Authors: Kenichi Suzuki, Takashi Yokokawa, Takashi Kawaguchi, Shinya Takada, Shinya Tamaki, Yohei Kawasaki, Takumi Yamaguchi, Kei Koizumi, Takuma Matsumoto, Yukio Sakata, Yuichiro Arakawa, Hideaki Ayuhara, Mari Hosonaga, Masakazu Yamaguchi, Daiki Tsuji
Categories: Article, Breast cancer, CINV, Cancer, Dexamethasone, Medical research, Olanzapine, Oncology, Phase II, Signs and symptoms
Source: Scientific Reports
Dexamethasone is an antiemetic drug widely used to prevent nausea and vomiting caused by anticancer drugs. However, dexamethasone can cause several side effects even after short-term administration. Therefore, the development of dexamethasone-free antiemetic therapies has been recognized as an important challenge. The objective of this study was to investigate the efficacy and safety of palonosetron, aprepitant, and olanzapine. Patients who were chemotherapy-naïve and scheduled to receive highly emetogenic chemotherapy for breast cancer were enrolled and assessed for nausea and vomiting occurring within 120 h after the start of chemotherapy. The primary endpoint was the total control (TC) rate of overall phases. Secondary endpoints included the complete response (CR) rate, which was evaluated during the acute, delayed, and overall phases. A total of 88 patients were enrolled from eight centers in Japan, of whom 84 were included in the analysis. The proportion of patients achieving TC throughout the overall period was 17.1%. Similarly, CR and CC rates for the overall period were 43.4% and 39.5%, respectively. Frequently reported adverse events were loss of appetite and constipation, with rates of 52.4% and 50.0%, respectively. The primary endpoint was not achieved. Therefore, antiemetic therapy without dexamethasone shows an inadequate effect on nausea, and it is generally advisable to avoid omitting dexamethasone. However, in the overall period, both CR and CC were comparable to conventional three-drug combination therapy. Thus, in patients unable to use dexamethasone, replacing it with olanzapine could be an option.
Trial registration number: UMIN 000038644, November 20, 2019. The date of first trial 13/03/2020.
Keywords: CINV, Olanzapine, Breast cancer, Dexamethasone, Phase II
**Subject ** Cancer, Medical research, Oncology, Signs and symptoms
A common side effect of cancer chemotherapy is chemotherapy-induced nausea and vomiting (CINV), and nausea is considered the symptom of greatest concern, particularly for patients^1^. In Japan, in the 1990s, 5-hydroxytryptamine 3 receptor antagonists (5HT3RAs) were introduced clinically, followed by the availability of neurokinin-1 receptor antagonists (NK1RAs) in 2009. Over the approximately 30 years since then, antiemetic effects in highly emetogenic chemotherapy (HEC) have clearly improved^2^. In 2010, palonosetron (PALO), a long-acting 5HT3RA, was launched in Japan. PALO has improved the control of nausea throughout the regimen and the complete response (CR) rate of delayed emesis specific to cisplatin (CDDP)-based regimens^3^. PALO is now preferred over the first-generation 5HT3RA granisetron (GRA) as the standard antiemetic therapy. Furthermore, in 2020, Hashimoto et al. demonstrated the antiemetic effect of four-drug combination therapy consisting of PALO, NK1RA, dexamethasone (DEX), and olanzapine (OLN) in CDDP, making this four-drug regimen widely used as the standard antiemetic therapy for HEC domestically and internationally^4^. For HEC, such as the combination therapy of anthracyclines and cyclophosphamide commonly used in breast cancer chemotherapy, many guidelines recommend the concomitant use of high doses of DEX along with 5HT3RA, NK1RA, and olanzapine^5,6^. However, concerns exist regarding DEX, including increased appetite and insomnia, and adverse effects such as decreased bone density and increased susceptibility to infections have been reported with short-term administration of DEX for antiemetic purposes in moderately emetogenic chemotherapy (MEC) for gastrointestinal cancer^7,8^. In cases of hepatitis, a two-drug combination antiemetic therapy (5HT3RA, NK1RA) without DEX has been implemented to avoid the recurrence of hepatitis caused by DEX, but a decrease in the CINV control rate was suggested^9^. Furthermore, cancer immunotherapy has recently been rapidly introduced for many cancer types. The effectiveness of immune checkpoint inhibitors (ICIs) has been confirmed in recurrent/metastatic triple-negative breast cancer and in neoadjuvant chemotherapy for triple-negative breast cancer, and they are already in clinical use. However, in cancer chemotherapy including ICIs, DEX, which inherently has immunosuppressive effects, is incorporated as standard antiemetic therapy, potentially counteracting the effects of ICIs when used concurrently. However, because DEX has immunosuppressive effects, such as suppressing T-cell activity^10^, it may attenuate the antitumor effects of ICIs by affecting immune function. Currently, the American Society of Clinical Oncology (ASCO) guidelines