Multiple-day dexamethasone for controlling delayed chemotherapy-induced nausea and vomiting with low-emetic-risk chemotherapy

Abstract

Background

Chemotherapy-induced nausea and vomiting (CINV) is a distressing adverse effect of chemotherapy that requires aggressive prevention. This study aimed to evaluate the antiemetic efficacy of a 4-day dexamethasone (DEX) regimen compared with the guideline-recommended 1-day DEX regimen against delayed CINV in patients receiving low-emetic-risk chemotherapy (LEC).

Methods

Pooled patient-level data from a multicenter, prospective, observational study, including 132 patients receiving LEC, divided into 1-day and 4-day DEX groups, were analyzed. An inverse probability of treatment weighting (IPTW) model was applied to balance patient characteristics. The primary endpoint was the incidence of delayed CINV, with secondary endpoints including symptom severity, food intake, and daily incidence patterns.

Results

After IPTW adjustment, the incidence of delayed nausea was significantly lower in the 4-day DEX group (3.19%) than in the 1-day group (22.78%, p = 0.001), with no significant difference in acute nausea between the groups (14.49% vs. 5.83%, p = 0.167). In a multivariate model, both female sex and the 1-day DEX regimen were independently associated with a higher risk of delayed CINV. The 4-day DEX group showed favorable trends in nausea incidence, severity, and food intake regardless of the presence of acute CINV.

Conclusions

Although 1-day DEX remains the standard approach recommended by antiemetic guidelines, a 4-day DEX regimen shows promise as an alternative strategy for managing delayed nausea, especially in patients who experience acute CINV or have inadequate symptom control.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-026-15555-9.

Introduction

Chemotherapy-induced nausea and vomiting (CINV) is a distressing adverse effect of chemotherapy that requires aggressive prevention. Interventions for controlling CINV have improved substantially with the introduction of new antiemetics and the development of clinical guidelines. However, the optimal antiemetic prophylaxis for low-emetic-risk chemotherapy (LEC) has not yet been established. Antiemetic therapy for cancer chemotherapy is classified according to the emetogenic risk of anticancer agents based on international guidelines. The Japan Society of Clinical Oncology [1] and the American Society of Clinical Oncology [2] guidelines recommend single-agent therapy with either a 5-HT₃ receptor antagonist (5-HT₃RA) or dexamethasone (DEX) for LEC. The Multinational Association of Supportive Care in Cancer/European Society for Medical Oncology guidelines [3] recommend the use of a 5-HT₃RA, DEX, or a dopamine receptor antagonist as a single agent. The National Comprehensive Cancer Network guidelines [4] recommend using 5-HT₃RA, DEX, procarbazine, or a dopamine receptor antagonist as a single agent. International guidelines [1–4] typically advise single-dose agents (e.g., DEX) on day 1 for LEC, without mandatory prophylaxis for the delayed phase. This approach, however, relies more on expert consensus than on robust data from randomized controlled trials or clinical studies. Previous reports showed no significant difference in antiemetic efficacy between the guideline-recommended single-dose DEX regimen on day 1 alone and combinations of multiple antiemetic agents [5, 6]. A previous report published by our group [7] found that 27.3% of patients receiving single-day antiemetic monotherapy still experienced delayed nausea; the severity of nausea and anorexia peaked around days 4 and 5 post-chemotherapy [7]. Management of delayed CINV remains difficult. As delayed CINV adversely affects quality of life, effective preventive strategies are needed [8]. The administration of DEX after day 2 of chemotherapy has been reported to alleviate delayed nausea and anorexia in high- and moderate-emetic-risk chemotherapy regimens [9–11]. Therefore, in this study, we compared the antiemetic efficacy of 1-day and 4-day DEX regimens to identify effective prophylaxis for delayed CINV in patients receiving LEC.

Methods

Study design

This study utilized a pooled analysis of individual patient data from a prospective, multicenter observational study conducted in Japan (UMIN000020800) [6]. Participants were patients scheduled to receive LEC regimens. The protocol was approved by the Independent Ethics Committee of the National Hospital Organization Kyushu Cancer Center (No. 2013-79). The research was performed in compliance with the Declaration of Helsinki and the Ethical Guidelines for Medical and Health Research Involving Human Subjects, and all participants provided written informed consent prior to participation in the study.

