2024 年 47 巻 6 号 p. 1189-1195
Although carboplatin (CBDCA) is classified as a moderately emetogenic agent, the majority of guidelines recommend the use of a neurokinin-1 receptor antagonist in addition to a 5-hydroxytryptamine type 3 receptor antagonist with dexamethasone (DEX) for CBDCA-containing chemotherapy because of its higher emetogenic risk. However, the additional efficacy of aprepitant (APR) in CBDCA-containing treatment remains controversial, and data on multiple-day treatments are limited. Etoposide (ETP) was administered on days 1–3 in the CBDCA + ETP regimen, and it is important to evaluate suitable antiemetic therapy for the regimen. Therefore, we evaluated the efficacy of additional APR in CBDCA + ETP. Patients were divided into two groups and retrospectively evaluated. One was the control group, which was prophylactically administered palonosetron (PALO) and DEX, and the other was the APR group, which received APR orally with PALO and DEX. The primary endpoint was complete response (CR) between the groups. The overall CR rates were 75.0 and 76.4% in the control and APR groups, respectively, with no significant difference (p = 1.00). In the acute phase, it was 88.9 and 97.2%, respectively, and 86.1 and 79.2% in the delayed phase, respectively, without significant differences (p = 0.10 and 0.38, respectively). The incidence and severity of nausea, vomiting, and anorexia were not significantly different between the two groups in the acute and delayed phases. Our findings suggest that combining APR with PALO and DEX does not improve the CR rate in CBDCA + ETP therapy.
It is estimated that 2.09 million cases of lung cancer were diagnosed, and 1.76 million patients died due to lung cancer worldwide in 2018.1) Small cell lung cancer (SCLC) accounts for 10–15% of all lung cancers.2,3) SCLC is highly sensitive to chemotherapy and radiotherapy, especially platinum-based chemotherapy, which can prolong overall survival.4) Platinum + etoposide (ETP) is the most common and effective regimen for SCLC treatment, and immune checkpoint inhibitors (ICIs), such as atezolizumab and durvalumab have been added to this regimen.5,6) Cisplatin or carboplatin (CBDCA) is used as a platinum agent and is selected depending on the patient’s age or performance status.7–10) Additionally, Platinum + ETP is recommended for treatment of neuroendocrine carcinomas.11)
Chemotherapy-induced nausea and vomiting (CINV) is one of the most frequent adverse effects of platinum-based regimens. Uncontrolled CINV limits the dose intensity of chemotherapy and decreases the patient’s QOL.12)
CBDCA has been classified as moderate emetogenic chemotherapy (MEC) in the majority guidelines for antiemetics,13–16) with the recommendation of a 5-hydroxytryptamine type 3 receptor antagonist (5-HT3RA) and dexamethasone (DEX) for prophylaxis. However, the Multinational Association of Supportive Care in Cancer (MASCC)/European Society for Medical Oncology (ESMO), American Society of Clinical Oncology (ASCO), National Comprehensive Cancer Network (NCCN), and the Japanese Society of Clinical Oncology (JSCO) have revised classification of MEC dividing CBDCA and other anticancer agents in antiemetic guidelines in 2016,13–16) and the combination of neurokinin-1 (NK1) receptor antagonist like aprepitant (APR) with 5-HT3RA and DEX became the recommended antiemetic regimen for CBDCA (area under the curve; (AUC) 4 or higher). Based on past clinical trials, nausea occurred in 20–34% and vomiting occurred in 12–17% of patients received the CBDCA + ETP.5,6,17)
Several studies have evaluated the additional antiemetic effects of APR; however, all CBDCA-containing regimens evaluated were single-day administrations.18–24) In contrast, ETP is administered on days 1–3 in CBDCA + ETP. Although ETP is classified as a low emetogenic chemotherapy (LEC),13–16) there is little evidence of the impact of APR addition on CBDCA with multiple-day chemotherapeutic administration.
APR is a well-tolerated medication with mild adverse events such as hiccups, constipation, and hepatic impairment.25) APR is costly, and its effect on CYP induces drug–drug interactions (DDI). Therefore, it is important to confirm its efficacy and safety for appropriate administration.
