Delayed nausea and vomiting from carboplatin doublet chemotherapy

Abstract

Background: Delayed nausea and vomiting following administration of carboplatin containing chemotherapy regimen remains a clinically significant problem for patients with cancer despite administration of standard antiemetic prophylaxis comprising of a 5-HT3 antagonist and dexamethasone. We performed a prospective study to define the incidence and risk factors for delayed chemotherapy-induced nausea and vomiting (CINV).

Methods: Previously untreated patients with newly diagnosed cancer scheduled to receive carboplatin containing chemotherapy (AUC 5 or above), but no prophylactic aprepitant were enrolled in the study. The primary endpoint was the incidence of delayed CINV after Cycle 1 of chemotherapy. Secondary endpoints included the incidence of CINV with the third chemotherapy cycle and gender differences in incidence of CINV. Patients completed the Functional Living Index Emesis (FLIE) questionnaires 24, 48, 72 and 96 hours after receiving chemotherapy. Telephone interviews were conducted 24–48 hours following chemotherapy to assess the severity and need for breakthrough medications for CINV.

Results: Between December 2006 and July 2009, 105 patients were enrolled onto this study. Delayed emesis following Cycle 1 of carboplatin was observed in 30% of patients. Of these, 14.1%, 22.4% and 23.5% of patients described CINV at 48, 72, and 96 hours, respectively. The incidence of delayed CINV following Cycle 3 dropped to 12.8%, 14.6% and 16% of patients at 48, 72 and 96 hours, respectively. No differences were observed in the incidence of CINV between men and women. A total of 20% of patients required use of breakthrough antiemetics with Cycle 1.

Conclusions: Without prophylactic aprepitant administration, 30% of patients receiving carboplatin containing regimen had moderate to severe delayed CINV.

Nausea and vomiting are among the most distressing side effects of chemotherapeutic agents. Inadequately controlled chemotherapy-induced nausea and vomiting (CINV) can lead to dehydration, electrolyte imbalances and malnutrition, which can jeopardize patient willingness to continue chemotherapy [1,2]. The incidence of CINV varies by chemotherapeutic agent with drugs, such as cisplatin, classified as highly emetogenic with CINV rates greater than 90%. CINV with moderately emetogenic agents can vary from 30% to 90%. Carboplatin falls in the higher range of moderately emetogenic chemotherapy. Other reported risk factors for CINV include younger age, female gender, emesis gravidarum, prior history of CINV, motion sickness, and sporadic alcohol use [2,3].

CINV is described as acute if it observed within the first 24 hours following chemotherapy administration, and delayed if seen after 24 hours. Delayed CINV peaks 48–72 hours following chemotherapy and subsides over the next 2–3 days [4]. As a result of effective prophylactic antiemetic combinations of 5-hydroxytryptamine type 3 (5-HT3) receptor antagonists, neurokinin-1 (NK-1) receptor antagonists and corticosteroids, CINV rates with highly emetogenic chemotherapy has dropped to 30% [5,6]. Current guidelines support the use of 5-HT3 receptor antagonists and corticosteroids for moderately emetogenic agents, plus or minus NK-1 receptor antagonists, as data supporting the use of all three antiemetics together is scant.

Carboplatin is a platinum salt which cross-links DNA, and is routinely used for the treatment of solid tumors, as an alternative to cisplatin due to favorable toxicity profile. Data regarding the incidence and severity of CINV with carboplatin are limited, as is its impact on quality of life [7,8]. Therefore, the optimal prophylactic antiemetic combination for use with carboplatin remains undefined.

We therefore performed a prospective observational study to determine the incidence, severity and risk factors for delayed CINV following the administration of carboplatin doublet chemotherapy.

Patients and methods

Patient selection and eligibility

This study was approved by Washington University School of Medicine Institutional Review Board and all study participants provided written informed consent for participation. This study was a prospective, observational trial conducted between December 2006 and July 2009. The study population included patients over the age of 18 with newly diagnosed, histologically or cytologically proven cancer, who were scheduled to receive chemotherapy with carboplatin at AUC 5 or above. Eligible patients received standard antiemetic prophylaxis prior to carboplatin, defined as 5-HT3 antagonist and dexamethasone. At the time of study enrollment the standard antiemetics at our institution involved ondansetron 32 mg IV or palonosetron 0.25 mg IV in combination with dexamethasone 10 or 20 mg IV on Day 1 of chemotherapy. Patients were ineligible if they received aprepitant as part of their prophylactic antiemetic regimen. Patients who had received prior cytotoxic chemotherapy within the last five years were also excluded, as were patients who were pregnant.