Patients and data collection

As the detailed methodology for patient recruitment and data acquisition has been published previously [6], it is briefly summarized here. Eligible participants were LEC-naïve adults (aged ≥ 20 years). Patients were scheduled to receive LEC regimens containing one or more specified injectable agents (e.g., docetaxel, paclitaxel, gemcitabine). Key exclusion criteria, also detailed previously [6, 7], included chronic systemic corticosteroid use, concurrent abdominopelvic radiotherapy, LEC administration within 120 h, brain metastases, or vomiting within 24 h preceding chemotherapy. As described in the original studies [6, 7], participants used a patient diary to self-record digestive symptoms (nausea, vomiting, and food consumption) for 5 days post-LEC. Nausea severity and food intake were assessed using a 100-mm visual analogue scale. Alcohol consumption was defined as consumption several times per week. A history of motion sickness was determined using patient interview data included in the pooled dataset [6].

Statistical analysis

An inverse probability of treatment weighting (IPTW) model based on logistic regression was used to adjust for observable characteristics between the 1-day and 4-day DEX groups, including sex, age, history of nausea, history of vomiting, alcohol consumption, motion sickness, morning sickness, opioid use, and chemotherapy (taxane). The observed incidence rates of nausea and vomiting were compared between groups using the chi-square test. Risk factors for delayed CINV were evaluated using a logistic regression model adjusted for sex, age, alcohol consumption, chemotherapy (taxane), and 1-day DEX. A two-sided p-value of < 0.05 was considered statistically significant. All analyses were performed using SAS 9.4 (SAS Institute, Cary, NC, USA).

Results

Patient characteristics

A total of 132 patients receiving LEC were included in the analysis. Baseline characteristics—sex, age, alcohol consumptions, history of motion sickness, morning sickness, nausea, vomiting, opioid use, and taxane—as well as unweighted and weighted characteristics stratified by the number of days of DEX administration are summarized in Table 1. After IPTW (ATT) adjustment, the effective sample sizes (ESS) were 89 for the 1-day DEX group and 94.5 for the 4-day DEX group. The distribution of IPTW weights (histogram, IQR/median) is shown in Supplementary Fig. 1. Although a slight imbalance was noted in the history of nausea, the standardized mean differences (SMDs) for the remaining factors were balanced between the groups after IPTW adjustment. DEX was administered intravenously at a dose of 8–12 mg on day 1 and orally at 4–8 mg from day 2 onward.

Table 1.

Unweighted and weighted baseline characteristics of patients

Characteristics	Unweighted					Weighted
	1-day (n = 89)		4-day (n = 43)		SMD	1-day	4-day	SMD
	N	%	N	%		%	%
Sex
Female	52	58.4	38	88.4	0.720	58.4	31.7	-0.267
Male	37	41.6	5	11.6		41.6	68.3
Age
<65	35	39.3	38	88.4	1.187	39.3	36.2	0.031
≥65	54	60.7	5	11.6		60.7	63.8
Alcohol consumption
No	60	67.4	18	41.9	0.531	67.4	65.3	0.022
Yes	29	32.6	25	58.1		32.6	34.7
Motion sickness
No	73	82.0	28	65.1	0.391	82.0	91.2	-0.092
Yes	16	18.0	15	34.9		18.0	8.8
Morning sickness
No	66	74.2	27	62.8	0.246	74.2	84.7	-0.105
Yes	23	25.8	16	37.2		25.8	15.3
Nausea history
No	68	76.4	17	39.5	0.805	76.4	37.8	0.386
Yes	21	23.6	26	60.5		23.6	62.2
Vomiting history
No	79	88.8	37	86.0	0.082	88.8	92.5	-0.037
Yes	10	11.2	6	14.0		11.2	7.5
Opioid use
No	79	88.8	43	100.0	0.503	88.8	100.0	0.112
Yes	10	11.2	0	0.0		11.2	0
Chemotherapy (taxane)
No	36	40.4	2	4.7	0.948	40.5	55.1	-0.146
Yes	53	59.6	41	95.3		59.5	44.9

