Real‐world emetic risk of chemotherapy and the corresponding antiemetic therapy in Japan: A study based on a nationwide database

Abstract Background Chemotherapy‐induced nausea and vomiting (CINV) is a major concern of patients with cancer, leading to suboptimal treatment. Aim This study assessed the emetic risk associated with intravenous and oral chemotherapy and the prophylactic antiemetic drugs by cancer type in a real‐world setting. Methods and Results We used the health services utilisation data for patients with cancer diagnosed in 2016. Patients aged at least 20 years at the time of diagnosis and who started their first course of chemotherapy were included. The emetic risk of chemotherapy was determined according to the cancer type and was classified based on clinical practice guidelines. The prescription of antiemetic drugs was assessed. Overall, 172 133 patients were evaluated, of whom 121 103 (70.4%) received intravenous chemotherapy. High‐emetic‐risk chemotherapy (HEC) was prescribed in 46 458 (27.0%) patients. HEC was prescribed most for patients with oesophageal cancer (80.3%), followed by malignant lymphoma (60.2%) and breast cancer (53.8%). Moderate‐emetic‐risk chemotherapy (MEC) was prescribed in 60 528 (35.2%) patients and was mostly prescribed for small cell lung cancer (59.9%). Meanwhile, more than 50% of the chemotherapy prescribed for patients with gastric, colorectal, and pancreatic cancer was low‐emetic‐risk chemotherapy. HEC was accompanied by three‐drug antiemetic prophylaxis in more than 90% of patients with small cell lung, non‐small cell lung, breast, and oesophageal cancer, whereas only 13.5% of patients with malignant lymphoma were administered CHOP (cyclophosphamide, doxorubicin, vincristine sulphate, and prednisolone) with prophylaxis. Conclusion The risk of CINV differs with cancer type. HEC was less prescribed compared with MEC. Most patients received the recommended anti‐emetic prophylaxis.


| INTRODUCTION
Chemotherapy-induced nausea and vomiting (CINV) is a serious adverse event of chemotherapy. [1][2][3] The frequency of CINV depends primarily on the emetic potential of the chemotherapeutic agent used.
There are effective antiemetic agents for the prevention of CINV, 4 which alleviate CINV significantly. 5 Vomiting is observed in more than 90% of patients receiving high-emetic-risk chemotherapy (HEC) without prophylaxis 6 ; this prevalence is reduced to approximately 30% when antiemetics are administered. [6][7][8] Several guidelines of antiemetic therapy for chemotherapy recommend prescriptions based on the emetic risk of the chemotherapeutic agent used. 9,10 In clinical practice, chemotherapy regimens are chosen depending on the cancer type, tumour stage, patient's general state and preference. 11 Most chemotherapy regimens are not highly emetic, and effective prophylaxis is available. Moreover, with an increasing number of molecular-targeted and immunological therapies, more agents with low and minimal emetic potentials are becoming available. 12 However, the side effects of chemotherapy remain a major concern of patients with cancer. 1,13 Some patients refuse chemotherapy 14,15 for fear of its side effects. 16 Previous studies have reported a chemotherapy non-compliance rate of 5%-18%. 14,17 In addition, avoidance of a treatment-related decrease of their quality of life was the primary reason for refusal, and the fear of nausea before chemotherapy initiation was found to be a strong predictor of subsequent nausea. 18 These data suggest that the patients' concern about treatment, which may be a result of their negative perception of high-risk emetic agents and the lack of antiemetic therapy in early days, can deprive them from receiving appropriate treatment. This fear may be due to limited knowledge of CINV and chemotherapy-specific antiemetic therapy.
Thus, awareness of the frequency of CINV and its management may help curb patients' negative perceptions of chemotherapy and ensure a more objective treatment decision-making. A nation-wide survey reported good compliance with the guidelines of anti-emetic therapy 19 ; however, the findings did not reflect those of a real-world setting as the registry contained clinical information of patients from selected institutions, and a detailed analysis according to cancer type was not performed.
Using health utilisation data from the Hospital-Based Cancer Registries (HBCR), this study aimed to describe the actual frequency of emetic chemotherapy use and the appropriate prophylaxis by cancer type.

