The cost of antiemetic therapy for chemotherapy-induced nausea and vomiting in patients receiving platinum-containing regimens in daily practice in Japan: a retrospective study
- First Online:
- Cite this article as:
- Hamada, S., Hinotsu, S., Hori, K. et al. Support Care Cancer (2012) 20: 813. doi:10.1007/s00520-011-1155-3
- 262 Views
The objective of this study was to estimate the cost of antiemetic therapy for chemotherapy-induced nausea and vomiting (CINV) in daily practice in Japan.
This was a retrospective observational study using medical records. Eligible patients were those with bladder or testicular cancer receiving platinum-containing highly emetogenic chemotherapy. The incidence of CINV on days 1–5 in single-day chemotherapy and on days 1–9 in multiple-day chemotherapy, and the costs of antiemetic therapy directly associated with the administration of antiemetics were estimated. The analysis of costs was performed from a hospital perspective.
A total of 54 patients or 169 chemotherapy courses were included. In all chemotherapy courses 5-HT3 receptor antagonists were used on the day(s) that platinum-containing agents were administered and frequently used on subsequent days. In contrast, the use of corticosteroids was infrequent. Acute CINV in single-day chemotherapy was well controlled, but the incidences of delayed CINV in single-day chemotherapy and CINV in multiple-day chemotherapy were relatively high. The costs for antiemetic therapy were $484.65 in courses with CINV and $318.56 in courses without CINV, and the difference was approximately $170 per chemotherapy course, which was considered to be mainly imputable to the prevalence of CINV.
The cost of antiemetic therapy for CINV is substantial in Japan as well as in other countries, and it is suggested that the onset of CINV is a possible cost driver. The improvements in antiemetic therapy may contribute not only to improved patient well-being but also to a reduction of economic burden.
KeywordsChemotherapy-induced nausea and vomitingEmesisCost of illnessHealth economicsPharmacoeconomics
Chemotherapy-induced nausea and vomiting
Chemotherapy-induced nausea and vomiting (CINV) is common and the most distressing side effect for patients receiving cancer chemotherapy [1–3]. Patients developing CINV may occasionally be forced to discontinue or postpone their chemotherapy, resulting in a loss of optimal clinical outcomes. The specified risk factors for CINV are, for example, female gender, younger age, dose, and intrinsic emetogenicity of chemotherapeutic agents given, history of low alcohol consumption, and pretreatment expectation of severe nausea . Regarding the severity of emetogenicity, chemotherapeutic agents are generally placed in one of four risk categories—minimal, low, moderate, and high—according to the expected frequency of emesis in the absence of effective antiemetic prophylaxis .
In research into CINV, chemotherapy regimens have been classified according to the dosing schedule of chemotherapeutic agents: single-day or multiple-day chemotherapy. Although the introduction of 5-HT3 receptor antagonists has dramatically improved the control of acute CINV (occurring within 24 h of chemotherapy) in patients receiving single-day chemotherapy, the control of delayed CINV (occurring more than 24 h after chemotherapy) remains a significant problem [6, 7]. Multiple-day chemotherapy is given to patients with germ cell cancer, sarcoma, leukemia, and non-Hodgkin’s lymphoma, such as the 5-day fractionated administration of cisplatin to patients with testicular cancer, and nearly half of patients continue to experience emesis despite prophylaxis with 5-HT3 receptor antagonists and corticosteroids .
Recently, several studies have demonstrated that CINV not only has clinical consequences but also has an economic impact [6, 9–11]. The costs of CINV including direct and/or indirect ones imputable to CINV were estimated in these studies, but such costs cannot be extrapolated directly to the setting of another country because of the differences in commonly used antiemetic therapies, drug pricing, and personnel costs. Several studies evaluating CINV-associated costs have been performed in Japan [12, 13], but the cost of antiemetic therapy for CINV itself has not yet been clarified. In general, the economic evaluations of medicine are limited in Japan, compared to research conducted in the United States or Europe . The purpose of the present study was to estimate the cost of antiemetic therapy for CINV and the impact of the onset of CINV on the cost in daily practice in Japan.
Patients and methods
Study design and setting
This was a retrospective observational study. The measurements were extracted from the medical records at two National University Hospitals (Hamamatsu University School of Medicine, University Hospital, Hamamatsu, Japan and Tsukuba University Hospital, Tsukuba, Japan).