In this study, we evaluated the prophylactic efficacy of adding APR to palonosetron (PALO) and DEX in CBDCA + ETP.
Patients who first received combination chemotherapy with CBDCA (AUC = 5, on day 1) and ETP (80 or 100 (in the case of ICI co-administration) mg/m2 on days 1–3) with or without ICIs (atezolizumab 1200 mg/body or durvalumab 1500 mg/body) every 3–4 weeks between December 2012 and August 2021 were enrolled in this retrospective observational study. All patients met the following baseline criteria: (1) age ≥20 years; (2) 0 to 2 Eastern Cooperative Oncology Group Performance Status; and (3) sufficient renal and liver function for treatment induction. Patients with baseline nausea or were regularly administered antiemetic drugs, such as olanzapine, metoclopramide, domperidone, prochlorperazine, lorazepam, hydroxyzine, or glucocorticoids, and those with insufficient data were excluded. Patients were divided into two groups: a control group that did not receive APR from December 2012 to June 2017, and an APR group that received APR from July 2017 to August 2021. Patients’ backgrounds were balanced in a propensity score-matched population (Table 1B). This study was approved by the Ethical Review Board for Life Science and Medical Research of Hokkaido University Hospital (Approval Number: 021-0169) and conducted in accordance with the Declaration of Helsinki and the Strengthening the Reporting of Observational Studies in Epidemiology statement. Given the retrospective nature of the study, informed consent was not mandated.
| (A) Patient’s characteristics (all-patient population) | |||
|---|---|---|---|
| Control group (n = 72) | APR group (n = 72) | p-Value | |
| Age (median, range) | 70 (54–84) | 70 (37–83) | 1.00 |
| Sex (male/female) | 57/15 | 48/24 | 0.13 |
| Body surface area (m2) (median, range) | 1.65 (1.30–2.18) | 1.64 (1.27–2.16) | 0.80 |
| Tumor types (n, %) | |||
| Small cell lung cancer | 56 (77.8) | 60 (83.3) | |
| Neuroendocrine carcinoma | 16 (22.2) | 12 (16.7) | 0.53 |
| Staging (n, %) | |||
| Stages I–III | 25 (34.7) | 30 (41.7) | |
| Stage IV, recurrence | 47 (65.2) | 42 (58.3) | 0.49 |
| Carboplatin dosage (AUC) (n, %) | |||
| 5 | 71 (98.6) | 69 (95.8) | |
| 4 (dose reduction) | 1 (1.4) | 3 (4.2) | 0.31 |
| Etoposide dosage (mg) (median, range) | 131 (78–174) | 131 (69–188) | 0.51 |
| Co-administration of ICIs | 0 | 24 | <0.01 |
| Drinking history (n, %) | |||
| None | 24 (33.3) | 22 (30.6) | |
| Occasional drinking | 17 (23.6) | 17 (23.6) | |
| Yes | 31 (43.1) | 33 (45.8) | 0.70 |
| Smoking history (n, %) | 64 (88.9) | 70 (97.2) | 0.10 |
| Regular opioid use (n, %) | 6 (8.3) | 8 (11.1) | 0.78 |
| Brain metastasis (n, %) | 9 (12.5) | 9 (12.5) | 1.00 |
| Creatinine clearance (mL/min) (median, range) | 74.3 (27.0–129.5) | 72.0 (29.9–171.9) | 0.45 |
| Total bilirubin (mg/dL) (median, range) | 0.6 (0.3–2.4) | 0.6 (0.3–2.8) | 0.61 |
| AST or ALT elevation (n, %) | |||
| Grades 1–2 | 10 (13.9) | 2 (2.8) | |
| Grades 3–4 | 1 (1.4) | 3 (4.2) | 0.14 |
| (B) Patient’s characteristics (propensity score-matched population) | |||
