FormalPara Key Summary Points

Cisplatin-based chemotherapy is highly emetogenic and clinical guidelines recommend the use of antiemetics to prevent chemotherapy-induced nausea and vomiting (CINV).

CINV can significantly decrease patient quality of life, disrupt treatment, and contribute to significant healthcare resource utilization (HCRU) and costs.

Real-world data showed that the rate of nausea and vomiting visits was significantly lower after fosnetupitant + palonosetron (NEPA) compared to fosaprepitant + palonosetron (APPA).

HCRU and costs were also lower for patients receiving NEPA compared to APPA.

These findings suggest that antiemetic regimens may not produce equivalent downstream resource use impact and that NEPA is associated with favorable economic outcomes compared to APPA.

Introduction

Chemotherapy-induced nausea and vomiting (CINV) is experienced by most patients with cancer treated with chemotherapy without antiemetic prophylaxis [1]. CINV can negatively impact patient quality of life and treatment schedules and can result in increased healthcare resource utilization (HCRU) and costs [2, 3]. CINV is most common for platinum-based chemotherapy regimens (i.e., cisplatin), which are considered highly emetogenic chemotherapy (HEC) [4,5,6]. For these reasons, guidelines recommend appropriate prophylactic antiemetics, including various combinations of a 5-hydroxytryptamine receptor 3 (5-HT3) antagonist, neurokinin-1 receptor antagonist (NK1RA), a corticosteroid such as dexamethasone, and olanzapine [1, 7]. The recommended dosing schedule for high emetic risk includes a combination prophylactic antiemetic regimen on day 1 before anticancer therapy is administered and continuing with antiemetics up to day 4 with the option of adding agent(s) from a different class of drugs for breakthrough nausea and vomiting [1].

There are many antiemetic options with clinical differences when used in different sub-populations of patients with different risk factors [8, 9]. Differences in clinical performance among antiemetic regimens may translate into downstream differences in treatment outcomes and economic outcomes. Published economic models have attempted to predict these differences [10,11,12,13]. According to one model, adding the antiemetic netupitant/fosnetupitant + palonosetron to a United States (US) payer or practice formulary led to a decrease in CINV-event related HCRU and costs, which corresponded to a net decrease in total budget [11]. Similar findings have been reported for other regions of the world [10, 13]. However, these studies relied on clinical trial data and may not reflect routine medical practice outcomes captured by real-world data. Using real-world data offers a way to directly characterize the economic impact of medical care during follow-up.

Understanding how the relative effects of different antiemetic regimens translate to real-world endpoints is essential for improving clinical practices that support the effective management of chemotherapy-related side effects. This retrospective, real-world study aimed to quantify the treatment outcomes, resource utilization, and costs associated with two antiemetic regimens, fosnetupitant + palonosetron (NEPA) and fosaprepitant + palonosetron (APPA), for the prevention of CINV in a broad US population treated with cisplatin-based chemotherapy. Both study regimens contained palonosetron but differed in the NK1RA components—fosnetupitant and fosaprepitant. We hypothesized that NEPA would be associated with better nausea and vomiting treatment outcomes and reduced resource utilization and costs, as higher response rates were reported for NEPA following cisplatin-based chemotherapy by Hata et al. [14].

Methods

Study Design and Data Sources

This retrospective database study utilized data from STATinMED RWD Insights, an all-payer medical and pharmacy claims data source from January 1, 2015 to December 31, 2020. STATinMED’s RWD was created using the all-payer claims databases (APCD) compiled and maintained by the Agency for Healthcare Research and Quality [15]. It provides insight to nearly 80% of the US healthcare system, with patient-level data from all provider types. These data are sourced directly from claims clearinghouses, which are responsible for managing claims transactions between payers and providers. The index date was defined as the date of the first intravenous (IV) antiemetic administration on or after the first observed cancer diagnosis date. The index date was also the same day that cisplatin treatment was initiated according to medical claims. The baseline period was defined as at least 12 months prior to the index date (excluding the index date), and the follow-up period was defined as 14 days after each cycle of chemotherapy, up to 12 months; patients could contribute multiple cycles of chemotherapy. Fourteen days were selected to capture the therapeutic effects of the antiemetics between cycles of chemotherapy. More than 15 days were required between each cycle of chemotherapy, as a sensitivity to the recommended guidelines of 21 days between cycles. Data were de-identified, and data use complied with the requirements of the Health Insurance Portability and Accountability Act (HIPAA) for the privacy and security of protected health information.

Patient Selection

Adults 18 years or older with at least one medical claim with a diagnosis code for any cancer during the study period were included, with the first observed cancer claim date during the study period designated as the initial cancer diagnosis date. In addition, included patients had at least one claim for NEPA or APPA and evidence of initiating cisplatin-based chemotherapy cancer treatment (inpatient or outpatient) during the identification period. Given the wide availability of dexamethasone and its enduring inclusion in clinical guidelines, its use as part of standard of care was assumed. Patients were excluded if there was insufficient data, defined as having fewer than two documented encounters during the 12 months before the initial diagnosis date. Patients with evidence of pregnancy during the study period were also excluded.

