CYP3A4, so it does not interact with other drugs that are substrates of CYP3A4, unlike aprepitant, where, for example, a concomitant dose of dexamethasone should be halved because the interaction reduces its clearance.^10,13 In a phase I trial of where rolapitant was co-administered with midazolam, a CYP 3A substrate, no interactions were seen. ¹⁴ The efficacy of rolapitant, however, can be decreased due to the reduction in plasma concentrations if given with potent CYP3A4 inducers such as rifampicin. ¹¹

Rolapitant is a moderate inhibitor of CYP2D6 and is contraindicated for use in patients receiving thioridazine where serious interactions can occur. It also inhibits P-glycoprotein transporter and breast cancer resistance protein transporter. ¹¹ When the adverse effects were compared between the rolapitant-containing antiemetic arm and the control arm of 4 randomized studies with 2800 patients who concomitantly received drugs that are metabolized by CYP2D6, BCRP, and CYP3A4, no increase in adverse events was seen. ¹¹

Rolapitant is largely excreted by the hepatobiliary route, but no dose adjustment is necessary for mild or moderate hepatic impairment. It is rapidly absorbed and reaches the circulation within 30 min of oral dosing, with its maximum concentration at 4 h after a dose. Food does not affect its absorption, and no dose adjustment is required for mild or moderate renal impairment. ¹¹

An intravenous formulation of rolapitant has been developed. In an open-label, single-center, parallel-group randomized study, the bioequivalence of a 180-mg oral dose (67 patients) was compared with a 166.5-mg 30-min intravenous infusion (71 patients). The study showed that the 2 formulations were bioequivalent. ¹⁵

The Efficacy of Rolapitant

There were 4 initial randomized studies of rolapitant for chemotherapy-induced emesis, 1 randomized phase II study, 2 randomized phase III trials for HEC, and the other a randomized trial with MEC and anthracycline and cyclophosphamide (AC).

Rolapitant with HEC

An initial randomized, double-blind, active-controlled international phase II trial accrued patients who were receiving their cisplatin >70 mg/m² for the first time as part of their chemotherapy. ¹⁶ A total of 454 patients were randomized to receive one of 4 doses of rolapitant—9, 22.5, 90, or 180 mg orally—or placebo. Both groups also received ondansetron and dexamethasone on day 1. All patients receiving any dose of rolapitant had an improved complete response (CR—no emesis and no use of rescue medication) rate over the placebo active control, but the 180-mg dose of rolapitant was associated with the greatest benefit compared with the control group in the overall phase (0-120 h) of 62.5% vs 46.7%, P = .045; in the acute phase of 87.6% vs 66.7%, P = .001; and in the delayed phase of 63.6% vs 48.9%, P = .045. Using the endpoint of no emesis and no significant nausea, the rolapitant 180 mg dose was significantly better than the control arm in the overall, acute, and delayed phases of postchemotherapy emesis.

The 2 randomized phase III HEC studies were reported together by Rapoport et al. ¹⁷ The 2 randomized, double-blind, active-controlled trials were multinational from 26 countries. The patient population was those with cancer, more than 18 years of age, and with a Karnofsky performance score of 60 or more who were due to receive their first ever dose of cisplatin at a dose of 60 mg/m² or more. They were judged to have a life expectancy of at least 4 months and adequate bone marrow, liver, and renal function. There was a stratification of patients by sex and then a web-based randomization to either receive 180 mg of oral rolapitant or an identical placebo. In addition, all patients received granisetron 10 µg/kg intravenously with dexamethasone 20 mg orally on day 1, 1 to 2 h prior to the chemotherapy, and then dexamethasone 8 mg orally twice daily on days 2 to 4. The patients received up to 5 subsequent cycles of at least 14 days’ duration using these antiemetics prior to their doses of HEC. There were 532 patients in the first trial and 555 in the second.

The primary endpoint was the percentage of patients achieving a CR, that is, no emesis and no use of rescue medication in the delayed phase of emesis (>24-120 h) which is where the NK1 RAs are likely to be most effective. The secondary endpoints included CR in acute (0-24 h) and overall (0-120 h) phases of postchemotherapy emesis. The rolapitant group yielded significantly more CRs. The pooled results showed 71% vs 60%; odds ratio, 1.6; 95% confidence interval (CI), 1.3 to 2.1; P = .0001. In the pooled analysis, the rolapitant groups yielded statistically significantly improved responses in all phases of emesis when compared with active controls: acute emesis (first 24 h) P = .0022, overall (>24-120 h) P < .0001. Moreover, nausea was statistically significantly better controlled in patients receiving rolapitant acute, delayed, and overall compared with the active control. For no nausea, overall the pooled results comparing rolapitant with control were 52% vs 42% (P = .0004) and for no clinically significant nausea, 72% vs 65% (P = .0174).