Assessment of delayed nausea and vomiting

Nausea and vomiting were defined as “acute” if occurred within the first 24 hours after carboplatin administration, and “delayed” if observed after 24 hours and assessed up to 96 hours after chemotherapy. The Functional Living Index Emesis (FLIE) standardized questionnaire was given to each patient during their scheduled clinic visit prior to their first and third cycle of carboplatin. This self-administered questionnaire captured the incidence and severity of nausea and vomiting in the preceding 24 hours.

Patients completed the questionnaire 24 hours after carboplatin administration to determine symptoms of acute CINV, and at the following time points to record incidence and severity of delayed nausea and vomiting: 48, 72 and 96 hours. A trained clinical research associate (CRA) explained the instructions to complete the questionnaire to the patient.

The FLIE questionnaire has 18 questions divided into two categories: Nausea (questions 1–9) and Vomiting (questions 10–18). Each of these questions was answered using a 100-mm (1–7 points) visual analog scale (VAS) with anchors to “not at all” and “a great deal” and tick marks dividing the scale into six equal zones. Questions within the category were weighted equally and totaled to create the category score. The two category scores were then combined to create a total score. Higher scores indicated less impairment on daily life as a result of nausea or vomiting.

Based on the total FLIE scores, each patient was assigned to one of the following groups: 1) severe delayed nausea and vomiting (score ≤ 45); 2) moderate delayed nausea and vomiting (score 46–99); and 3) no impact of nausea and vomiting on daily life (score ≥ 100). A FLIE score less than 100 (severe and moderate delayed nausea and vomiting) constituted an emetic episode and was used to calculate the incidence of delayed nausea.

Patients were sent home with prescriptions for breakthrough antiemetic medications as per physician preference. Patients were also interviewed by a trained clinical research assistant or research nurse over the telephone 24–48 hours following carboplatin administration to assess the severity of delayed nausea and vomiting and to assess the need for any breakthrough antiemetic medications. Breakthrough antiemetics used as well as frequency of use were recorded.

Statistical considerations

The goal of this study was to prospectively characterize the incidence and severity of delayed emesis following administration of carboplatin containing chemotherapy regimens despite the appropriate administration of standard 5-HT3 receptor antagonist antiemetic prophylaxis after the first and third cycle of chemotherapy. The primary endpoint was to examine the incidence of delayed nausea and vomiting following Cycle 1 of carboplatin administration. Secondary endpoints included examining the incidence of delayed nausea and vomiting following Cycle 3 of carboplatin, differences in incidence of emesis by gender, and patterns of use of breakthrough anti-emetic medication use. A relatively large sample size (N = 100) was proposed to allow the estimation of the overall incidence of delayed emesis with a less than 5% measurement error. This allowed sufficient numbers for the possible estimations within subgroups, but the study was not powered for subgroup analysis.

The analysis for this study was primarily descriptive. Demographic and clinical characteristics of the sample, occurrence and severity of the delayed emesis, as well as numbers of emetic episodes and time to the onset of emesis, were summarized using descriptive statistics. The incidence of the emesis (overall and by severity) was calculated for both Cycle 1 and 3 at the following time points: 24, 48, 72 and 96 hours. The overall incidence of delayed nausea and vomiting was described by single patient reports during the study time periods for Cycles 1 and 3. χ²-test was used to tabulate and compare the difference incidence of delayed nausea and vomiting between male and female participants. A multivariate logistic regression analysis was also fitted to explore the possible associations of other demographic/clinical characteristics to the overall incidence of delayed emesis.

Results

Patient characteristics

Between December 2006 and July 2009, 105 patients were enrolled onto this study. Of these participants, 99 patients completed Cycle 1 FLIE questionnaires, while only 50 completed Cycle 3 FLIE questionnaires. The patient characteristics are listed in Table 1. Of the patients enrolled, 62 were male and 43 were female with a median age of 64.8 years. The majority of patients were Caucasian (83/105, 79%), followed by African American (19/105, 18%), Native American or Alaskan Nation (2/105, 2%), and Asian (1/105, 1%). Seventy-nine percent of the patients were diagnosed with non-small cell lung cancer. The remaining 21% of the patients had the following tumors: small cell lung cancer 6.7%, esophageal 6.7%, melanoma 1.9%, mesothelioma 4.8%, and transitional cell 1.0%. The majority of the patients had stage III and IV disease (21.6% and 72.7%, respectively). In terms of smoking status, 33% of the patients were current smokers while 61% were former smokers and 6% had never smoked.

Table 1. Demographics of patients enrolled on study.

