Evaluating the potential to predict chemotherapy-induced nausea and vomiting in carboplatin-containing treatment based on symptoms induced by prior cisplatin-containing treatment.

Save

Chemotherapy-induced nausea and vomiting (CINV) is a serious adverse effect of cisplatin (CDDP)- and carboplatin (CBDCA)-containing treatments. In this study, we aimed to assess the potential to predict CINV in subsequent CBDCA-containing chemotherapy based on symptoms induced by prior CDDP-containing treatments. Patients with thoracic cancer who received CDDP followed by CBDCA treatments (n = 52) were divided into a control group, including patients with total control (TC) of CINV during prior CDDP-containing treatment period, and a CINV-experience group, including patients who did not achieve TC of CINV during the CDDP-containing treatment period.

Patients were retrospectively evaluated. The TC rates during all evaluation periods (0–120 h) were significantly lower in the CINV-experience group than in the control group (45.5% and 86.7%, respectively, P = 0.002), which met the primary endpoint. The incidence rates of all-grade nausea and anorexia were also significantly higher in the CINV-experience group.

In conclusion, our study suggests that CINV prediction in subsequent CBDCA-containing treatment based on the symptoms induced by prior CDDP-containing treatment is achievable.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-36488-0.

Introduction

Chemotherapy-induced nausea and vomiting (CINV) is one of the most significant adverse effects of chemotherapy that requires cautious monitoring and appropriate management. CINV can lead to undesirable symptoms, such as dehydration, electrolyte disorders, and malnutrition, significantly decreasing patient adherence to treatment1. Several guidelines define emetic risks of different chemotherapeutic agents, including molecular-targeted medicines, and recommend management strategies, such as prophylactic and counteractive interventions2–5.

Platinum agents, including cisplatin (CDDP) and carboplatin (CBDCA), play a critical role in the treatment of thoracic cancer in both the perioperative and metastatic phases6,7. Therefore, we sometimes encounter patients receiving both CDDP- and CBDCA-containing treatments in the treatment process. In these patients, CDDP is typically administered prior to CBDCA, taking into account the perioperative–recurrence progression and potent adverse effects of CDDP, such as CINV, nephrotoxicity, and hearing impairment.

According to the antiemetic guidelines, CDDP is classified as highly emetogenic chemotherapy (HEC), whereas CBDCA is classified as highly moderately emetogenic chemotherapy (MEC)2–4, although the National Comprehensive Cancer Network guidelines classify CBDCA as HEC5. Neurokinin-1 (NK1) and serotonin 3 (5-hydroxytryptamine 3; 5HT3) receptor antagonists are typically administered along with dexamethasone for CINV prophylaxis in HEC and high MEC regimens, whereas olanzapine has recently been recommended for CINV prevention in HEC regimens4. There is much evidence supporting the clinical practice in CINV management; however, there is limited information regarding the association between CINV and chemotherapeutic treatments.

Since efficient CINV management contributes to safer, less onerous, and more effective treatment, evaluating the potential to predict CINV with CBDCA-containing treatment based on symptoms induced by prior CDDP-containing treatment is of significant value.

In this study, we aimed to compare CINV development during CBDCA-containing treatment between patients who had and who had not experienced CINV during prior CDDP-containing treatments, to assess the potential to predict CINV in the subsequent treatment based on symptoms induced during the prior chemotherapeutic treatment in real-world thoracic cancer treatment settings.

Methods

Patients

Patients with thoracic cancer who received CBDCA-containing treatment between September 2014 and January 2024 at Hokkaido University Hospital were enrolled in this retrospective single-center observational study. All participants met the following baseline inclusion criteria (at the initial CBDCA-containing treatment cycle): (1) age ≥ 20 years, (2) Eastern Cooperative Oncology Group performance status (ECOG-PS) of 0–2, (3) receiving CDDP- followed by CBDCA-containing treatments, (4) acceptable laboratory renal and liver function for chemotherapy induction in clinical practice, and (5) receiving prophylactic CINV treatment consisting of aprepitant, dexamethasone, and palonosetron during both the CDDP- and CBDCA-containing treatment regimens. Patients who received regular corticosteroids or antiemetics, had baseline nausea, received a dose reduction of chemotherapy from treatment initiation, were transferred to another hospital during the first cycle, or with no sufficient data, in either setting were excluded.