Rolapitant with MEC and AC

The third trial was a randomized trial of rolapitant for MEC and for the combination of AC which is now classified as HEC. ¹⁸ Again, this was an international study from 23 countries with similar eligibility criteria to the HEC studies except that the patients could not have received prior HEC or MEC. They were stratified by sex. They were randomized to receive either rolapitant 180 mg or placebo 1 to 2 h before the chemotherapy, along with granisetron 2 mg and dexamethasone 20 mg orally on day 1 except those receiving taxanes where the dose of dexamethasone was that used for taxane premedication. On days 2 and 3, they received 2 mg of oral granisetron. They received up to 5 cycles of therapy of at least 14 days’ duration.

Of the 1369 patients randomized, 703 received AC and 629 received non-AC MEC. Again, the primary endpoint was response in the delayed phase, and a significantly greater percentage of patients receiving rolapitant responded (71% vs 62 % in cycle 1 of therapy; odds ratio, 1.6; 95% CI, 1.2-2.0; P = .0002). For the AC group, the response was better for those receiving rolapitant (67% vs 60%; odds ratio, 1.4; 95% CI, 1.0-1.9; P = .465). Those receiving rolapitant for non-AC MEC also had better control of delayed emesis, (76% vs 64%; odds ratio, 1.8; 95% CI, 1.3-2.5; P = .0008). In the acute phase of emesis, there was no significant difference between the arms (P = .9659) for AC but a better response to rolapitant for non-AC MEC (P = .0163). Overall (0-120 h), the response was better in the rolapitant arm for AC (P = .0332) and non-AC MEC (P = .0003). Although rolapitant was superior in controlling vomiting, there was no difference between rolapitant and placebo groups in controlling nausea.

Rolapitant with carboplatin

In the MEC study, 30.1% (401) patients received a carboplatin-based chemotherapy, and this group has been analyzed separately. ¹⁹ Compared with the active control group, the patients receiving rolapitant had a significantly improved CR rate overall (80.2% vs 64.6%, P < .001) and in the delayed phase of emesis (82.3% vs 65.6%, P < .001), but there was no significant difference between groups for acute emesis where only around 10% experienced chemotherapy-induced nausea and vomiting. An improvement in nausea seen for rolapitant did not reach statistical significance; however, over the 120 h, the time to first emesis or use of rescue medication was significantly longer in the rolapitant group (P < .001).

The Toxicity of Rolapitant

In the 2 HEC studies, rolapitant was well tolerated. The frequency of toxicities reported was similar in the rolapitant and active control groups. The most commonly reported toxicities were dyspnea, headache, and constipation, but the incidence of these side effects was less than 2%. ¹⁷

In the MEC and AC study, there was no difference in the occurrence of side effects between the rolapitant and placebo arms. The most common side effects ascribed to the treatment were fatigue, dizziness, and headache. ¹⁸

Over multiple cycles of chemotherapy, the incidence of treatment-related side effects was similar between the rolapitant and control arms of the studies. ²⁰ The side effects were the same as those reported in cycle 1, did not increase with subsequent cycles, and so there was no evidence of any cumulative toxicity.

Quality of Life and Rolapitant

In antiemetic studies, the impact of the treatment on quality of life is often assessed using the Functional Living Index–Emesis (FLIE) in which the patients report how the symptoms of the nausea and vomiting affect their daily life. In the MEC and AC study, more of those receiving rolapitant reported that the chemotherapy-induced nausea and vomiting had no effect on their activities of daily living compared with those receiving the active control (73% vs 67%; odds ratio, 1.3; 95% CI, 1.0-1.7; P = .0270). ¹⁸ For the carboplatin subgroup, 86.1% reported no impact on their daily life compared with 80.4% of patients receiving active control, and this difference is not statistically significant (P = .145)

In the 2 HEC studies, rolapitant improved the FLIE total score (114.5 vs 109.3, P < .001), vomiting score (59.2 vs 55.8, P < .001), and nausea (55.3 vs 53.5, P < .05); however, the percentage of patients with no impact on daily life did not reach significance for rolapitant over control (odds ratio 1.3; 95% CI, 1.0-1.7; P = .082). ²¹