SMD, standardized mean difference

Incidence of nausea and vomiting

The incidence of nausea and vomiting after IPTW adjustment is shown in Fig. 1. The incidence of delayed nausea was significantly lower in the 4-day DEX group than in the 1-day DEX group (1.5% vs. 23.6%, p < 0.001), with an odds ratio of 0.05 (95% confidence interval: 0.009–0.279). No significant difference in acute nausea was observed between the groups (13.5% vs. 4.4%, p = 0.057). Vomiting occurred only in the 1-day DEX group during the delayed phase (3.4% vs. 0%, p = 0.223).

Fig. 1.

Incidence of chemotherapy-induced nausea and vomiting. Incidence of nausea (a) and vomiting (b) during the acute and delayed phases

As balance in nausea history was not sufficiently achieved even after IPTW adjustment, the incidence of CINV was examined after stratification by nausea history. Irrespective of nausea history status, the 1-day DEX group consistently demonstrated a higher incidence of delayed CINV than the 4-day DEX group (Supplementary Table 1).

Black bars represent the 1-day dexamethasone (DEX) group, and white bars represent the 4-day DEX group. After IPTW adjustment, the effective sample sizes (ESS) were 89 for the 1-day DEX group and 94.5 for the 4-day DEX group. The incidence of delayed nausea was significantly lower in the 4-day DEX group than in the 1-day DEX group (p = 0.001).

Risk factors for delayed CINV

Univariate and multivariate logistic regression analyses were conducted with sex, age, alcohol consumptions, chemotherapy (taxanes), and 1-day DEX as covariates for delayed CINV (Table 2). Female sex and the 1-day DEX were identified as factors associated with delayed CINV.

Table 2.

Risk factors for delayed chemotherapy-induced nausea and vomiting

Characteristics	Univariate			Multivariate
	OR	95%CI	p	OR	95%CI	p
Sex, female	10.218	2.909–35.891	< 0.001	6.352	1.860–21.697	0.003
Age, < 65	1.005	0.404–2.502	0.990	1.776	0.589–5.353	0.308
Alcohol consumption, no	1.750	0.627–4.885	0.285	1.752	0.558–5.503	0.337
Chemotherapy, taxane	1.169	0.481–1.070	0.730	1.272	0.433–3.734	0.662
DEX, 1-day	19.957	3.585–111.094	< 0.001	13.931	2.418–80.256	0.003

OR, odds ratio; CI, confidence interval; DEX, dexamethasone

Pattern of incidence of nausea

Figure 2 presents the incidence of nausea. From day 2 onward, the incidence was lower in the 4-day DEX group than in the 1-day DEX group.

Fig. 2.

Pattern of nausea occurrence. Solid and dotted lines represent the 1-day and 4-day dexamethasone groups, respectively. From day 2 onward, the incidence of nausea was lower in the 4-day group than in the 1-day group

Pattern of nausea incidence by presence or absence of acute CINV

The pattern of nausea incidence from day 2 onward, stratified by the presence or absence of acute CINV, is shown in Fig. 3. In both subgroups, the incidence of nausea tended to be lower in the 4-day DEX group.

Fig. 3.

Analysis stratified by presence or absence of acute chemotherapy-induced nausea and vomiting (CINV). The incidence pattern of nausea from day 2 onward, stratified by the presence (a) or absence (b) of acute CINV. Solid and dotted lines represent the 1-day and 4-day dexamethasone groups, respectively

Trends in nausea severity and food intake, stratified by presence or absence of acute CINV

Nausea severity and food intake from day 2 onward, stratified by the presence or absence of acute CINV, are presented in Fig. 3. Among patients with acute CINV, nausea severity on days 4 and 5 was milder in the 4-day DEX group (Fig. 4). Food intake also showed a more favorable trend in this group (Fig. 5). In patients without acute CINV, both nausea severity and food intake also tended to be better in the 4-day DEX group, although the differences were less pronounced (Figs. 4 and 5).

Fig. 4.