The patients were divided into a control group, which included patients who achieved total control (TC; no emetic episodes, no rescue medication use, and no nausea) in the first cycle of a prior CDDP-containing treatment regimen between February 2018 and January 2024, and a CINV-experience group, which included those who experienced nausea and/or vomiting in the first cycle of CDDP-containing treatment between September 2014 and December 2023.

This study was approved by the Ethical Review Board for Life Science and Medical Research of Hokkaido University Hospital (approval number: 024–0478), and was performed in accordance with the Declaration of Helsinki and the STROBE statement. As the present study was retrospectively conducted, written informed consent from the participants was not mandated by the Ethical Review Board for Life Science and Medical Research of Hokkaido University Hospital.

Treatment methods

CBDCA (administered at an area under the blood concentration-time curve [AUC] of 5–6, day 1) + pemetrexed (PEM, 500 mg/m2, day 1), CBDCA (AUC 6, day 1) + nanoparticle albumin-bound paclitaxel (100 mg/m2, days 1, 8, 15), CBDCA (AUC 5, day 1) + etoposide (ETP, 80–100 mg/m2, days 1–3), or CBDCA (AUC 6, day 1) + paclitaxel (200 mg/m2, day 1) were intravenously administered every 3 weeks. The CBDCA dosage was calculated using the Calvert formula, and the maximum creatinine clearance, calculated using the Cockcroft-Gault formula, was set at 125 mL/min. Bevacizumab (15 mg/kg), atezolizumab (1200 mg/body), pembrolizumab (200 mg/body), or durvalumab (1500 mg/body) were co-administered every 3 weeks with the above-described CBDCA-based regimens.

Nivolumab (360 mg/body, every 3 weeks) and ipilimumab (1 mg/kg, every 6 weeks) were also concurrently administered with two cycles of CBDCA-based regimens. Prior CDDP-containing treatments comprised CDDP (80 mg/m2, day 1) + vinorelbine (20–25 mg/m2, days 1, 8) ± radiation; CDDP (75 mg/m2, day 1) + PEM (500 mg/m2, day 1) ± bevacizumab (15 mg/kg, day 1), atezolizumab (1200 mg/body, day 1), or pembrolizumab (200 mg/body, day 1); CDDP (80 mg/m2, day 1) + ETP (80 mg/m2, days 1–3); CDDP (60 mg/m2, day 1) + irinotecan (60 mg/m2, days 1, 8, 15); and CDDP (80 mg/m2, day 1) + gemcitabine (1000 mg/m2, days 1 and 8). All participants received prophylactic antiemetic treatments, including palonosetron 0.75 mg infusion on day 1, along with dexamethasone (9.9 mg intravenous on day 1, and 8 mg oral on days 2–4) and oral aprepitant (125 mg on day 1, and 80 mg on days 2 and 3) during the CDDP-containing chemotherapy period, according to the national antiemetic guidelines8.

With CBDCA-containing regimens, palonosetron 0.75 mg infusion on day 1, dexamethasone (6.6 mg intravenous on day 1, and 4 mg oral on days 2 and 3), and oral aprepitant (125 mg on day 1, and 80 mg on days 2 and 3) were administered as prophylaxis. In several patients, additional olanzapine (5 mg/day, between days 1–4) was administered along with the CBDCA-containing chemotherapy based on the severity of CINV in the prior CDDP-containing treatment period. Supportive care for adverse effects was administered at the physician’s discretion.