Rolapitant Over Multiple Cycles of Chemotherapy

Patients who were enrolled in the 2 randomized HEC trials, the MEC/AC trial and the randomized phase II trial of cisplatin HEC trial, continued the same randomized treatment irrespective of their response in cycle 1 of their respective study, and the outcomes were reported in a separate paper. ²⁰ They continued for up to 5 further cycles. On days 6 to 8 of the additional cycles of treatment, the patients were asked 2 questions: (1) Have you had any episode of vomiting or retching since your chemotherapy started in this cycle? (2) Have you had any nausea since your chemotherapy started in this cycle that interfered with normal daily life? From these questions, the endpoints that were reported were no emesis, no interfering nausea, time to first emesis, and a combination of no emesis or nausea that interfered with daily life. Safety was monitored throughout. It is important to note that the proportion of patients discontinuing their chemotherapy was the same in the rolapitant and active control arms. The main reason for discontinuation was that the patient had completed a prescribed course of 4 cycles of therapy (34% vs 37.2%), and discontinuation was not due to safety or toxicity issues with the antiemetics.

Compared with the active controls, the rolapitant group had less emesis or nausea that interfered with their daily activities, over multiple cycles of therapy. The difference reached statistical significance in cycle 2 (P = .006), cycle 3 (P < .001), cycle 4 (P = .001), and cycle 5 (P = .021). The difference in vomiting was significant for cycles 2 to 6 but not significant in cycle 6 for nausea. The difference in the time to first emesis, which favored the rolapitant arm, was maintained in the subsequent cycles (P < .001).

Rolapitant and Age

The 3 randomized phase III trials (2 with HEC and 1 with MEC/AC) were analyzed to ascertain the efficacy and safety by age of adding rolapitant to a 5HT3 RA and dexamethasone to control postchemotherapy emesis. Across the studies, 1755 patients (73.1%) were under 65 years and 647 (26.9%) were 65 years or older. ²² Chemotherapy-induced nausea and vomiting appeared equally likely in both age ranges; the overall incidence was close to 40% in both groups. When 180 mg of rolapitant is added to a 5HT3 RA and dexamethasone, the CR rate and rates of no emesis and complete protection, which have better outcomes in the delayed and overall phases compared with control, were similar in the under 65 and 65 and over groups. The CR rate in the acute phase for younger and older patients was 76.8% vs 76.1%, for the delayed phase was 59.8% vs 61.3%, and for the overall phase was 58.5% for both. In the MEC/AC trial, the placebo arm showed a trend toward a lower CR rate in all phases consistent with young age being a risk factor and the group containing more females, which also predicts poorer control of emesis. As reported in the overall analysis of the 3 studies, the addition of rolapitant improved CR in both age groups in the HEC patients for acute, delayed, and overall emesis and in the delayed and overall phases of the MEC/AC trial. It is very well tolerated by both older and younger patients.

Discussion

Rolapitant is the first long-acting NK1 RA. In the 2 HEC trials and the MEC/AC, more patients on rolapitant who achieved a CR in the acute phase continued with a CR in the delayed phase of emesis than those receiving the active control.^17,18 This suggests that rolapitant continues to be active beyond day 1. Even those patients receiving rolapitant who did not achieve a CR in the acute phase were more likely than the control group to achieve a CR in the delayed phase, demonstrating the activity of rolapitant in preventing delayed emesis. Although rolapitant showed activity in both the acute and delayed phases of emesis, the primary endpoint of the studies was delayed emesis based on the current consensus that acute emesis is predominantly mediated by the 5HT3 receptor and delayed emesis by the NK1 receptor, and the impact of an NK1 RA is greater in the delayed phase.

In both the HEC studies and the MEC/AC study, those on rolapitant who were free of emesis and the need for rescue at 48 h (or in one of the HEC studies at 72 h) continued that way for the remainder of 120 h, again suggesting a long-term benefit for rolapitant over the active control.