do not recommend omitting dexamethasone (DEX) when using ICIs in cancer chemotherapy^6^. However, recent reports have shown that early administration of corticosteroids shortens survival in patients with low tumor mutation burden^11^. Therefore, the development of antiemetic therapies that omit DEX as much as possible is an important challenge. A clinical trial comparing the efficacy and safety of three-drug combination antiemetic therapy including PALO, NK1RA, and OLN, with DEX omitted and replaced with OLN, was conducted based on concerns about adverse effects and the impact on ICIs. NK1RA, aprepitant (APR), and PALO, comparing the 3-day DEX administration group with the 1-day administration group, reported that the two groups showed comparable antiemetic effects^12^. In addition, it has been suggested that OLN, a medication used for conditions such as schizophrenia, is effective against CINV in several previous studies and has been available as an antiemetic in Japan since 2017^13^. In 2024, Minatogawa investigated the antiemetic efficacy of four-drug combination therapy including olanzapine (OLN), in which dexamethasone (DEX) was reduced to a single day, targeting cisplatin^14^. Recently, there has been increasing interest in omitting DEX because of concerns about its side effects. Given OLN’s potent antiemetic effects, this prospective, multicenter, phase II trial was carried out to evaluate the antiemetic efficacy and safety of omitting DEX from the traditional three-drug combination therapy, instead using a combination of palonosetron (PALO), aprepitant (APR), and OLN.
This multicenter, phase II trial aimed to investigate the efficacy and safety of triplet antiemetic therapy with PALO, APR, and OLN in patients scheduled to undergo HEC as their initial treatment for malignant tumors, specifically breast tumors.
This study was a collaborative effort involving eight domestic institutions. Approval for this research was obtained from the ethics review committees at the Cancer Institute Hospital of JFCR and each participating facility. This study was registered with the University Hospital Medical Information Network (UMIN) in Japan with the registration number UMIN 000038644. In addition, an independent Data and Safety Monitoring Committee oversaw the monitoring of efficacy and safety throughout this clinical trial.
The phase II trial was conducted from July 2019 to June 2022. Patients scheduled to undergo initial HEC treatment for breast cancer were enrolled.
Other eligibility criteria included age 20 years or older, plans for the administration of first-line chemotherapy at standard doses, no restriction on treatment history, and Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0–2. The eligible regimens were as AC (doxorubicin + cyclophosphamide, including dose-dense AC), EC (epirubicin + cyclophosphamide, including dose-dense EC), CAF (cyclophosphamide + doxorubicin + 5-fluorouracil), FEC (5-fluorouracil + epirubicin + cyclophosphamide), and TAC (docetaxel + doxorubicin + cyclophosphamide, with the condition that dexamethasone was not used as pre-medication before docetaxel), all of which are classified as HEC in several antiemetic guidelines^5,6,15^.
Patients were required to have sufficient organ function, with the following results within the two weeks preceding study alanine aminotransferase less than 100 IU/L; aspartate aminotransferase less than 100 IU/L; total bilirubin concentration less than 2.0 mg/dL; and creatinine clearance calculated by the Cockcroft–Gault formula of 55 mL/min or more. Written, informed consent was obtained from all patients. The exclusion criteria for this study were as history of allergy to the drugs and similar compounds used in this study; steroid drugs, excluding inhaled and topical steroids; evident vomiting symptoms, such as with brain metastasis or severe gastrointestinal passage disorders; symptomatic ascites or pleural effusion requiring therapeutic puncture; gastrointestinal passage disorders, such as gastric outlet obstruction or intestinal obstruction; seizure disorders requiring treatment with antiepileptic drugs; currently receiving adrenaline or pimozide; started treatment with strong opioid drugs within the 48 h before registration; abdominal or pelvic radiotherapy within the 6 days before registration or planned to receive it within 6 days after initiating anticancer drug administration; regular use of antiemetic drugs with antiemetic effects other than the investigational drug; pregnant, lactating, or potentially pregnant women or patients not intending to use contraception; diabetes mellitus treated with antidiabetic drugs or an HbA1c (NGSP) level of 6.5% or higher (6.1% or higher for JDS) within the last 28 days before registration; smoking regularly, although eligible if they quit smoking 1 week before registration; difficulties performing operations such as electronic data capture (EDC) input; or deemed unsuitable for this trial by the principal investigator for other reasons.