| Control group (n = 59) | APR group (n = 59) | p-Value | |
| Age (median, range) | 70 (54–82) | 70 (37–83) | 0.74 |
| Sex (male/female) | 44/15 | 44/14 | 0.67 |
| Body surface area (m2) (median, range) | 1.64 (1.30–2.18) | 1.67 (1.27–2.16) | 0.20 |
| Tumor types (n, %) | |||
| Small cell lung cancer | 47 (79.7) | 49 (83.1) | |
| Neuroendocrine carcinoma | 12 (20.3) | 10 (16.9) | 0.81 |
| Staging (n, %) | |||
| Stages I–III | 19 (32.2) | 25 (42.4) | |
| Stage IV, recurrence | 40 (67.8) | 34 (57.6) | 0.34 |
| Carboplatin dosage (AUC) (n, %) | |||
| 5 | 58 (98.3) | 57 (96.6) | |
| 4 (dose reduction) | 1 (1.7) | 2 (3.4) | 1.00 |
| Etoposide dosage (mg) (median, range) | 129 (78–174) | 134 (69–188) | 0.05 |
| Co-administration of ICIs | 0 | 20 | <0.01 |
| Drinking history (n, %) | |||
| None | 20 (33.9) | 17 (28.8) | |
| Occasional drinking | 14 (23.7) | 16 (27.1) | |
| Yes | 25 (42.4) | 26 (44.1) | 0.68 |
| Smoking history (n, %) | 53 (90.0) | 58 (98.3) | 0.11 |
| Regular opioid use (n, %) | 3 (5.1) | 4 (6.8) | 1.00 |
| Brain metastasis (n, %) | 7 (11.9) | 7 (11.9) | 1.00 |
| Creatinine clearance (mL/min) (median, range) | 73.6 (27.0–129.5) | 72.8 (29.9–171.9) | 0.57 |
| Total bilirubin (mg/dL) (median, range) | 0.6 (0.3–2.4) | 0.6 (0.3–2.8) | 0.78 |
| AST or ALT elevation (n, %) | |||
| Grades 1–2 | 10 (16.9) | 1 (1.7) | |
| Grades 3–4 | 1 (1.7) | 3 (5.1) | 0.07 |
AUC, area under the curve; ICIs, immune checkpoint inhibitors; AST, aspartate aminotransferase; ALT, alanine aminotransferase. AST or ALT elevation was evaluated using the Common Terminology Criteria for Adverse Events version 5.0.
Patients in the control group were intravenously administered PALO 0.75 mg on day 1 and DEX 6.6 mg on days 1–3. Those in the APR group were treated with APR 125 mg orally with PALO 0.75 mg and DEX 6.6 mg intravenously on day 1, followed by oral APR 80 mg and intravenous DEX 3.3 mg on days 2 and 3. Antiemetic drugs, such as metoclopramide 5–10 mg, domperidone 10 mg, prochlorperazine 5 mg, olanzapine 5 mg, hydroxyzine 25 mg, or lorazepam 0.5 mg were administered as rescue medications according to the physician’s decision.
Evaluation CriteriaAll patients were hospitalized and the required information was obtained from their daily medical records. The evaluation periods were: days 1–5 (overall phase), day 1 (acute phase), and days 2–5 (delayed phase) in the first cycle of CBDCA + ETP, according to previous studies.18–24) The primary endpoint was an evaluation of complete response (CR), which was defined as the absence of emetic events, vomiting, and the need for rescue antiemetic treatment in the overall phase of the first cycle between the groups. The secondary endpoints were the CR rate in the acute and delayed phases and the incidence and severity of nausea, vomiting, and anorexia in each phase. The symptoms were evaluated based on real-time daily assessments by physicians and/or pharmacists according to the Common Terminology Criteria for Adverse Events version 5.0, and the worst grade was described.