Trends in nausea severity, stratified by the presence or absence of acute chemotherapy-induced nausea and vomiting (CINV). Severity of nausea from day 2 onward, stratified by the presence (a) or absence (b) of acute CINV. Solid and dotted lines represent the 1-day and 4-day dexamethasone groups, respectively. Daily mean visual analog scale (VAS) scores of nausea severity (100 mm, worst nausea; 0 mm, no nausea) and food intake (100 mm, no oral intake; 0 mm, eating as usual) on days 2–5 after chemotherapy

Fig. 5.

Trends in decreased food intake, stratified by the presence or absence of acute chemotherapy-induced nausea and vomiting (CINV). Decrease in food intake from day 2 onward, stratified by the presence (a) or absence (b) of acute CINV. Solid and dotted lines represent the 1-day and 4-day dexamethasone groups, respectively. Daily mean visual analog scale (VAS) scores of nausea severity (100 mm, worst nausea; 0 mm, no nausea) and food intake (100 mm, no oral intake; 0 mm, eating as usual) on days 2–5 after chemotherapy

Discussion

This study compared the efficacy of 4-day DEX prophylaxis with the guideline-recommended 1-day DEX in preventing delayed CINV in patients receiving LEC. The 4-day DEX group demonstrated significantly better control of delayed nausea than the 1-day DEX group. In the multivariate analysis, 1-day DEX emerged as a factor associated with delayed CINV, as did female sex, a widely recognized patient-related risk factor.

In the original prospective multicenter observational study [6], the guideline-inconsistent practice (GICP) group—defined by antiemetic intensification—did not demonstrate a clear additive benefit. However, the GICP group with intensified antiemetic prophylaxis was heterogeneous, comprising diverse regimens (e.g., 5-HT₃RA + 1-day DEX, 5-HT₃RA + multiple-day DEX, multiple-day DEX alone). In contrast, in the current study, we excluded patients receiving single-day administration of multiple antiemetics and narrowed the analysis specifically to patients receiving 1-day versus 4-day DEX. This focused approach clarified the benefit of extended DEX administration, particularly in patients who experienced acute CINV.

The presence of acute CINV is an established predictor of delayed CINV [12, 13]. Our stratified analysis indicated that the 4-day DEX regimen was particularly beneficial for individuals who experienced acute CINV, suggesting that it may be a valuable strategy for this high-risk subgroup.

Additionally, recognizing that a history of nausea has been identified as a risk factor for CINV in prior research [6], we conducted a stratified analysis of delayed CINV incidence based on the presence or absence of a history of nausea to account for potential confounding. Despite the limited number of events, the 1-day DEX group exhibited consistently higher incidence rates of delayed CINV compared with the 4-day DEX group, irrespective of nausea history.

While international guidelines recommend only single-day antiemetic prophylaxis for LEC, our findings highlight a potential gap between guideline recommendations and real-world patient experiences. In our study, patients in the 1-day DEX group experienced a delayed nausea incidence of nearly 23%. This suggests that the “low emetic risk” classification may underestimate the burden of delayed symptoms, particularly nausea, which is often more difficult to control than vomiting. The significant reduction in delayed nausea observed with the 4-day DEX suggests that, for a subset of patients, the standard 1-day DEX prophylaxis is insufficient to maintain quality of life during the days following chemotherapy.

The mechanism by which DEX prevents CINV remains unclear [14]; however, its prophylactic effect—particularly against nausea—has been demonstrated in several clinical studies [14, 15]. Consistent with our findings, other clinical studies have suggested that extending DEX administration beyond day 1 can help mitigate delayed nausea as well as related symptoms such as anorexia and fatigue [9, 10]. The potential benefits of prolonged DEX must be weighed against its known adverse effects, including insomnia, dyspepsia, agitation, and hyperglycemia, which may occur with extended use [16, 17].

Delayed CINV is largely mediated by central pathways involving the neurotransmitter substance P, a mammalian tachykinin peptide; therefore, neurokinin-1 receptor antagonists (NK1RAs) are considered therapeutic candidates [18]. Although NK1RAs are effective against vomiting, their efficacy in controlling nausea is limited [19, 20]. While palonosetron is an effective, though costly, agent for CINV prevention [21], it does not typically improve associated anorexia or fatigue. This observation aligns with previous reports suggesting that corticosteroids may exert a broader therapeutic effect on the “symptom cluster” of nausea, fatigue, and appetite loss, possibly through anti-inflammatory mechanisms or modulation of blood–brain barrier permeability, thereby offering advantages that specific receptor antagonists alone may not provide.