Platinum-containing compounds such as cisplatin and carboplatin are chemotherapeutic agents widely used for the treatment of malignancies, including head and neck, lung, ovarian, bladder, and testicular cancer. Cisplatin and carboplatin have relatively high emetogenicity, and are listed as high and moderate risks when used as single agents, respectively, in emetogenic schema . Some of us are urologists with a great deal of experience in chemotherapy, and so we decided in this study to focus on patients with bladder or testicular cancer receiving platinum-containing highly emetogenic chemotherapy in a hospital setting between January 2005 and December 2007. Chemotherapy regimens included in the study were as follows: MVAC (methotrexate, vinblastine, doxorubicin, and cisplatin), GC (gemcitabine and cisplatin or carboplatin), and ITP (ifosfamide, paclitaxel, and cisplatin) for bladder cancer, and BEP (bleomycin, etoposide, and cisplatin), EP (etoposide and cisplatin), TIP (paclitaxel, ifosfamide, and cisplatin), and VIP (vinblastine, ifosfamide, and cisplatin) for testicular cancer. The standard doses of cisplatin in the hospitals were 70 mg/m2 for the regimens for bladder cancer, and 100 mg/m2 as single-day chemotherapy and 20 mg/m2/day as a 5-day fractionated multiple-day chemotherapy for the regimens for testicular cancer. These regimens are conventional rather than experimental [15, 16]. All of the chemotherapeutic agents in combination with platinum-containing agents are listed as minimal to moderate risks in the emetogenic schema . The combination of gemcitabine plus carboplatin corresponded to a high risk of emetogenicity according to the algorithm proposed by Hesketh et al. . Consequently, all chemotherapy regimens included in this study were regarded as highly emetogenic. In addition, platinum-containing agents might be the main contributors to the onset of CINV since the emetogenicity of platinum is high compared to the other agents used concomitantly.
Patients were included regardless of their experience of chemotherapy. Patients who were previously diagnosed as having malignancies located in the central nervous system or gastrointestinal tract and patients with metastasis of cancer in these organs were excluded because the onset of CINV might be affected by these conditions.
The research protocol was approved by the ethical review boards at Kyoto University, Hamamatsu University School of Medicine, and Tsukuba University. This study was performed in accordance with the Declaration of Helsinki and the Ethical Guidelines for Epidemiological Research in Japan.
Data were obtained by a structured chart review. We extracted the following information from the medical records: age, sex, chemotherapy setting (inpatient or outpatient), dates of admission and discharge, type of cancer, histopathological diagnosis, grade of tumor, presence and location of metastasis, previous treatment (chemotherapy, surgery, and radiation), chemotherapeutic agents and antiemetics (dose, route, and time and date of administration of agents), and episodes of nausea and vomiting. No information on the use of benzodiazepines was collected because drugs in this class were not used as antiemetics in the two hospitals. The day or the initial day on which platinum-containing agents were administered was set as day 1.
Incidence of CINV
In this study, nausea and vomiting on days 1–5 in single-day chemotherapy and on days 1–9 in multiple-day chemotherapy were defined as chemotherapy-induced. All episodes of nausea and vomiting in this period were collected per chemotherapy course regardless of the intensity or number of episodes. The incidences of nausea, vomiting, nausea and vomiting, or neither were estimated on day 1 (the acute phase), and on days 2–5 (the delayed phase) for single-day chemotherapy, and on days 1–5 (the acute phase) and on days 6–9 (the delayed phase) in multiple-day chemotherapy. For the purpose of the following cost analysis, chemotherapy courses were classified into those with at least one episode of nausea and/or vomiting and ones without any episodes.
Cost of CINV
All patients scheduled to receive highly emetogenic chemotherapy were given predetermined antiemetic therapy, standardized within each hospital, for the purpose of prevention of CINV. This type of antiemetic therapy was referred to as routine prophylaxis in this study. Routine prophylaxis was basically set before emetogenic chemotherapy, and occasionally until a few days after chemotherapy. In addition to routine prophylaxis, antiemetic therapy was conducted in certain other situations, such as cases where the patient was considered to be susceptible to CINV because of the onset of CINV in a previous course or having actually developed CINV. These types of antiemetic therapy were referred to as additional medication in this study. The overall costs of antiemetic therapy were divided into the two categories—routine prophylaxis and additional medication—and calculated separately.
The cost of antiemetic therapy for CINV per chemotherapy course was calculated by summing the following costs associated with the administration of antiemetics directly: drug, administration device, and staff time to prepare and administer antiemetics. We calculated the costs of antiemetic therapy using drug prices in 2007, and the costs of administration devices and personnel costs reported by Ishimaru et al. . The unit cost for an administration device was added to the cost of routine prophylaxis and/or additional medication when the device was used for each intended purpose at least once, irrespective of the number of administration devices used. Personnel cost was added depending on the number of times the antiemetics were administered.
A prospective payment system based on the diagnostic procedure combination was implemented with a partly inherited fee for service system in Japan in 2003 . Hospitals cover the economic burden in acute inpatient care for all cancer-related treatments including the cost of antiemetic therapy for CINV. Therefore, the analysis of costs in this study was performed from a hospital perspective.
All costs (Japanese yen) were converted to US dollars using the mean currency exchange rate in 2007 ($1 = \113.15 yen).