Statistical AnalysisDifferences in patient characteristics between the control and APR groups were assessed using Fisher’s exact test for categorical variables and the Mann–Whitney U test for continuous variables. Differences in CR rate and incidence of gastrointestinal symptoms between the two groups were analyzed using Fisher’s exact test. Differences in the severity of gastrointestinal symptoms were assessed using the Mann–Whitney U test. Using propensity score matching, we compared the two groups adjusted by risk factors for CINV, including age, sex, and alcohol consumption, to compensate for insufficient statistical power. A 1 : 1 matching (without replacement) between the two groups was achieved using the nearest neighbor method with a 0.20-width caliper of the standard deviation of the logit of propensity scores. Additional evaluation was performed in a propensity score-matched population to confirm the results in an all-patient population. All analyses were performed using JMP version 13.2.1 statistical software (SAS Institute, Tokyo, Japan). A p-value <0.05 was considered statistically significant for all tests.
A total of 178 patients were enrolled in this study and 34 patients were excluded. Finally, 72 patients were evaluated in each group (Fig. 1). Patient backgrounds are shown in Table 1. There were no significant differences between the control and APR groups in terms of age, sex, body surface area, tumor type, staging, dosage of CBDCA (AUC) and ETP (mg), drinking habits, smoking history, regular opioid use, or presence of brain metastasis. The number of patients with renal dysfunction (creatinine clearance calculated using the Cockcroft–Gault formula of less than 60 mL/min) and those with liver dysfunction (grade 1 or higher aspartate transaminase, alanine aminotransferase, and total bilirubin levels) did not differ between the groups.
However, the inclusion of patients who were co-administered ICIs was significantly more in the APR group than in the control group, although this was unlikely to affect gastrointestinal symptoms as ICIs are classified as having minimal emetic risk in each guideline for antiemetics.13–16) In addition, no problematic immune-related adverse events were observed during the evaluation period (data not shown).
Comparison of CR RateThe overall CR rates were 75.0 and 76.4% in the control and APR groups, respectively, with no significant differences (p = 1.00, Fig. 2). The percentages in the acute and delayed phases were 88.9 and 97.2%, and 86.1 and 79.2%, respectively, with no significant differences (p = 0.10 and 0.38, respectively). Similar results were obtained between the propensity score-matched populations. As the APR group had 80 or 100 mg/m2 on doses of ETP due to ICIs combination, we also examined this influence, and no differences was observed (data not shown).
The incidence and severity of gastrointestinal symptoms are shown in Table 2. The incidence of all-grade nausea in the control and APR groups was 16.7% in the overall phase, 4.2 and 0% in the acute phase, and 11.1 and 16.7% in the delayed phase, respectively, with no significant differences (p = 1.00, 0.25, and 0.47, respectively). The incidence of vomiting was 2.8 and 0% in the overall phase, 1.4 and 0% in the acute phase, and 1.4 and 0% in the delayed phase, respectively, with no significant differences (p = 0.50, 1.00, and 1.00, respectively). The incidence of anorexia was 12.5 and 8.3% in the overall phase, 0 and 1.4% in the acute phase, and 12.5 and 6.9% in the delayed phase, respectively, with no significant differences (p = 0.59, 1.00, and 0.40, respectively). Additionally, the severity of nausea, vomiting, and anorexia did not differ between groups in each phase (Table 2). Similar results were obtained in the propensity-matched population.