The clinical implications of our findings advocate for a risk-stratified approach rather than a uniform protocol change for all LEC patients. Given the potential adverse effects of extended corticosteroid use, such as insomnia or hyperglycemia, the 4-day DEX should be targeted. Our multivariate analysis identified female sex as a factor associated with CINV, and stratified analysis confirmed the benefit of extended DEX specifically in patients who experienced acute CINV. Therefore, clinicians might consider a step-up strategy: adhering to the standard 1-day regimen for low-risk patients, while preemptively prescribing or escalating to a 4-day DEX for women or for those who experience CINV in the acute phase. Although safety data were not directly assessed in this pooled analysis, the SPARED trial [9], which compared the administration of DEX for four days versus one day as part of a four-drug antiemetic regimen for cisplatin-based chemotherapy, reported no clinically significant difference in adverse events with 4-day DEX, supporting the feasibility of this targeted approach.

This study has some limitations. First, it was not a blind, randomized controlled trial; therefore, causality could not be established. Second, IPTW adjustment resulted in a slight imbalance in patient background characteristics, such as a history of nausea. Although consistent results were obtained in the analysis stratified by history of nausea, it is possible that a history of nausea confounds the association between the duration of DEX administration and delayed CINV. Moreover, the small ESS may have affected the stability of the treatment effect estimation. Third, the safety profile could not be assessed because adverse event data were unavailable in the pooled dataset. However, in the SPARED trial [9], no clinically significant differences were observed between the two groups with respect to adverse effects of concern associated with DEX. Therefore, although safety data are not available in the present study, we believe that 4-day DEX is unlikely to cause serious problems. Fourth, since this cohort consisted mainly of patients treated with taxanes (particularly docetaxel), extrapolation of these results to non-taxane LEC regimens should be done with caution. Finaly, the study focused solely on the 4-day DEX regimen; therefore, the efficacy of shorter DEX courses remains uncertain. Although international guidelines do not recommend prophylaxis for delayed CINV in patients receiving LEC, the risk of delayed CINV should be carefully considered. The 4-day DEX regimen appears to be a viable option for managing delayed nausea associated with LEC, particularly in patients with poor nausea control or preceding acute CINV.

In conclusion, while 1-day DEX remains the standard approach for LEC, as recommended by antiemetic guidelines, a 4-day DEX represents a promising alternative for managing delayed nausea.

Abbreviations

NK1RAs

neurokinin-1 receptor antagonists

CINV

chemotherapy-induced nausea and vomiting

LEC

low-emetic-risk chemotherapy

5-HT₃RA

5-HT₃ receptor antagonist

DEX

dexamethasone

IPTW

inverse probability treatment weighting

ESS

effective sample size

Authors’ contributions

Conceptualization: T.H.; Analysis of the claims data: A.Y., T.H. and M.S.; Statistical analysis: M.S.; Data interpretation and manuscript preparation: A.Y., T.H., M.S., S.U., C.S., T.M., Y.K., and T.E.; Initial manuscript drafting: A.Y., T.H., and M.S.; Manuscript critical revision: A.Y., T.H., M.S., Y.H., Y.N., M.U., S.U., C.S., T.M., Y.K., K.K., K.M. and T.E. All authors reviewed the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data that support the findings of this study are available from the study group. However, restrictions apply to the availability of these data, as they were used under a license for the current study and are therefore not publicly available. The data are, nevertheless, available from the corresponding author, Toshinobu Hayashi (toshinobu@fukuoka-u.ac.jp), upon reasonable request and with permission from the study group.

Declarations

Ethics approval and consent to participate

The study was approved by the Independent Ethics Committee of National Hospital Organization Kyushu Cancer Center (2013-79) and conducted in accordance with the Declaration of Helsinki and the Ethical Guidelines for Medical and Health Research Involving Human Subjects. Written informed consent was obtained from the participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.