| Study design and data source
This observational study used data from the Quality Indicator project, 20 which involved health utilisation data linked with the HBCR.
The project was conducted to evaluate the quality of health care of patients with cancer. In brief, the National Cancer Centre collects HBCR data from designated cancer care hospitals and voluntarily participating hospitals nationwide in Japan, covering approximately 70% of all patients with newly diagnosed cancer. 21 Among these hospitals, 475 hospitals participated in the Quality Indicator project. We analysed health utilisation data collected from 1 January 2016 through 31 December 2017 for patients diagnosed with cancer in 2016. Collection of health utilisation data was part of the governmental survey that assesses the effect of the introduction of the diagnosis procedure combination-based payment system. The survey data included information equivalent to fee-for-service insurance claims that cover all billable health services (e.g., diagnostic tests, imaging workup, procedures, treatments, and prescribed drugs) for both inand out-patients. These data were linked to the HBCR data of each patient in the participating hospitals. The data of approximately 49% of incident cancer cases in Japan were included in this study.

| Patient selection
The subjects of this study were selected as follows. First, patients with cancer who were at least 20 years old at the time of diagnosis in 2016 and who were initiated on chemotherapy (intravenous and/or oral) were included in the study. Second, information on chemotherapeutic agents and antiemetic drugs were extracted from the utilisation data. Patients who received chemotherapy with interferon-alpha were excluded as interferon-alpha is used in other conditions like hepatitis C virus infection. For patients who received chemotherapy more than once, data of the first chemotherapy session after diagnosis was analysed. Many chemotherapeutic agents were administered within the first 8 days in one course (e.g., for patients with gastric cancer, cisplatin was added on Day 8 after S-1 [tegafur/gimeracil/oteracil potassium] initiation). Thus, all drugs administered within the first 8 days were systematically included in one combination regimen in the emetic risk classification. Patients in whom chemotherapy was administered on the same day as surgery, thoracic drainage, abdominal drainage, and/or pericardial drainage were excluded because the chemotherapy drugs might have been administered topically. Patients who received chemotherapy drugs requiring arterial injection were excluded. Further, patients who received haematopoietic stem cell transplantation within 3 weeks of chemotherapy were also excluded because the emetic risks and prophylactic anti-emetic drugs are different from those for regular chemotherapy.

| Emetic risk classification
The emetic risk of chemotherapy was classified using the Japan Society of Clinical Oncology guidelines (JSCO). 22 The guidelines of the National Comprehensive Cancer Network (NCCN), 6 American Society of Clinical Oncology (ASCO), 23 and Multinational Association Supportive Care in Cancer (MASCC) were used to classify the acute emetogenicity of chemotherapy drugs. 24 The major differences between the guidelines are presented in Table 1. The emetic risks of some drugs differed according to dosage. Thus, the emetic risk of these drugs was based on the average Japanese adult's body surface area of 1.48 m 2 . 13,25 Finally, we classified cyclophosphamide administered at >1500 mg/m 2 as HEC and methotrexate sodium administered at >250, 50-250, and <50 mg/m 2 as HEC, MEC, and LEC, respectively. Some of the emetic risks were defined based on the combination of the drugs. 22 That is, fluorouracil, levoholinato, oxaliplatin, and irinotecan (FOLFOXIRI) for colorectal cancer and oxaliplatin, irinotecan, fluorouracil, and levoholinato (FOLFIRINOX) for pancreatic cancer were classified as HEC; gemcitabine and S1 (GS) and gemcitabine and nab-paclitaxel (GEM/nab-PTX) for pancreatic cancer as MEC; ifosfamide, carboplatin, and etoposide (ICE) for malignant lymphoma as HEC; and oral etoposide, nimustine, and ranimustine for malignant lymphoma as MEC.

