Some patients experience nausea and/or vomiting (NV) before receipt of chemotherapy. Our objective was to evaluate the impact of prior chemotherapy-induced NV (CINV) on the incidence of anticipatory NV in later cycles. This multicenter, prospective non-interventional study enrolled chemotherapy-na < ve adults scheduled to receive highly or moderately emetogenic chemotherapy (HEC/MEC) for cancer in six Asia Pacific countries, excluding those with emesis within 24 h before cycle 1 chemotherapy. On day 1 before chemotherapy, patients answered four questions regarding emesis in the past 24 h, nausea, expectation of post-chemotherapy nausea, and anxiety in the past 24 h, the latter three scored from 0-10 (none-maximum). Multivariate logistic regression was used to assess the impact of prior CINV on anticipatory NV in cycles 2 and 3. Five hundred ninety-eight patients (59 % female) were evaluable in cycle 2 (49 % HEC, 51 % MEC). The incidence of anticipatory emesis was low before cycles 2 and 3 (1.5-2.3 %). The incidence of clinically significant anticipatory nausea (score of a parts per thousand yen3) was 4.8, 7.9, and 8.3 % before cycles 1, 2, and 3, respectively, with adjusted odds ratio (OR), 3.95 (95 % confidence interval (CI), 2.23-7.00; p < 0.001) for patients with clinically significant nausea in prior cycles, compared with none. The adjusted ORs for other anticipatory NV endpoints ranged from 4.54-4.74 for patients with prior CINV. The occurrence of clinically significant anxiety in the prior cycle also resulted in a significantly increased likelihood of anticipatory nausea. These findings highlight the importance of preventing CINV in cycle 1 to reduce anticipatory NV in subsequent cycles.