Evaluation of CINV

All required information was obtained from patients’ daily medical records. All patients were hospitalized during the first cycle of CDDP- and CBDCA-containing treatments. Nausea, vomiting, and anorexia were evaluated daily according to the Common Terminology Criteria for Adverse Events, version 5.0, as applied in clinical practice.

The evaluation periods were during the first five days of the first cycle, and the worst grade of symptoms was recorded.

The primary endpoint of the present study was the assessment of TC rates during all evaluation periods (0–120 h from initiating chemotherapy) in the two groups. Secondary endpoints included the evaluation of TC rates in the acute (0–24 h) and delayed (24–120 h) phases, complete response (CR; no emetic episodes and rescue medication use), and the incidence of nausea, vomiting, and anorexia in the two groups. Furthermore, propensity score matching was performed to adjust for baseline patient characteristics between groups, and the matched data were additionally analyzed to assess the robustness of the results.

Statistical analysis

We hypothesized that the TC rates during the overall evaluation periods in the first cycle were 85% in the control group and 45% in the CINV-experience group, based on previous reports as well as our clinical experience9–11. To achieve 80% power with a 5% alpha error and a patient ratio of 4:3, the required sample size was calculated to be 30 patients in the control group and 22 patients in the CINV-experience group.

Assessment of baseline patient characteristics in the two groups was performed using Fisher’s exact test for categorical variables, and Mann–Whitney U test for continuous variables. The TC rates and incidence of nausea, vomiting, and anorexia were evaluated using the Fisher’s exact test. Propensity score matching was performed using age and sex, both of which are considered obvious CINV risk factors8.

One-to-one nearest neighbor matching without replacement was conducted between the two groups, with a caliper width of 0.20 standard deviations of the logit of propensity scores, to reduce bias from potential confounding factors.

All statistical analyses were performed using the JMP statistical software (version 17.0; SAS Institute, Tokyo, Japan). Statistical significance was set at P < 0.05.

Results

Patient characteristics

In total, 52 patients were enrolled in this study (Fig. 1). Baseline patient characteristics are presented in Table 1.

No statistically significant differences in patient characteristics were observed between the two groups, although more patients in the control group received co-administration of immune-checkpoint inhibitors (ICIs). Females accounted for 26.6% and 36.4% of the patients, and the median age was 66 (range: 48–74) and 64 (42–76) years in the control and CINV-experience groups, respectively. None of the patients had bowel obstruction, vestibulopathy, clinically significant electrolyte abnormalities, uremia, gastric atony, or problematic mental disorders, all of which have been suggested as potential CINV risk factors5.

Two patients in the CINV-experience group received prophylactic olanzapine. Additionally, no significant differences were observed between the two groups in the propensity score-matched population (Supplemental Table 1).

Fig. 1Study design CDDP, cisplatin; CINV, chemotherapy-induced nausea and vomiting.