JMP, version 8.0.1 (SAS Institute, Cary, NC, USA)  was used for statistical analysis. For cost analysis, mean, standard deviation, median, and range were calculated as descriptive statistics because costs were expected not to be normally distributed and to be highly variable due to differences in chemotherapeutic and antiemetic regimens.
Description of patients and chemotherapy regimens
Patient characteristics and chemotherapy regimens
No. of patients
Age (years), median (range)
Presence of metastasis, n (%)
Previous chemotherapy, n (%)
No. of chemotherapy courses
Usage of antiemetics
Usage of antiemetics
Single-day (n = 141)
Multiple-day (n = 28)
5-HT3 receptor antagonists
Dopamine D2 receptor antagonists
According to the classification for cost estimation, routine prophylaxis consisted of 5-HT3 receptor antagonists and corticosteroids on the day(s) of administration of platinum agents, and partly of 5-HT3 receptor antagonists from day 2 in single-day chemotherapy, and of metoclopramide. The other antiemetics were used as additional medication.
Incidence of CINV
Incidence of chemotherapy-induced nausea and vomiting per chemotherapy course
Single-day (n = 141)
Multiple-day (n = 28)
Day 1 (Acute phase)
Days 2–5 (delayed phase)
Days 1–5 (overall)
Days 1–5 (acute phase)
Days 6–9 (delayed phase)
Days 1–9 (overall)
Nausea and vomiting
Cost of CINV
Cost of chemotherapy-induced nausea and vomiting per chemotherapy course
Courses without CINV
Courses with CINV
Type of cost
(n = 44)
(n = 125)
146.11 ± 166.25
168.05 ± 157.10
6.81 ± 2.27
7.52 ± 2.38
Staff time (personnel cost)b
1.03 ± 1.53
1.58 ± 1.29
153.95 ± 167.20
177.16 ± 157.04
161.87 ± 235.45
300.15 ± 261.66
1.71 ± 2.19
3.81 ± 2.05
Staff time (personnel cost)b
1.03 ± 1.60
3.53 ± 3.37
164.61 ± 238.72
307.50 ± 265.33
318.56 ± 275.15
484.65 ± 271.38
In this study, we investigated the usage of antiemetics and estimated the incidence of CINV and the cost of antiemetic therapy for CINV in patients receiving platinum-containing highly emetogenic chemotherapy in daily practice. Patients with bladder or testicular cancer were included to grasp current situations associated with CINV in multiple-day chemotherapy as well as single-day chemotherapy. As far as we know, this was the first study to estimate the cost of antiemetic therapy for CINV itself and the impact of the onset of CINV on the cost of therapy in Japan.
The use of antiemetics for the prophylaxis of CINV was highly correlated with the onset of CINV. Acute CINV in single-day chemotherapy was well controlled by 5-HT3 receptor antagonist-based prophylaxis in the present study as well as in other studies [6, 7, 9, 10]. It should be noted that the use of corticosteroids was infrequent or even rare from day 2 in single-day chemotherapy and in the overall observational period in multiple-day chemotherapy, but the use of 5-HT3 receptor antagonists was relatively frequent. This is despite the guidelines available at that time recommending dexamethasone-based antiemetic therapy against delayed emesis induced by high doses (≥50 mg/m2) of cisplatin  or the combination of 5-HT3 receptor antagonists plus dexamethasone against emesis induced by multiple-day chemotherapy . This was considered to be one of the possible reasons for a tendency toward a higher incidence of delayed CINV in single-day chemotherapy in the present study compared to studies performed in other countries [6, 7, 9, 10]. Sato et al. reported a similar situation in another hospital in Japan, where 5-HT3 receptor antagonists tended to be used frequently instead of corticosteroids in patients receiving moderately emetogenic chemotherapy . Although available data on the use of antiemetics in emetogenic chemotherapy are limited, antiemetic therapy in Japan could be improved by the appropriate application of guidelines formulated in Japan and overseas. The first antiemetic guideline in Japan was published in May 2010 . On the other hand, it was reported that about half of the patients receiving multiple-day chemotherapy achieved a complete response (no vomiting and no rescue medication) , and the incidence of chemotherapy-induced vomiting in patients with or without additional medications in the present study was comparable with other studies despite the rare use of corticosteroids. This result does not contradict recommendations to use corticosteroids because our study was observational, with a limited sample size, and was not designed to evaluate the efficacy of antiemetic therapy.