All eligible patients were administered PALO 0.75 mg (intravenous injection 30 min before the start of chemotherapy on day 1), OLN 5 mg (oral administration once daily at bedtime from day 0 to day 4, i.e., for a total of 5 days from bedtime on the day before chemotherapy to 4 days after chemotherapy), and APR (oral administration of 125 mg 60 min on day 1, before the start of chemotherapy and 80 mg orally on the morning of day 2 and of day 3) or fosaprepitant (intravenous injection of 150 mg 60 min on day 1, before the start of chemotherapy). If nausea, vomiting, or retching occurred within 120 h after chemotherapy and at the patient’s request, additional rescue antiemetic medication was allowed for these symptoms. Rescue antiemetic medications, such as domperidone or metoclopramide, were administered by healthcare providers at each facility according to the study protocol. Because OLN has the side effect of sedation, it is preferable to administer it after dinner or before bedtime to avoid daytime drowsiness or dizziness. However, when not using DEX, there is a possibility of inadequate control if OLN is administered after dinner or before bedtime on the first day of chemotherapy. In addition, it has been suggested that the incidence of acute-phase CINV is higher than that of delayed-phase CINV in AC/EC therapy^15^. Therefore, to expect sufficient effects from the first day, considering the half-life (approximately 28 h) of OLN, OLN was administered before bedtime on the day before chemotherapy. Furthermore, it has been suggested that pre-chemotherapy insomnia affects the onset of CINV^16^, and taking medication from the day before treatment starts is also considered rational for avoiding insomnia.
The chemotherapy targeted in this study was administered in outpatient settings. Therefore, as specified in the research protocol, all patients had episodes of nausea, vomiting, and the use of rescue medication meticulously recorded using EDC for 120 h after chemotherapy. The data center is located in the Department of Clinical Assessment, School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, Tokyo, Japan. Enrolment, data collection, and monitoring were performed using the EDC system Viedoc me (Viedoc Technologies). Data on patient-reported vomiting or nausea status and the use of rescue medication were collected electronically from patients via an electronic device. Data entry to the electronic case report form was performed by investigators using EDC at each site. Patient-reported outcome (PRO) data were collected electronically from patients through an electronic tablet device. No personally identifiable information was entered into the EDC, and the data center did not collect personal information.
Rescue antiemetic medications such as metoclopramide, domperidone, alprazolam, and dexamethasone were prescribed on the day of chemotherapy or before, following the usual procedures at each facility. In addition, adverse events were evaluated using the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 and Patient-Reported Outcomes version of the CTCAE (PRO-CTCAE) version 1.0. Patients recorded predetermined PRO-CTCAE items (constipation, diarrhea, loss of appetite, hiccups, insomnia, and drowsiness) in the EDC on the day before treatment initiation and on the 7th day after treatment initiation, as outlined in the research protocol. In addition, this study adopted “composite grading” in the PRO-CTCAE^17^.
The primary endpoint of this study was the total control (TC) rate of nausea and vomiting throughout the overall phase (0–120 h after chemotherapy), defined as the absence of nausea, vomiting, and the use of rescue medication. Secondary endpoints included the TC rate during the acute phase (0–24 h after chemotherapy) and the delayed phase (24–120 h after chemotherapy), CR of vomiting (defined as no vomiting episodes and no use of rescue medication), complete control (CC) of nausea and vomiting (defined as no vomiting episodes, no use of rescue medication, and no moderate or severe nausea), and the proportions of “no nausea” during the acute phase, delayed phase, and overall phase. The severity of nausea was measured on a 4-point Likert scale (0, no nausea; 1, mild; 2, moderate; 3, severe)^18^. The other secondary endpoints included time to treatment failure (TTF; defined as the time to the first vomiting episode or administration of rescue medication) and treatment-related adverse events.