Table 1Patient characteristics.Control group(n = 30)CINV-experience group(n = 22)P-valueGender (n) Male/female22/814/80.55Age (years) (median, range) Patients < 55 years old (n, %)66 (48–74)5 (16.7)64 (42–76)4 (18.2)0.461.00Performance status (ECOG) (n) 0–1/229/121/11.00BMI (kg/m2) (median, range)22.8 (17.5–30.8)23.5 (16.8–30.6)0.99Recurrence after surgery (n, %)8 (26.7)10 (45.5)0.24Staging (IV or recurrence) (n, %)21 (70.0)14 (63.6)0.77Cancer type (small/non-small) (n)8/22 3/19 0.32Brain metastasis (n, %)6 (20.0)7 (31.8)0.35Liver dysfunction (n, %)2 (6.7)2 (9.1)1.00Serum albumin (g/dL) (median, range)4.1 (2.5–4.8)4.0 (2.8–4.5)0.66CCr (mL/min) (median, range)76.8 (46.3–134.0)76.4 (50.7–115.3)0.63Alcohol intake (≥ 5 days in a week) (n, %)17 (56.7)10 (45.5)0.58Smoking history (Former or Current) (n, %)27 (90.0)19 (86.4)1.00Regular antacid administration (n, %)15 (50.0)13 (59.1)0.58Antihistamines (n, %)6 (20.0)3 (13.6)0.72Narcotics (n, %)5 (16.7)6 (27.3)0.49AUC of carboplatin (mg/mL/min) (n, %) 5 612 (40.0)18 (60.0)10 (45.5)12 (54.5)0.78Concomitant bevacizumab (n, %)3 (10.0)3 (13.6)1.00Concomitant ICIs (n, %)18 (60.0)7 (31.8)0.05Treatment regimens (n, %) PEM-based regimens CBDCA + PEM + bevacizumab CBDCA + PEM + ICIs CBDCA + PEM Non-PEM based regimens CBDCA + nab-PTX CBDCA + ETP + ICIs CBDCA + ETP CBDCA + nab-PTX + ICIs CBDCA + PTX + ICIs CBDCA + PTX + bevacizumab CBDCA + PTX + bevacizumab + atezolizumab2 (6.7)6 (20.0)06 (20.0)5 (16.7)3 (10.0)4 (13.3)3 (10.0)1 (3.3)02 (9.1)3 (13.6)3 (13.6)8 (36.4)2 (9.1)2 (9.1)01 (4.5)01 (4.5)Prior CDDP-based treatment (n, %) PEM-based regimens CDDP + PEM CDDP + PEM + bevacizumab CDDP + PEM + ICIs Non-PEM based regimens CDDP + VNR CDDP + ETP CDDP + IRI CDDP + GEM1 (3.3)2 (6.7)018 (60.0)6 (20.0)3 (10.0)04 (18.2)1 (4.5)2 (9.1)12 (54.5)1 (4.5)1 (4.5)1 (4.5)Interval of CDDP and CBDCA administration (days) (median, range)521 (24–3290)658 (23–2590)0.16ECOG, Eastern Cooperative Oncology Group; BMI, body mass index; CCr, creatinine clearance; AUC, area under the blood concentration-time curve; ICIs, immune checkpoint inhibitors; CBDCA, carboplatin; nab-PTX, nanoparticle albumin-bound paclitaxel; ETP, etoposide; PEM, pemetrexed; PTX, paclitaxel; CDP, cisplatin; VNR, vinorelbine; IRI, irinotecan; GEM, gemcitabine.Liver dysfunction: (grade 1 or higher) aspartate aminotransferase, alanine aminotransferase, and total bilirubin level.ICIs include atezolizumab, pembrolizumab, durvalumab, or nivolumab + ipilimumab.

Evaluation of TC and CR rates, nausea, vomiting, and anorexia

The TC rates as well as CR rates and the incidence of nausea, vomiting, and anorexia are described in Figs. 2, 3 and 4. The TC rates during all evaluation periods were significantly lower in the CINV-experience group than in the control group (45.5% and 86.7%, respectively; P = 0.002; Fig.

2). No patients experienced acute nausea, therefore, the rates in the delayed phase were also 45.5% and 86.7%, respectively. As all patients who used rescue medication exhibited nausea, the CR rates were similar to TC rates (Fig.

3). The incidence rates of all-grade nausea were 54.6% and 13.3% (P = 0.002) and those of anorexia were 50.0% and 16.7% (P = 0.02) in the CINV-experience and control groups, respectively, both of which were significantly higher in the CINV-experience group (Fig. 4).

In contrast, the incidence rates of grade 3 nausea and all-grade vomiting showed no statistical differences between the two groups (9.1% and 0% for both, P = 0.17). Grade 3 anorexia was only observed in the CINV-experience group (18.2% vs. 0%, P = 0.03, Fig.

4).