The randomized studies compared rolapitant with a placebo added to a 5HT3 RA and dexamethasone. For the HEC studies, this is now ethically problematic because current international guidelines specify triple antiemetic therapy for HEC. This design only allows confirmation that adding rolapitant, the long-acting NK1 RA, to a 5HT3 RA and dexamethasone adds efficacy, as occurred when adding the short-acting NK1 RAs to this combination. It does not provide information on whether rolapitant offers any advantage over its short-acting NK1 RAs such as aprepitant. The mitigating factors are that studies show that triple therapy is not always used in clinical practice, rescue medication was allowed, and a patient could be withdrawn from the study and given an NK1 RA if necessary. ²³ With MEC, most guidelines do not routinely recommend an NK1 RA, so a randomized trial against 5HT3 RA and dexamethasone is appropriate. Some guidelines allow for NK1 RA in MEC in selected patients, such as those receiving carboplatin or who have additional risk factors for postchemotherapy nausea and vomiting.^24–26

The degree of benefit that would warrant changing guidelines is set for the Multinational Association of Supportive Care in Cancer (MASCC)/European Society of Medical Oncology guidelines are at greater than 10% because this is considered clinically meaningful. ²⁴ This was achieved with rolapitant with some MEC regimens. For example, in the group of patients on the rolapitant arm prior to carboplatin, the absolute benefit for CR in the delayed phase was 16.7%. Similarly, the addition of other NK1 RA to 5HT3 RAs and dexamethasone has been reported as achieving a 10% to 14% increase in benefit. At present, rolapitant is included in the MASCC guidelines as one of the options when an NK1 RA is indicated.

Given that the current studies compare the addition of rolapitant to adding a placebo to 5HT3 RAs and steroids, what potential benefit would rolapitant have over other NK1 RAs for use to prevent chemotherapy-induced emesis? One potential advantage is that it is long acting, and therefore, a single dose on day 1 may be sufficient to have impact on both acute and delayed emesis. If it is more effective against delayed emesis, it may be possible to construct antiemetic regimens that do not expose the patients to the high doses of steroids, which not only have their own side effects but may also interact with immunotherapeutic drugs. Another difference with the short-acting NK1 RAs is that by not inducing or inhibiting CYP3A4, there will be reduced propensity for drug interactions and this includes not having to dose adjust the concomitantly administered dexamethasone as is required with short-acting NK1 RAs. There has been no clinically important drug interaction with cytotoxics and aprepitant, so this reduced propensity for drug interaction will mainly be important with other concomitant medications.

A difficulty in deciding the current place of rolapitant in antiemetic regimens for chemotherapy-induced nausea and vomiting is that we do not have large randomized controlled trials comparing short- and long-acting NK1 RAs. A long-acting NK1 RA should allow a single dose to cover the delayed emesis of single-day highly emetogenic chemotherapy and multiple-day chemotherapy, particularly with high-dose regimens prior to stem cell or bone marrow transplants, but further trials are needed. It is worth noting, by comparison, that with the long-acting 5HT3 RA, palonosetron, the initial studies comparing it with its short-acting predecessors showed noninferiority and slowly more data accumulated so that a recent meta-analysis indicates superiority and greater safety for palonosetron. However, it is still not specifically recommended over the other 5HT3 RAs in major international guidelines. ²⁷

Two network meta-analyses have been reported for NK1 RAs. One examined randomized controlled trials of 9 triple antiemetic therapy regimens for cisplatin-based emesis. The compliance with rolapitant, granisetron, and dexamethasone was best because of its low toxicity. However, overall from 9 regimens, NEPA (netupitant, palonosetron, and dexamethasone) was found to be the most effective. ²⁸

Another network meta-analysis examined the comparative effectiveness of NK1 RAs from randomized studies evaluating aprepitant, fosaprepitant, netupitant (NEPA), casopitant, and rolapitant in preventing highly emetogenic chemotherapy. In summary, most of the regimens which contained an NK1 RA outperformed regimens without an NK1 RA in achieving CR in the overall phase. ² Although recognizing the pitfalls of comparing across trials, the casopitant and aprepitant regimens were more effective than the rolapitant regimens in achieving CR. There were no significant differences found between regimens in the rates of nonsignificant nausea.

Rolapitant is a long-acting NK1 RA which is effective, particularly for delayed chemotherapy-induced nausea and vomiting in both HEC and MEC regimens, although in the latter NK1 RAs are not routinely recommended. It is very well tolerated with less propensity for drug interactions because it does not inhibit or induce CYP3A4. Further studies will define its place among other NK1 RAs and its inclusion in guidelines based on its clinically relevant differences between rolapitant regimens and controls in chemotherapy-induced emesis. Specifically, studies are required that test rolapitant against short-acting NK 1RAs, including in the control of emesis with multiple-day chemotherapy. At present, it is included in the MASCC 2016 guidelines along with other NK1 RAs for prevention of acute and delayed emesis in HEC regimens, and the choice of NK1 RAs is left to the clinician based on convenience and cost. ²⁴ It is Federal Drug Administration approved for prevention of delayed emesis of chemotherapy-induced nausea and vomiting.