Patient Characteristics

The two study cohorts were patients who received cisplatin-based chemotherapy and received NEPA or APPA. Patient characteristics were measured on the index date, including age, sex, and payer channel (commercial, Medicare Fee-for-Service/Medicare Advantage Plan, Medicaid). Clinical characteristics included baseline National Cancer Institute (NCI) Comorbidity Index score and multiple baseline comorbidities were determined during the 12 months before the index date (Table 1).

Table 1 Baseline demographic characteristics and comorbidities
Full size table

Outcomes

Nausea and vomiting visit data were collected within 1–7 days and 8–14 days after chemotherapy. Total values 1–14 days after chemotherapy were also evaluated. Nausea and vomiting events in medical claims were identified by ICD-9 and ICD-10-CM codes. A CINV-specific claim was defined as any post-chemotherapy claim during the 14-day follow-up period with nausea or vomiting in any position of the claim. In addition, all-cause and CINV-related HCRU for inpatient and outpatient settings were assessed, and costs were adjusted to 2021 US dollars using the medical care component of the Consumer Price Index (CPI) [16]. Inpatient visits and costs were those incurred while the patient received medical care during hospital admissions. Outpatient visits and costs included medical care provided to patients without being admitted.

Statistical Analysis

All baseline patient and clinical characteristics were reported descriptively. For continuous variables, mean and standard deviation (SD) were generated. For categorical variables, counts (frequencies) and percentages were reported. Logistic regression was used to evaluate the treatment outcomes in the follow-up period for each cycle. P values, odds ratios, and 95% confidence intervals (CIs) were presented. The logistic model controlled for age group, sex, payer type, and baseline NCI Comorbidity Index score. Generalized linear models (GLMs) were used to examine all-cause and CINV-related healthcare resource utilization and costs. GLMs with Poisson distribution and log link function were applied for visit count variables, binomial distribution with log link function was used for binary outcome variables, and gamma distribution with log link function was applied for cost outcomes. GLMs were controlled for age group, sex, payer type, and baseline NCI Comorbidity Index score. p values for the comparison of the group means were provided for each HCRU and cost outcome.

Compliance with Ethics Guidelines

This study analyzed claims databases that are privately held by STATinMED, LLC. Data were de-identified, and data use complied with the requirements of the Health Insurance Portability and Accountability Act (HIPAA) for the privacy and security of protected health information. Institutional review board approval was not required for this study, as this study does not involve human or animal subjects.

Results

Patient Characteristics

A total of 4,504,995 patients were identified with a cancer diagnosis during the identification period from January 1, 2016 to October 31, 2020. After all selection criteria were applied, 1627 patients were in the NEPA cohort, and 46,340 patients were in the APPA cohort. After final consideration of patients with cancer on cisplatin therapy with same-day IV antiemetic medication and a gap of at least 15 days between the cycles of treatment, there were 421 patients included in the NEPA cohort and 15,275 patients in the APPA cohort (Table 1). The study cohorts had a similar mean age on the index date (NEPA, 61.1 years; APPA, 61.0 years). A greater proportion of patients in both cohorts were in the age category of 55–74 years (67–68%) and were slightly more likely to be male (59%). Payer type was most commonly commercial (41–48%) or Medicare (38–39%) for both cohorts. The NCI Comorbidity Index scores were 0.82 and 0.86 for NEPA and APPA, respectively. The most common baseline comorbidities were chronic obstructive pulmonary disease, diabetes, peripheral vascular disease, mild liver disease, and renal disease.

Nausea and Vomiting

As represented in Fig. 1 and Table 2, the rate of nausea and vomiting visits per 100 cycles was significantly lower in the NEPA group compared to APPA during all time periods post-chemotherapy, including 1–7 days (2.87% vs. 5.92%; adjusted odds ratio (AOR) 1.24, 95% CI 1.00–1.52; p = 0.0467), 8–14 days (1.32% vs. 3.10%; AOR 1.86, 95% CI 1.33–2.62; p = 0.0003), and 1–14 days (4.19% vs. 9.01%; AOR 1.43, 95% CI 1.19–1.71; p = 0.0001).

Fig. 1
figure 1

Rate of nausea and vomiting visits per 100 cycles. p values obtained by logistic regression, adjusted for age group, sex, payer type, and baseline NCI Comorbidity Index score. *Significantly different at p = 0.0001; §Significantly different at p = 0.0467; †Significantly different at p = 0.0003; NEPA fosnetupitant + palonosetron, APPA fosaprepitant + palonosetron. The total value may be different from the sum of the rates as a result of rounding

Full size image
Table 2 Nausea and vomiting visits by cycle
Full size table

HCRU

The mean number of all-cause inpatient visits per patient was significantly lower for NEPA vs. APPA (4.83 vs. 5.41, p = 0.0195). These differences corresponded to 57% of NEPA patients and 67% of APPA patients having one or more inpatient visits (p = 0.0002). The mean number of outpatient visits per patient was statistically similar between groups (39.23 vs. 42.90; p = 0.0835) (Table 3) with every patient in each cohort having at least one outpatient visit. The mean numbers of CINV-related inpatient (0.20 vs 0.32) and outpatient (1.32 vs 2.49) visits were lower for NEPA compared to APPA (p < 0.0001) (Table 3). These differences corresponded to 5% of NEPA patients and 7% of APPA patients with one or more inpatient visits (p = 0.0963), and 29% of NEPA patients and 39% of APPA patients with one or more outpatient visits (p = 0.0002).