Fig. 2Comparison of TC rates TC, total control; CINV, chemotherapy-induced nausea and vomiting.

Fig. 3Comparison of CR rates CR, complete response; CINV, chemotherapy-induced nausea and vomiting.

Fig. 4Evaluation of incidence of nausea, vomiting, and anorexia CINV, chemotherapy-induced nausea and vomiting.

Similar results were observed in the propensity score-matched population, although several values for anorexia did not reach statistical significance due to reduced statistical power (Supplemental Fig. 1–3).

Discussion

Management of CINV is crucial, as it is one of the most significant frequently occurring adverse effects of anticancer treatment. Appropriate prophylactic and treatment strategies according to the current guidelines are necessary for successful CINV control. Matsuoka et al. have reported that experiencing CINV during previous chemotherapies was significantly associated with the incidence of CINV during trifluridine/tipiracil + bevacizumab treatment for metastatic colorectal cancer10.

Other reports have also suggested that a history of CINV during prior treatment is a potential risk factor for CINV in subsequent chemotherapy treatment12,13. However, the detailed emetogenic risk of the previous treatments has not been described in these reports, and therefore the association between CINV development and specific emetogenic risk regimens remains unclear. Predicting CINV incidence based on the symptoms induced by a prior treatment can be beneficial in managing the symptoms of the subsequent treatment.

Therefore, we aimed to assess the potential to predict CINV in CBDCA-containing treatment based on the induced symptoms of prior CDDP-containing treatment in real-world thoracic cancer treatment settings.

Notably, patients who experienced CINV with prior CDDP-containing treatment regimens exhibited CBDCA-induced CINV at a significantly higher rate than those who did not experience CINV with the prior CDDP-containing regimens. Since two patients in the CINV-experience group received additional prophylaxis using olanzapine in the subsequent treatment regimen, we reassessed the data following their exclusion, which also resulted in similar outcomes (data not shown), supporting the primary results as well as the results in the propensity score-matched population. To the best of our knowledge, this is the first study demonstrating the potential to predict CINV based on symptoms induced by prior treatments between high and highly moderate emetogenic treatments.

Because patients experiencing CINV during CDDP-containing treatment can exhibit similar symptoms during CBDCA-containing treatment, enhancing CINV prophylaxis in this population would be necessary. CBDCA was categorized as MEC in several antiemetic guidelines, and has been categorized as HEC or high MEC since the mid-2010s2–5. Consequently, the prophylactic antiemetic strategy was modified to a triple combination of NK1 and 5HT3 receptor antagonists along with dexamethasone2–5.

Furthermore, a previous study reported that additional administration of olanzapine improved CINV control in patients receiving CBDCA-containing chemotherapy, although the study did not restrict the types of 5HT3 receptor antagonists used, and omitted dexamethasone on days 2 and 314. In contrast, additional efficacy of aprepitant to palonosetron with dexamethasone was not confirmed in CBDCA + PTX regimen for gynecological cancer or CBDCA + ETP regimen for small-cell lung cancer, which suggests that the selection of 5HT3 receptor antagonists might affect the efficacy of other prophylactic antiemetic agents15,16, which is a controversial topic in the guidelines as well4. Therefore, further evaluation is needed to elucidate the proper use of prophylactic antiemetics (such as the type of NK1 and 5HT3 receptor antagonists, or the dose of olanzapine) in CBDCA-containing treatment regimens, particularly in high-risk populations.

The involvement of medical personnel is also crucial for appropriate CINV management, particularly delayed nausea, which many patients hope to control successfully17,18, to improve patient satisfaction. Since clinicians alone might miss detecting CINV in several cases19, intervention by a medical team is necessary to identify patient complaints and to provide appropriate treatment. Several studies have reported the benefits of involving pharmacists and nurses in CINV management20–22.

Additionally, medical intervention methods include medical interviews, telephone support, and use of electronic devices23–25. Therefore, CINV management should be enhanced using the above-mentioned methods and prophylactic medication, particularly in patients with high CINV risk.