| All-patient population | Propensity score-matched population | |||||
|---|---|---|---|---|---|---|
| Control group (n = 72) | APR group (n = 72) | p-Value | Control group (n = 59) | APR group (n = 59) | p-Value | |
| Incidence of gastrointestinal symptoms (n, %) | ||||||
| Overall phase (days 1–5) | ||||||
| Nausea all-grade | 14 (16.7) | 12 (16.7) | 1.00 | 12 (20.3) | 12 (20.3) | 1.00 |
| G1 | 11 (15.2) | 9 (12.5) | 10 (16.9) | 9 (15.3) | ||
| G2 | 0 | 3 (4.2) | 0 | 3 (5.1) | ||
| G3 | 1 (1.4) | 0 | 0.95 | 1 (1.7) | 0 | 0.77 |
| Vomiting all-grade | 2 (2.8) | 0 | 0.50 | 2 (3.4) | 0 | 0.50 |
| G1 | 2 (2.8) | 0 | 2 (3.4) | 0 | ||
| G2 | 0 | 0 | 0 | 0 | ||
| G3 | 0 | 0 | 0.16 | 0 | 0 | 0.16 |
| Anorexia all-grade | 9 (12.5) | 6 (8.3) | 0.59 | 8 (13.6) | 4 (6.8) | 0.36 |
| G1 | 4 (5.6) | 3 (4.2) | 4 (6.8) | 2 (3.4) | ||
| G2 | 3 (4.2) | 3 (4.2) | 3 (5.1) | 2 (3.4) | ||
| G3 | 2 (2.8) | 0 | 0.4 | 1 (1.7) | 0 | 0.22 |
| Acute phase (day 1) | ||||||
| Nausea all-grade | 3 (4.2) | 0 | 0.25 | 3 (5.1) | 0 | 0.24 |
| G1 | 3 (4.2) | 0 | 3 (5.1) | 0 | ||
| G2 | 0 | 0 | 0 | 0 | ||
| G3 | 0 | 0 | 0.09 | 0 | 0 | 0.08 |
| Vomiting all-grade | 1 (1.4) | 0 | 1.00 | 1 (1.7) | 0 | 1.00 |
| G1 | 1 (1.4) | 0 | 1 (1.7) | 0 | ||
| G2 | 0 | 0 | 0 | 0 | ||
| G3 | 0 | 0 | 0.32 | 0 | 0 | 0.33 |
| Anorexia all-grade | 0 | 1 (1.4) | 1.00 | 0 | 0 | 1.00 |
| G1 | 0 | 1 (1.4) | 0 | 0 | ||
| G2 | 0 | 0 | 0 | 0 | ||
| G3 | 0 | 0 | 0.32 | 0 | 0 | 0.33 |
| Delayed phase (days 2–5) | ||||||
| Nausea all-grade | 8 (11.1) | 12 (16.7) | 0.47 | 8 (13.6) | 12 (20.3) | 0.46 |
| G1 | 7 (9.7) | 9 (12.5) | 7 (11.9) | 9 (12.5) | ||
| G2 | 0 | 3 (4.2) | 0 | 3 (5.1) | ||
| G3 | 1 (1.4) | 0 | 0.32 | 1 (1.7) | 0 | 0.31 |
| Vomiting all-grade | 1 (1.4) | 0 | 1.00 | 1 (1.7) | 0 | 1.00 |
| G1 | 1 (1.4) | 0 | 1 (1.7) | 0 | ||
| G2 | 0 | 0 | 0 | 0 | ||
| G3 | 0 | 0 | 0.32 | 0 | 0 | 0.33 |
| Anorexia all-grade | 9 (12.5) | 5 (6.9) | 0.40 | 8 (13.6) | 4 (6.8) | 0.36 |
| G1 | 4 (5.6) | 2 (2.8) | 4 (6.8) | 2 (3.4) | ||
| G2 | 3 (4.2) | 3 (4.2) | 3 (5.1) | 2 (3.4) | ||
| G3 | 2 (2.8) | 0 | 0.40 | 1 (1.7) | 0 | 0.22 |
CBDCA + ETP is a chemotherapy used for the treatment of SCLC4–6) and neuroendocrine carcinoma.11) CINV is a management-requiring adverse event, and its control is important for maintaining patients’ QOL and dose intensity of anticancer agents.12) CBDCA-containing regimens are classified as MEC, although they are similar to high emetogenic chemotherapy. Thus, revisions in the recommendations were made to add APR to 5-HT3RAs and DEX in the guidelines.13–16) All concomitant drugs administered with CBDCA in previous studies that evaluated the efficacy of additional APR were administered on a single day alone.18–24) However, ETP was administered for three days in combination with CBDCA. Although ETP is classified as a LEC, it is important to assess the usability of additional APR for CBDCA-containing multiple-day administration regimens. Therefore, we aimed to evaluate the prophylactic efficacy of APR for CBDCA + ETP.