The TC rate of the overall phase with chemotherapy, using the standard antiemetic therapy of PALO, APR, and DEX, was reported as 23.2% in a Phase III trial conducted in Japan^12^. In addition, a Phase III trial investigating a four-drug combination antiemetic therapy, which included adding OLN to PALO, APR, and DEX, was conducted outside Japan. The proportion of “no nausea” in the overall phase was reported to be 37.3%, showing an improvement of 15.4% compared with the non-OLN combination group^13^. In Japan, although large-scale trials implementing four-drug combination antiemetic therapy targeting anthracycline-based chemotherapy and cyclophosphamide have not been conducted, observational studies with 92 patients reported a CC rate of approximately 40%, showing an improvement of approximately 10% compared with the non-OLN combination group^19^. According to these reports, the threshold TC rate in the overall phase was set at 23%, with an expected TC rate of 38%, a one-sided significance level (α error) of 0.025, and power of 80%, resulting in a required sample size of 74 cases. In this trial, the administration of OLN before bedtime on day 0 was likely to be performed by the patients themselves at home, raising concerns about medication non-adherence. Therefore, considering a dropout rate of approximately 20%, the target registration sample size was set at 89.
For the primary endpoint, the TC rate during the overall phase, along with its 95% confidence interval, was calculated. In addition, a one-sided binomial test was conducted to verify the null hypothesis that the TC rate is less than 23% and the alternative hypothesis that it is greater than or equal to 23%. The significance level for the one-sided test was set at 2.5%. For secondary endpoints, the number and proportion, along with the 95% confidence interval, were calculated for binary data according to the characteristics of the data. For survival time data, the number of events and censored events were calculated, and survival curves were estimated using the Kaplan–Meier method. Furthermore, the 25th percentile, median, 75th percentile values, and 95% confidence interval for the time to event occurrence based on the Kaplan–Meier method were estimated. Comparison of TC rate and adverse events classified by age in 10-year increments were used Fisher’s exact test.
From eight facilities in Japan, 89 breast cancer patients scheduled for HEC, anthracycline-containing regimens, were registered. Of them, 5 cases were excluded from the analysis of efficacy and safety because of reasons such as insufficient records in ePRO. Furthermore, 8 cases with incomplete implementation of the prescribed antiemetic therapy according to the protocol were excluded. A total of 76 cases with evaluable data for the primary endpoint were included in the full analysis set (FAS) (Table 1).
The TC rate in the overall phase, which was the primary endpoint, was 17.1% (95% CI, 9.4-27.5%; p = 0.275). Since the lower limit of the 95% CI fell below the pre-set criterion of 23%, the primary endpoint was not met. The TC rates in the acute and delayed phases were 25.0% and 26.3%, respectively (Table 2). The CR rates in the overall, acute, and delayed phases were 43.4%, 53.9%, and 63.2%, respectively. Similarly, the CC rates were 39.5%, 47.4%, and 57.9% during the overall, acute, and delayed phases, respectively. The TTF curve 120 h after chemotherapy initiation was estimated using the Kaplan–Meier method (Fig. 1).
Fig. 1 Kaplan–Meier curve of time to treatment failure.
In this study, there were 84 evaluable patients for safety assessment. There were no treatment discontinuations due to adverse events, and no unexpected severe adverse events related to antiemetic drugs occurred during the observation phase. The most common treatment-related adverse events were anorexia and constipation. Most adverse events were Grade 2 or lower, with only one each of Grade 3 anorexia and somnolence. Somnolence, considered to be due to OLN, was reported in 38.1% of cases (Table 3). In addition to adverse event assessment by medical staff using CTCAE ver5.0, patient-reported adverse event assessment based on PRO-CTCAE version 1.0 was conducted^20^. Patients were evaluated for adverse events related to symptoms experienced in the past 7 days, categorized by adverse event category, and rated on a 5-point scale on the day before chemotherapy initiation (day 0) and on the 7th day of treatment initiation (day 7). Subsequently, the difference in adverse event reporting before and after treatment was calculated using the subtraction method of baseline values^17^, and an overall grade was calculated by comparing PRO-CTCAE descriptions. In the assessment by PRO-CTCAE, the frequency of adverse events generally exceeded that of medical staff evaluations. Similarly, in the evaluation by PRO-CTCAE, there was a higher proportion of Grade 3 or higher adverse events, and Grade 4 was observed for five items except for insomnia. (Table 4)
Age is an important risk factor that significantly impacts the occurrence of CINV. Therefore, the eligible patients were divided by 10 years of age and analyzed for efficacy and adverse events. The TC rate, the primary endpoint of this study during the overall phase is shown in Fig. 2. The incidence of TC divided by age into 10-year intervals was 11.1%, 12.9%, and 25.9% for the groups aged 44 years and younger, 45–54 years, and 55 years and older, respectively, indicating that the younger the patient, the lower the TC rate, but not statistically significant. With regard to adverse events, the incidence of any grade of adverse events tended to be higher in younger patients for anorexia, diarrhea, and somnolence, which was statistically significant (Table 5).