Variable	Total (%) N = 105
Age (median)	64.8 years
Gender
Male	62 (59%)
Female	43 (41%)
Race
White	83 (79%)
Black or African-American	19 (18%)
Native American or Alaskan Nation	2 (2%)
Asian	1 (1%)
Tumor type
Non-small cell lung	83 (79.0%)
Esophageal	7 (6.7%)
Small cell lung	7 (6.7%)
Mesothelioma	5 (4.8%)
Transitional cell	1 (1.0%)
Stage
I	2 (2.3%)
II	3 (3.4%)
III	19 (21.6%)
IV	64 (72.7%)
Smoking status
Current	32 (33.0%)
Former	59 (60.8%)
Never	6 (6.2%)

Incidence of nausea and vomiting following Cycle 1 of carboplatin

The mean FLIE score for patients following Cycle 1 were 118.8 (standard deviation/SD12.2), 116.6 (SD 14.9), 112.8 (SD 19.3) and 111.3 (SD 23.1) at 24, 48, 72 and 96 hours, respectively. A total of 30 patients reported delayed CINV at any time point following chemotherapy up to 96 hours during Cycle 1. Ten percent of patients described acute nausea and vomiting within 24 hours following carboplatin administration, which was categorized as moderate (FLIE score 46–99). None of the patients had severe (FLIE score ≥ 100) acute CINV. Delayed emesis was observed in 14.1% (moderate), 22.4% (21.4% moderate and 1% severe) and 23.5% (18.4% moderate and 4.1% severe) of patients at 48, 72 and 96 hours, respectively.

Incidence of nausea and vomiting following Cycle 3 of carboplatin

The mean FLIE score for patients following Cycle 3 were 121 (SD 9.1), 115 (SD 22.7), 115.8 (SD 19.7) and 115.5 (SD 20.2) at 24, 48, 72 and 96 hours, respectively. A total of 13 patients of 50 patients described delayed CINV at any time point following chemotherapy and up to 96 hours during Cycle 1. Of these, 4.1% described acute emesis within 24 hours following Cycle 3 of carboplatin, which was categorized as moderate. Delayed emesis was observed in 12.8% (8.5% moderate and 4.3% severe), 12.4% (10.4% moderate and 2% severe) and 16% (14% moderate and 2% severe) of patients at 48, 72 and 96 hours, respectively.

Differences in delayed emesis by gender and other risk factors

Tables 2 and 3 show the incidence of acute and delayed nausea and vomiting for both men and women, at the time points 24, 48, 72 and 96 hours, respectively, following Cycles 1 and 3 of carboplatin. Based on univariate analysis, no differences were observed in the incidence of carboplatin containing chemotherapy-induced emesis between men and women (Table 4). Age was not a predictor for moderate-severe emesis in our dataset. Smoking status appeared to be a predictor on univariate analysis with the incidence of chemotherapy-induced delayed moderate to severe emesis being higher in never or former smokers compared to current smokers. However, this trend was no longer observed on multivariate analysis. Therefore, no significant clinical predictors for moderate-severe emesis were observed in this dataset.

Table 2. Incidence of acute and delayed nausea and vomiting in patients following Cycle 1 of carboplatin containing chemotherapy according to gender.

Type of emesis	Time points	Overall (%)	Male (%)	Female (%)
Acute	24 hours	10 (10%)	7 (11.9%)	3 (7.3%)
Delayed	48 hours	14 (14.1%)	10 (16.9%)	4 (10%)
	72 hours	22 (22.4%)	14 (24.1%)	8 (20%)
	96 hours	23 (23.5%)	16 (27.1%)	7 (17.9%)

Table 3. Incidence of acute and delayed nausea and vomiting in patients following Cycle 3 of carboplatin containing chemotherapy according to gender.

Type of emesis	Time points	Overall (%)	Male (%)	Female (%)
Acute	24 hours	2 (4.1%)	1 (3.6%)	1 (4.8%)
Delayed	48 hours	6 (12.8%)	4 (14.3%)	2 (10.5%)
	72 hours	7 (14.6%)	4 (14.3%)	3 (15%)
	96 hours	8 (16%)	5 (17.2%)	3 (14.3%)

Table 4. Risk factors for delayed nausea and vomiting following Cycle 1 of carboplatin containing chemotherapy based on univariate analysis.

			Incidence
Covariate	Statistics	Level	No emesis	Moderate-severe emesis	Parametric p-value*
Gender	N (%)	Male	39 (56.52)	20 (66.67)	0.344
	N (%)	Female	30 (43.48)	10 (33.33)
Smoking status	N (%)	Never/Former	38 (61.29)	24 (82.76)	0.041
	N (%)	Current	24 (38.71)	5 (17.24)
Age	N		69	30	0.704
	Mean		64.85	64
	Std dev		10.38	9.88

*The parametric p-value is calculated by ANOVA for numerical covariates and χ²-test for categorical covariates.