When the costs of antiemetic therapy for CINV directly associated with the administration of antiemetics were estimated, we divided antiemetic therapy into routine prophylaxis, which all patients received, and additional medication, which some patients received depending on their situation. As expected, the costs of antiemetic therapy were highly varied. Since there had been no antiemetic guideline in Japan until the first one was published, antiemetic therapies were basically standardized only within each hospital and the physicians could change the regimens at their manuals in the hospitals included in this study. These were considered major reasons why the variations of costs for routine prophylaxis and additional medication were wide. The overall costs of antiemetic therapy of the courses with or without CINV were $484.65 or $318.56, respectively, and the difference of approximately $170 between courses with and without CINV was principally attributed to the additional medication. Namely, this incremental cost was considered to be mainly imputable to the prevalence of CINV in previous or current chemotherapy courses. Several studies have reported on the costs of CINV per patient per chemotherapy course: costs incurred by third-party payers and hospital providers were €49 and €48, respectively, in Germany , and incremental costs in courses with CINV were CAN $61 in Canada  and €36 in Italy . The overall costs and incremental costs estimated in the present study were higher than those in the above mentioned studies partly because of the frequent use of 5-HT3 receptor antagonists instead of corticosteroids, the use of brand-name drugs despite generic drugs existing, and the calculation of costs using retail rather than wholesale prices. The use of 5-HT3 receptor antagonists more than 24 h after chemotherapy for prevention of delayed emesis has not been justified in terms of clinical efficacy or cost-effectiveness . The use of not only corticosteroids but also metoclopramide and dopamine D2 receptor antagonists could be considered for the prophylaxis of delayed emesis or rescue medication in cases where patients have developed CINV despite prophylactic use of 5-HT3 receptor antagonists. The improved control of CINV and the adjusted cost of antiemetic therapy for CINV might also be achieved as a result.
Two major limitations existed in this study. First, detailed data were limited in the study because of its retrospective nature. The incidence of CINV might be underestimated if all episodes are not reported in the medical records. In addition, quality of life or discomfort of the patients could not be predicted although the incidence of CINV was estimated. Although the costs of drugs, administration devices, and personnel directly associated with the administration of antiemetics were included in the present study, CINV actually affects not only these direct costs but also other costs. From the patient or social perspective, the other direct costs (e.g., transportation, over-the-counter drugs, and alternative medication) and indirect costs (i.e., loss of productivity in patients and caregivers) should be included when considering the cost of CINV. Taking into account the results of studies from other countries [6, 9, 10], these direct costs seem to be relatively low, and the indirect costs were highly variable according to the calculation methods.
The other limitation was a possible selection bias. The patients included in the study well reflected the characteristics of bladder and testicular cancer patients in Japan. Complementary studies, however, are needed to be performed in more extended patient groups with other types of cancer or chemotherapy to achieve a more accurate estimate of incidence and cost.
Most recently, two novel antiemetics, aprepitant, and palonosetron, were approved in Japan. These agents were approved in the United States in 2003. Aprepitant is a member of the family of neurokinin-1 receptor antagonists and selectively blocks the neurokinin-1 receptor, which is the binding site of substance P . The Japanese antiemetic guideline  recommends the use of aprepitant for the prophylaxis of acute and delayed emesis in patients receiving single-day highly emetogenic chemotherapy as well as the MASCC guideline . In addition, the Japanese guideline suggests that aprepitant may contribute to the improvement of the quality of life of patients receiving multiple-day chemotherapy despite the evidences of this drug limited to the cases in single-day chemotherapy. Palonosetron is a second-generation 5-HT3 receptor antagonist with a longer half-life and greater receptor-binding affinity than first-generation 5-HT3 receptor antagonists . Although the effectiveness of replacing the first-generation drugs with palonosetron has not been fully clarified, palonosetron has been shown to have improved efficacy in the prophylaxis of delayed emesis induced by moderately [24, 25] and highly  emetogenic chemotherapy. The improved control of CINV, particularly delayed CINV in single-day chemotherapy and CINV in multiple-day chemotherapy, possibly can be achieved by optimizing antiemetic therapy with these new agents and the existing drugs. The cost of antiemetic regimens that include these new agents might be higher than the current therapy. However, there is also the possibility that the overall cost of antiemetic therapy for CINV decreases by preventing the onset of CINV because the present study suggested that the onset of CINV might increase the cost of antiemetic therapy. It should be noted that the cost of antiemetic therapy for CINV in this study did not include all costs associated with CINV, and patient discomfort was not taken into account.
In conclusion, the cost of antiemetic therapy for CINV is substantial in Japan as well as in other countries, and it is suggested that the onset of CINV is a possible cost driver. Although the goal of antiemetic therapy is to prevent the onset of CINV completely, current antiemetic therapy does not fulfill this ultimate goal. There is much room for improving antiemetic therapy, such as the use of corticosteroids, and other existing and/or new agents. The improved practice of antiemetic therapy may contribute not only to an improvement in patient well-being but also to a reduction of economic burden.
This study was partly supported by a grant from Pfizer Health Research Foundation (2009).
Conflict of interest
No conflict of interest is declared.