| Statistical analysis
The frequency of chemotherapy use by emetic risk was calculated in both the overall population and by cancer type. The prescription rate of the prophylactic antiemetic drugs according to the type of cancer and typical regimens in each cancer was also calculated. In general, antiemetic drugs administered on the same day as the first chemotherapy were regarded as prophylactic. For patients who received HEC after the initiation of chemotherapy with lower emetic risks (e.g., S1 + cisplatin is added on Day 8 for gastric cancer) or antiemetic drugs that were prescribed on the same day as HEC were considered prophylactic. To capture oral antiemetic drugs administered in preparation for the initiation of chemotherapy, oral drugs that were prescribed 30 days before chemotherapy initiation were also considered prophylactic. The disease stage was assessed by combining the clinical and pathological stages; the pathological stage was used for patients who underwent surgical resection, whereas the clinical stage was used for patients with unavailable data on the pathological stage of the tumour. 26 All statistical analyses were performed using Stata software (ver. 15.0; Stata Corporation, Texas, USA).

| RESULTS
In total, 172 133 patients receiving chemotherapy were identified (  Among the patients who received HEC, the prescription rates of antiemetic drugs differed with cancer type (Figure 2 Patients with cancer and an indication for chemotherapy often experience pre-treatment psychological distress. 27 A previous study showed that pre-chemotherapy education can decrease treatmentrelated concerns and improve physical/psychological outcomes. 28 Therefore, psychoeducational support can be an effective intervention for managing CINV. 29,30 To avoid treatment refusal due to strong concerns about CINV, health care providers should establish a system to educate patients on the CINV risk of their planned chemotherapy regimen. Further, it should be emphasised that chemotherapy should be accompanied by a recommended anti-emetic therapy. Moreover, the actual risk may depend on the patients' characteristics (cancer and treatment type, patient's age and sex), 31 and thus these factors may need to be incorporated in the educational materials.
A nationwide survey reported a good compliance to the guidelines for anti-emetic therapy. 19 Using the health utilisation data linked with the HBCR, this study described the actual frequency of using  emetic chemotherapy and the appropriate prophylaxis by cancer type and typical regimens. The overall adherence to prophylactic antiemetic drugs for intravenous chemotherapy in this study was higher than that in previous studies in Japan. 32,33 However, this study found that only 13.5% of patients with malignant lymphoma treated with CHOP received the recommended antiemetic therapy; this was consistent with a previous study. 34 CHOP therapy use high dose prednisolone administration. Therefore, it may be recognised in clinical practice that many patients receiving CHOP do not suffer CINV.
Aapro et al. reported that guideline-consistent antiemetic therapy alleviates CINV significantly. 5 Healthcare professionals should consider using recommendations from guidelines.
Assuming that the effectiveness of the antiemetic drugs was similar to that in previous reports (i.e., in patients receiving both intravenous HEC and the recommended antiemetic, the frequency of vomiting was 30%) and that 90% of the patients received the antiemetic prophylaxis for HEC, the frequency of vomiting in the patients who received HEC without antiemetic prophylaxis was approximately twice that of the patients who took HEC with prophylaxis. Furthermore, considering that more than 90% of patients with non-small cell lung, breast, and oesophageal cancers who received HEC with the rec- patients with a high-to-moderate risk of CINV. 8 Meanwhile, the MASCC guideline recommends a two-drug combination of a serotonin receptor antagonist and dexamethasone for the same patients. 37 This study simply describes the current status of prophylactic antiemetic drugs prescribed for patients receiving oral chemotherapy. These findings may be used for further research on the appropriate antiemetic therapy for these patients.
This study had some limitations. First, this study did not measure the frequency of CINV due to unavailability of the CINV incidence data. Therefore, we estimated the frequency using data from a previ-

| CONCLUSION
Overall, HEC was less prescribed than MEC, and prophylactic antiemetic drugs were generally prescribed. Healthcare professionals should educate patients about emetic risks before chemotherapy initiation to avoid patients' concerns about CINV that may lead to treatment termination. This type of survey should be repeated to observe the improvements in compliance with recommended anti-emetic therapy.