The results showed no significant difference in the CR rates between the control and APR groups in the overall, acute, and delayed phases. In addition, the incidence and severity of nausea, vomiting, and anorexia did not differ between the groups in any of the evaluation phases. These results suggest that APR did not improve CINV in CBDCA + ETP-containing regimens, similar to our previous evaluation of a single-day CBDCA + paclitaxel regimen.23)
Since ETP is categorized as LEC and the prophylactic antiemetic treatment recommended for LEC is a single agent prophylaxis such as DEX or 5-HT3RAs,13–16) we consider that the CINV caused by ETP was adequately controlled with prophylactic antiemetics for MEC.
Moreover, the results of the present study may be associated with superior antiemetic effects of PALO on first-generation 5-HT3RAs.26) PALO has approximately 30 times higher affinity for 5-HT3 receptors and an 8–10 times longer half-life (40 h) than first-generation 5-HT3RAs, such as granisetron (GRA) and ondansetron.27) In addition, PALO inhibits crosstalk between NK1 and 5-HT3 receptors without binding to the NK1 receptor because of the inhibition of substance P-induced calcium ion release.28,29) Studies on triple-drug prophylaxis for MEC regimens have shown conflicting results18–24) which may be due to the different properties of 5-HT3RAs. All studies demonstrating the positive efficacy of APR used first-generation 5-HT3RAs, such as ondansetron or GRA.18–22) In contrast, the addition of APR was not effective in the studies on PALO.23,24) Previous studies have suggested that PALO is as effective as APR + first-generation 5-HT3RAs in CBDCA regimens.23,24) The present study indicated similar outcomes, even though other chemotherapeutic agents were administered for multiple days.
NK1 receptor antagonists, including APR, can cause DDI. APR is an inhibitor of CYP3A4 and an inducer of CYP2C9.30) In addition, netupitant inhibits CYP3A4 and rolapitant inhibits CYP2D6, P-glycoprotein, and breast cancer resistance proteins.31,32) Therefore, in cases where medications are affected by NK1 receptor antagonists, dose adjustment or a change to alternative drugs may be necessary. Patients with cancer often have polypharmacy, with an average of 5–9 medications prescribed. DDI is a troublesome problem, especially among elderly patients with organ dysfunction.33–35) Consequently, we believe that it is better to avoid using additional APR in view of DDI.
Additionally, PALO and APR are expensive drugs; for example, APR increases costs by 50% when co-administered with PALO + DEX. Furthermore, when comparing PALO and GRA in combination with APR + DEX, PALO costs 80% more than GRA. Owing to the increasing medical costs in Japan, it is better to select lower-cost drugs if their efficacies are equivalent.
The current study has some limitations. First, it was a retrospective study with a small sample size from a single institution. In this study, we hypothesized that the CR rate in the first cycle would be 60% in the control and 80% in the APR groups based on previous reports.18–21) The calculated required sample size was 91 participants per group to achieve 80% power with an alpha error of 5%. However, the sample size was 72 participants per group, the maximum collected during possible evaluation periods. This suggests that this study may have less power to evaluate, requiring a cautious interpretation of the results. Therefore, it is necessary to conduct a large-scale, multicenter prospective study to confirm these findings. Second, several background factors that should have been evaluated, such as motion and morning sickness, were not assessed. Third, in Japan, 0.75 mg of PALO is adopted although 0.25 mg is the international standard dosage, suggesting that these results should be applied cautiously. Fourth, we evaluated the efficacy of APR in the first course, because several factors may affect CINV in subsequent courses. Evaluation of multiple courses of chemotherapy may reveal other outcomes, warranting further research. Fifth, the DEX dosage on day one differed from the current guidelines for patients with APR administration.16) Sixth, patients with ICI coadministration were significantly more included in the APR group than in the control. Although the impact might be low, it could have affected the results. Finally, we adopted a healthcare worker-based evaluation of CINV in this retrospective study; however, a subjective assessment by patients would provide a better evaluation of antiemetic therapy.
Our findings suggest that combining APR with PALO and DEX does not improve the CR rate in CBDCA + ETP therapy.
The authors declare no conflict of interest.