Table 3 Per-patient healthcare resource utilization and costs
Full size table

Costs

Mean all-cause outpatient costs were significantly lower in the NEPA cohort compared to the APPA cohort ($83,070 vs $126,263 respectively, p < 0.0001), while mean all-cause inpatient costs were not significantly different for the two cohorts ($32,457 vs. $30,735; p = 0.0665) (Table 3). Mean CINV-related inpatient costs were significantly lower for NEPA when compared to APPA ($33 vs $163, p < 0.0001); mean CINV-related outpatient costs were numerically lower for NEPA but were not significantly different between study groups ($7848 vs. $8690, p = 0.4339) (Table 3).

Discussion

The current analysis compared CINV outcomes of two antiemetic regimens containing fosnetupitant or fosaprepitant, in combination with palonosetron. The comparative analysis found significant differences in the study outcomes despite a large difference in the number of patients in the study groups. As hypothesized, NEPA was associated with lower rates of nausea and vomiting following HEC than APPA. Of note, APPA had 86% greater odds of nausea and vomiting events during the second week (8–14 days) after chemotherapy (OR = 1.86; p = 0.0003). This improved effectiveness corresponded to lower all-cause HCRU and costs for patients treated with NEPA including a lower mean number of all-cause inpatient visits (p = 0.0195), as well as CINV-related inpatient and outpatient visits (p < 0.0001). All-cause outpatient costs (NEPA, $83,070; APPA, $126,263) and CINV-related inpatient costs (NEPA, $33; APPA, $163) were also significantly lower for NEPA (p < 0.0001).

These findings correspond to those from several clinical trials that have demonstrated the safety and efficacy of NEPA in patients undergoing chemotherapy [17,18,19,20,21]. One trial included a head-to-head comparison of NEPA with APPA and found evidence of a better safety profile for NEPA and noninferiority for overall 0–120 h complete response (defined as no emetic event and no rescue medication) [21]. In addition, existing trials have shown non-inferiority or superiority of orally administered NEPA to oral administered aprepitant + granisetron or palonosetron for CINV outcomes [17, 19, 20]. It could be the case that NEPA's effects might extend beyond the traditional 5-day period of CINV monitoring after cisplatin-based chemotherapy as the half-life of NEPA (mean of 80 h) relative to APPA (maximum of 13 h) is longer [22, 23]. Evidence aligning with longer-lasting effects of NEPA was reported in an exploratory analysis of an existing clinical trial [14, 21] which found that NEPA was 6.6% more effective than APPA in preventing CINV associated with cisplatin-based HEC during an extended 7-day period following chemotherapy. Extended effects of an antiemetic treatment could result in improved outcomes for patients.

The finding that NEPA was associated with reductions in HCRU and costs is also supported by existing research showing that hospitalization costs are significantly lower for patients who receive NEPA than those who received APPA or aprepitant + ondansetron (APON) (p < 0.0001 for both comparisons) [24]. The observed cost difference may be driven by higher mean cost per hospitalized patient and higher number of repeat hospitalizations in the APPA and APON groups. Thus the findings of this study add to the growing body of evidence suggesting that the use of NEPA is potentially more cost-effective which aligns with several existing economic analyses [12, 13, 25,26,27].

Strengths and Limitations

A primary strength of the present study is its reliance on APCD for medical and pharmacy claims, which provides insight into nearly 80% of the US healthcare system. These data are comparable with other all-payer claims repositories as they are sourced directly from claims clearinghouses. Limitations include reliance on medical claim diagnostic codes, which can be incorrectly coded or included as rule-out criteria rather than the actual disease. Not all the clinical risk factors for nausea and vomiting are available in claims data and thus could have influenced outcomes. In addition, the present study could not track the use of low-cost medications such as olanzapine that might have been recommended along with the antiemetic regimen without the submission of an insurance claim. Similarly, dexamethasone administration as well as its duration of use cannot be tracked within claims data. Lastly, statistical analyses for variables with a sample size less than 10 were not reported in accordance with the Agency for Healthcare Research and Quality’s (AHRQ) data use agreement.

Conclusion

On the basis of these real-world data, NEPA was associated with fewer nausea and vomiting visits after cisplatin-based chemotherapy, as well as reduced HCRU and costs compared to APPA. NEPA might also provide benefits from being a longer-acting treatment and offer patients extended CINV control. Overall, findings suggest that the downstream economic impact of antiemetic regimens is not equivalent and that NEPA could deliver better resource use and cost outcomes.