Neoadjuvant plus adjuvant bevacizumab in early breast cancer (NSABP B-40 [NRG Oncology]): secondary outcomes of a phase 3, randomised controlled trial

Background Highly emetogenic chemotherapy induces emesis in almost all patients in the absence of prophylaxis. Guidelines recommend use of a neurokinin-1 (NK-1) receptor antagonist in conjunction with a 5-HT 3 receptor antagonist and corticosteroid in patients receiving highly emetogenic chemotherapy. We aimed to assess rolapitant, an NK-1 receptor antagonist, for prevention of chemotherapy-induced nausea and vomiting in patients with cancer after administration of cisplatin-based highly emetogenic chemotherapy. Methods We conducted two global, randomised, double-blind, active-controlled, phase 3 trials (HEC-1 and HEC-2) at 155 cancer centres (76 in HEC-1 and 79 in HEC-2) in 26 countries (17 in HEC-1 and 14 in HEC-2). We enrolled patients with cancer aged 18 years or older, who had not previously been treated with cisplatin, with a Karnofsky performance score of 60 or higher, and a predicted life expectancy of 4 months or longer. We used an interactive web-based randomisation system to randomly assign patients to treatment. Patients were stratified by sex and randomly allocated to either oral rolapitant (180 mg dose; rolapitant group) or a placebo that was identical in appearance (active control group) about 1-2 h before administration of highly emetogenic chemotherapy. All patients received granisetron (10 mu g/kg intravenously) and dexamethasone (20 mg orally) on day 1, and dexamethasone (8 mg orally) twice daily on days 2-4. Every cycle was a minimum of 14 days. In up to five subsequent cycles, patients were allowed to receive the same study drug they were assigned in cycle 1, unless removed at the clinician's discretion. Patients could also choose to leave the study at any point. Efficacy analysis was done in the modified intention-to-treat population (comprising all patients who received at least one dose of study drug at a cancer centre compliant with Good Clinical Practice [GCP]). The primary endpoint was the proportion of patients achieving a complete response (no emesis or use of rescue medication) in the delayed phase (>24-120 h after initiation of chemotherapy) in cycle 1. These studies are registered with ClinicalTrials.gov, numbers NCT01499849 and NCT01500213. Both studies have been completed. Findings Between Feb 21, 2012, and March 12, 2014, 532 patients in HEC-1 and 555 patients in HEC-2 were randomly assigned to treatment. 526 patients in HEC-1 (264 rolapitant and 262 active control) and 544 in HEC-2 (271 rolapitant and 273 active control) received at least one dose of study drug at a GCP-compliant site and were included in the modified intention-to-treat population. A significantly greater proportion of patients in the rolapitant group had complete responses in the delayed phase than did patients in the active control group (HEC-1: 192 [73%] vs 153 [58%]; odds ratio 1.9, 95% CI 1.3-2.7; p=0.0006; HEC-2: 190 [70%] vs 169 [62%]; 1.4, 1.0-2.1; p=0.0426; pooled studies: 382 [71%] vs 322 [60%]; 1.6, 1.3-2.1; p=0.0001). The incidence of adverse events was similar across treatment groups. The most commonly reported treatment-related treatment-emergent adverse events in the rolapitant versus active control groups were headache (three [<1%] vs two [<1%]), hiccups (three [<1%] vs four [<1%]), constipation (two [<1%] vs three [<1%]), and dyspepsia (two [<1%] vs three [<1%]). For cycle 1, the most common grade 3-5 adverse events in patients allocated rolapitant versus active control were neutropenia (HEC-1: nine [3%] vs 14 [5%]; HEC-2: 16 [6%] vs 14 [ 5%]), anaemia (HEC-1: one [<1%] vs one [<1%]; HEC-2: seven [ 3%] vs two [<1%]), and leucopenia (HEC-1: six [2%] vs two [<1%]; HEC-2: two [<1%] vs two [<1%]). No serious treatment-emergent adverse events were treatment related, and no treatment-related treatment-emergent adverse events resulted in death. Interpretation Rolapitant in combination with a 5-HT 3 receptor antagonist and dexamethasone is well-tolerated and shows superiority over active control for the prevention of chemotherapy-induced nausea and vomiting during the at-risk period (120 h) after administration of highly emetogenic cisplatin-based chemotherapy.