This study had several limitations. First, it was conducted retrospectively, and included a relatively small patient population from a single institution. Additionally, patients in the control group received more ICIs, although it did not show a significant statistical difference.

Second, we assessed CINV daily through active medical interviews by a medical team consisting of physicians, nurses, and pharmacists in clinical practice; however, our evaluations could be objective. Since the medical personnel and patients evaluate CINV based on different perspectives16,26,27, the severity of symptoms in some patients may have been misjudged. Third, we were unable to assess several factors such as nausea gravidarum and motion sickness, due to the unavailability of data, which might have affected the results.

Fourth, the dosage of dexamethasone on day 1 in the CBDCA-containing treatment regimens was higher than that recommended by the guidelines4. In addition, in the CINV-experience group, prophylactic antiemetics were enhanced using additional olanzapine in several patients, depending on the severity of symptoms induced by prior CDDP-containing treatments. These differences might have affected the results.

Additionally, as prophylactic olanzapine has recently been recommended for HEC regimens4, assessment using the prophylaxis is required. Fifth, all CDDP-containing treatments were administered prior to the CBDCA-containing treatments, and approximately one-third of the CDDP-based treatments were initiated in the perioperative setting, suggesting that the condition of the patients varied between the treatment periods. Additionally, we did not set other control arms that were necessary for the specific CINV interaction between CDDP and CBDCA.

Consequently, our outcomes should be interpreted as a potential prediction of CINV development with CBDCA-containing treatments based on symptoms induced by prior CDDP-containing treatments, rather than the direct association between both treatments. Sixth, as we evaluated multiple concomitant chemotherapeutic agents, and PEM or irinotecan can worsen CINV2–5,28, assessment in the integrated regimen will offer further comprehension. Finally, because assessments were confined to first cycles and 0–120 h post-administration without Patient Reported Outcomes (PROs), anticipatory nausea/vomiting was not evaluated; thus, our findings indicate risk stratification from prior symptoms rather than anticipatory mechanisms, requiring further study evaluating anticipatory nausea/vomiting using PROs or with longer evaluation periods.

Therefore, our preliminary results should be validated in future studies.

In conclusion, our study demonstrated that patients experiencing CINV during prior CDDP-containing treatments exhibited significant symptoms with subsequent CBDCA-containing treatments, suggesting that predicting CINV in subsequent treatments based on the symptoms induced by prior treatments is achievable. CINV can be successfully managed by considering the symptoms developed during preceding treatments; therefore, further studies are needed to investigate the associations between similar emetogenic risks and the effectiveness of prophylactic strategies.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Supplementary Material 4

Related articles

PubMed

Incidence of Nausea and Vomiting in Patients With Cervical Cancer Undergoing Chemoradiotherapy With Steroid-Sparing.

Background/Aim Although dexamethasone (DEX) is essential for preventing chemotherapy-induced nausea and vomiting during concurrent chemoradiotherapy (CCRT) for cervical cancer, prolonged administration carries risks of adverse events. Therefore, we evaluated the effect of a shortened DEX duration (steroid-sparing therapy) on the total control (TC) of nausea and vomiting in patients with cervical cancer undergoing CCRT comprising cisplatin 40 mg/m2 to avoid the risk of adverse effects associated...

PubMed

Collagenous Gastritis Presenting as Chronic Nausea and Vomiting.

ABSTRACT Collagenous gastritis (CoG) is a rare disorder characterized by subepithelial collagen band deposition in the gastric mucosa. In this report, we describe a case of a 33-year-old woman presenting with chronic, episodic, postprandial nausea and vomiting later found to have biopsy proven CoG. She symptomatically improved following proton pump inhibitor and budesonide initiation and cessation of nonsteroidal anti-inflammatory drugs. This case underscores the insidious nature of CoG; it sho...