In this work, data from two phase III studies were pooled to further evaluate the NK 1 antagonist aprepitant for prevention of cisplatin induced nausea and vomiting. One thousand and forty three patients receiving cisplatin (⩾70 mg/m 2) were randomised to receive either a control regimen (32 mg intravenous ondansetron [O] and 20 mg oral dexamethasone [D] on day 1; 8 mg D twice daily on days 2–4) or an aprepitant (A) regimen (125 mg A plus 32 mg O and 12 mg D on day 1, 80 mg A and 8 mg D once daily on days 2–3, and 8 mg D on day 4). The primary endpoint was no emesis and no rescue therapy. Potential correlations between acute and delayed emesis were assessed, as were frequency of emetic episodes by time interval and effects on nausea and quality of life as measured by the functional living index emesis (FLIE) questionnaire. In the aprepitant group, there was statistically significantly less nausea over the study period as well as higher functioning on the FLIE questionnaire in both the nausea and vomiting domains. Patients without acute emesis were more likely to have no emesis in the delayed phase. Compared with control, the aprepitant regimen improved prevention of delayed emesis by 16% points in patients without acute emesis, and by 17% points in patients with acute emesis. Among patients who did not have complete response, the frequency of emesis at various intervals over 5 days was consistently lower in patients receiving aprepitant. Analyses of this combined Phase III population further characterized the clinical profile of the aprepitant regimen, showing that delayed emesis is correlated with, but not entirely dependent on, the presence of acute emesis, and that aprepitant has a favorable effect against nausea throughout 5 days postchemotherapy. In addition, even among patients who had emesis or needed rescue therapy, aprepitant was associated with a lower frequency of these events compared with the control regimen.