Fig. 2 TC rates in 10-year age intervals over the overall period.
In the present study, OLN was administered before bedtime for 4 days starting from the day before chemotherapy. In clinical trials of treatments for CINV, the CR rate has long been used as a primary evaluation criterion. Although the CR rate is an objective measure, it does not consider patients’ subjective complaints of nausea. With advances in antiemetic therapy, the incidence of vomiting has decreased, and patients undergoing cancer chemotherapy are more concerned with improving nausea than vomiting. In recent clinical trials of CINV, several have used the proportion of “no nausea” as a primary endpoint^13^. Therefore, since nausea is of greater concern to patients receiving chemotherapy, this study chose the TC rate as the primary evaluation criterion. Unfortunately, in antiemetic therapy without DEX, nausea could not be adequately controlled, and the TC rate for the overall duration was 17.1%. Considering that the TC rate of the conventional standard three-drug combination therapy without OLN, set as the threshold in the present study, was 23%, antiemetic therapy without DEX is considered insufficient to achieve the target TC rate. However, in terms of the primary evaluation criterion of the CR rate, which has been emphasized in many previous studies, in the present study, it was 43.4% for the overall phase, which was almost equivalent to the CR rate of the conventional three-drug combination antiemetic therapy including DEX (41.1%^12^) reported in Japan for AC/EC therapy. Similarly, the delayed-phase CR rate was 63.2% (53.0%^8^), and the CC rate for the overall phase and the delayed phase were 39.5% (39.1%^12^) and 57.9% (52.3%^12^), respectively, all of which were comparable to or higher than those reported in previous Japanese studies^12^. One reason for the low control rate of nausea might be the patients’ lack of understanding of the use of rescue medications, and it is possible that the instructions given by pharmacists at the time of prescription were insufficient. In addition, although the severity of nausea was evaluated during the observation period, even mild nausea can cause a significant physical burden on patients if it persists for a long time. However, the present study did not include a temporal evaluation of nausea, which is considered a limitation of this research.
However, it is considered that this three-drug combination antiemetic therapy may be beneficial for patients in whom the use of DEX is undesirable, such as those with peptic ulcer disease, hepatitis B virus-positive patients, or elderly women at high risk of osteoporosis. Regarding adverse events, in the three-drug antiemetic therapy including OLN, Grade 3 anorexia and somnolence were observed in one patient each, but most symptoms were mild, Grade 2 or below. Sedation, believed to be caused by OLN, occurred in 38.1%. Because there is a suggested discrepancy between healthcare providers’ and patients’ evaluations of chemotherapy-induced nausea and vomiting, this study included patient self-assessments using PRO-CTCAE in addition to healthcare providers’ evaluations of adverse events. Similar to previous studies, PRO-CTCAE had a higher incidence and severity than healthcare providers’ assessments, highlighting the importance of careful observation.
In summary, although the present study did not demonstrate any advantages of antiemetic therapy replacing DEX with OLN due to the TC rate being below the threshold, since the CR rate and CC rate were equivalent to or higher than those of the three-drug combination therapy including DEX, this antiemetic therapy may be one of the treatment options for patients in whom DEX should be avoided.
The authors would like to acknowledge the contributions of all those who contributed to this article and all the participants. The authors would also like to thank FORTE Science Communications (https://www.forte-science.co.jp/) for English language editing.
All authors contributed to the study conception and design. Material preparation and data collection were performed by TK. YK and TY performed the statistical analysis. The first draft of the manuscript was written by KS and DT and TY. All authors read and approved the final manuscript.
The datasets used and analyzed in this study are available from the corresponding author upon reasonable request.
The authors declare no competing interests.
This study was conducted in accordance with Ethical Guidelines for Medical and Health Research Involving Human Subjects published by Japan’s Ministry of Education, Culture, Sports, Science and Technology and the Ministry of Health, Labour and Welfare, Japan, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Written, informed consent was obtained from all participants.
The datasets used and analyzed in this study are available from the corresponding author upon reasonable request.