Use of breakthrough medication for nausea and vomiting

Table 5 describes the frequency of breakthrough medication for nausea and vomiting in patients following Cycle 1 of carboplatin containing chemotherapy. A total of 20/99 (20.2%) patients required use of breakthrough antiemetics. Prochlorperazine was the most commonly used antiemetic (15/20) either alone, or combination with other agents, such as lorazepam or metoclopramide.

Table 5. Breakthrough anti-emetic use among patients following Cycle 1 of carboplatin containing chemotherapy.

Breakthrough anti-emetic	Frequency	Cumulative frequency
Lorazepam	2	2
Lorazepam and prochlorperazine	1	3
Prochlorperazine	13	16
Prochlorperazine and metoclopramide	1	17
Metoclopramide	2	19
Dexamethasone	1	20

Discussion

This prospective study, evaluating the incidence and severity of delayed nausea and vomiting with carboplatin containing regimens, was performed prior to aprepitant being commonly used. Unlike most studies which evaluate breakthrough emesis as a dichotomous variable, we utilized the validated FLIE questionnaire to capture not only the severity of breakthrough emesis, but also its impact on the quality of life [9–11]. This emphasis on patient-reported outcomes is a major strength of this study. In addition, the FLIE questionnaire was administered at several time points, and captures date pertaining to the preceding 24 hours, which reduces recall bias. The prophylactic antiemetics administered to patients in this study were uniform, based on our institutional practice.

Investigating the incidence and severity of emesis with Cycles 1 and 3 provided a snapshot of the effectiveness of pharmacologic interventions to manage breakthrough emesis. The reduction in incidence of both acute and delayed emesis with Cycle 3 of chemotherapy, compared to Cycle 1, reflects the successful use of rescue antiemetics, given that dose reductions in carboplatin were rare. The patient dropout rate of 50% is similar to rates observed in other quality of life studies [12,13]. It is sobering that even with breakthrough anti-emetic use, delayed emesis was observed in over 12% of patients even with Cycle 3, underscoring the importance of better prophylactic antiemetic regimens.

Following our initial report, a phase II study randomly assigned 134 chemotherapy-naïve Japanese patients with advanced non-small cell lung cancer receiving carboplatin-based chemotherapy to receive standard prophylactic antiemetics (5-HT3 receptor antagonist and corticosteroid) with or without aprepitant [14,15]. The primary endpoint of this study was to evaluate the complete response rate, defined as no vomiting and no rescue therapy during the 120 hours post-chemotherapy. The study showed trend towards improved emesis control in the aprepitant group compared to the non-aprepitant group (complete response rate 80.3% vs. 67.2%, odds ratio 0.50; 95% CI 0.22–1.10; p = 0.085) [15]. Though this benefit of aprepitant did not meet statistical significance, improvement in emesis control remains an issue of importance.

Several patient-related risk factors for CINV have been described, including younger age, female gender, sporadic alcohol use and prior chemotherapy [16]. The association with smoking observed in our study is interesting, and needs to be explored further. While our study did not specifically aim to evaluate individual risk factors, but rather focus on quality of life measures, we did examine the role of gender, but did not find a significant relationship in our study. Recent randomized studies of aprepitant versus placebo in female Japanese patients receiving carboplatin have shown mixed results regarding efficacy. Tanioka and colleagues demonstrated a numerically higher complete response (no emesis or rescue therapy needed) in the aprepitant group, though this did not meet statistical significance [17]. Yahata and colleagues, however, did observe a statistically significant improvement in “no vomiting” in the aprepitant group compared to placebo [18].

Further studies are needed to validate the utility of aprepitant with carboplatin-based regimens in other ethnic patient subgroups. Despite the absence of compelling data favoring prophylactic aprepitant use to control emesis with carboplatin, current guidelines list its use as “an option” to help control this feared side effect of chemotherapy [1,19].

Funding information

This publication was supported by the National Cancer Institute of the National Institutes of Health (NIH), Grant Number 1K12CA167540 and the Clinical Translational Science Award (CTSA) program of the National Center for Advancing Translational Sciences at the National Institutes of Health, Grant Number UL1RR024992. This study was supported in part by Cancer Center Support Grant P30 CA91842 and the Clinical Trials Core of the Siteman Cancer Center at Washington University and Barnes-Jewish Hospital in St. Louis, Missouri.

Disclosure statement

The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.