Budget impact of netupitant/palonosetron for the prevention of chemotherapy-induced nausea and vomiting

Abstract

Background: Chemotherapy-induced nausea and vomiting (CINV) are among the most common and debilitating side-effects patients experience during chemotherapy, and are associated with considerable acute care use and healthcare cost. It is estimated that 70–80% of CINV could be prevented through appropriate use of CINV prophylaxis; however, suboptimal CINV compliance and control remains an issue in clinical practice. Netupitant/palonosetron (NEPA) is a fixed combination of serotonin-3 (5-HT₃) and neurokinin-1 (NK₁) receptor antagonists (RAs), respectively, indicated for the prevention of acute and delayed nausea and vomiting associated with highly emetogenic chemotherapy (HEC) and moderately emetogenic chemotherapy (MEC). Phase 3 clinical trials showed a significantly higher complete response rate in both acute and delayed CINV in chemotherapy-naïve patients receiving NEPA compared to patients receiving palonosetron.

Objective: The objective of this study was to estimate the budgetary impact of adding NEPA to a US payer or practice formulary for CINV prophylaxis.

Methods: A model was developed to estimate the impact of adding NEPA to the formulary of a hypothetical US payer with 1.15 million members, including 150,000 (13%) Medicare beneficiaries. The model compared the annual total costs of CINV-related events and CINV prophylaxis in two scenarios: base year (no NEPA) and comparator year (10% and 5% NEPA usage in HEC and MEC patients, respectively). A univariate sensitivity analysis was conducted to explore the effect of variability in model parameters on the budget impact.

Results: A total of 2,021 patients were eligible to receive CINV prophylaxis. With NEPA, CINV prophylaxis costs increased by 0.7% ($3,493,630 vs $3,518,760) while medical costs associated with CINV events decreased by 3.9% ($15,118,639 vs $14,532,442), resulting in a net cost saving of $561,067 (3.0%) for the health plan ($18,612,269 vs $18,051,202), or $0.04 per member per month. This was equivalent to saving $5,011 per patient moved to NEPA. Among all 5-HT₃ RA + NK₁ RA regimens, NEPA was associated with the lowest CINV-related costs, leading to the lowest total cost of care.

Conclusions: Adding NEPA to a payer or practice formulary results in a net decrease in the total budget due to a substantial reduction in CINV event-related resource utilization and medical costs, and an increase in pharmacy costs <1%, saving over $5,000 per patient.

Keywords:

• CINV
• nausea
• chemotherapy
• cost
• netupitant
• palonosetron
• economics

JEL CLASSIFICATION CODES:

• I10
• I19

Introduction

Chemotherapy-induced nausea and vomiting (CINV) are among the most common and debilitating side-effects patients experience during chemotherapy^1–3. CINV is associated with poor outcomes and impaired dose intensity of chemotherapy^4–7, and CINV-related hospitalization and other healthcare resource use can incur substantial healthcare costs. A recent study of over 300,000 patients receiving chemotherapy found nausea and emesis to be among the 10 chemotherapy toxicities associated with avoidable emergency department (ED) visits and hospitalizations⁸, leading to their inclusion in OP-35, the first national oncology outcome measure instituted by the US government⁹.

Antiemetic guidelines from the American Society of Clinical Oncology (ASCO) and the National Comprehensive Cancer Network (NCCN) recommend CINV prophylaxis including a serotonin-3 receptor antagonist (5-HT₃ RA), a neurokinin-1 receptor antagonist (NK₁ RA), and dexamethasone (dex) for all patients treated with highly emetogenic chemotherapy (HEC)¹⁰. For patients undergoing moderately emetogenic chemotherapy (MEC), suggested prophylaxis includes a 5-HT₃ RA with or without an NK₁ RA^1,11. It is estimated that 70–80% of CINV could be prevented through appropriate prophylaxis¹², but suboptimal physician adherence to antiemetic prophylaxis guidelines remains an issue in clinical practice^13,14.

NEPA (netupitant/palonosetron) is a fixed combination of netupitant (an NK₁ RA) and palonosetron (a 5-HT₃ RA) that is indicated and included within national guidelines for the prevention of acute and delayed nausea and vomiting associated with both HEC and MEC.

The cost of CINV prophylaxis is often borne by the pharmacy function within a healthcare organization, while the benefits of avoiding CINV-associated medical consequences and costs typically accrue to the medical function. Increasingly, organizations are evolving to evaluate cancer on a “total cost of care” basis, particularly when the potential exists for medical savings that outweigh pharmacy expenditures. However, payers, at-risk oncology practices, and other groups selecting agents for CINV prophylaxis often lack information on the total cost of CINV that incorporates the costs of both the prophylaxis and avoided CINV events. The objective of this study was to estimate the budgetary impact of adding NEPA to a US payer or practice formulary for CINV prophylaxis.

Methods

Overview

A model was developed to estimate the impact of adding NEPA to the formulary of a hypothetical US payer with 1.15 million members, including 150,000 (13%) Medicare beneficiaries. The model compared the annual total costs of CINV-related events and CINV prophylaxis from a payer perspective, by comparing two scenarios: (1) a base year without the use of NEPA and (2) a comparator year with 10% and 5% usage of NEPA in patients undergoing HEC and MEC, respectively. Market shares of competing antiemetic agents in the comparator year were reduced proportionally according to their market share in the base year.

Model parameters

Population

US epidemiological data, guideline recommendations, and guideline compliance rates were used to estimate the model’s target population. In 2014, the prevalence of actively treated cancer was estimated at 0.4% among patients covered by commercial health plans and 2.6% among Medicare beneficiaries¹⁵; 39.8% and 64.6% of patients in each respective group received infused chemotherapy¹⁵. Nearly half (49%) of patients treated with infused chemotherapy received MEC or HEC¹⁶. Treatment guidelines recommend all HEC patients receive a three-drug regimen (5-HT₃ RA + NK₁ RA + dex) for CINV prophylaxis. With 90% guideline compliance, as reported in prior literature¹⁷, it was assumed that the remaining 10% of patients receive a two-drug regimen consisting of 5-HT₃ + dex. For patients receiving MEC, it was assumed that guideline recommendations would result in 10% of patients receiving a 5-HT₃ RA + NK₁ RA + dex regimen and the remainder a 5-HT₃ RA + dex regimen. Guideline compliance among patients with MEC was assumed to be 73% and 98% for the three-drug and two-drug regimen, respectively¹⁷. Non-guideline-recommended therapy was assumed to consist of 5-HT₃ RA + dex for patients eligible for the recommended 5-HT₃ RA + NK₁ RA + dex regimen, and dex monotherapy for patients eligible to receive 5-HT₃ RA + dex.

Market share

The base year treatment mix of CINV prophylaxis agents other than NEPA among patients was based on unpublished market research data for patients undergoing HEC and MEC, due to the unavailability of published references (Table 1). In the comparator year, we assumed that 10% of patients with HEC and 5% of patients with MEC would receive NEPA; these patients were moved proportionately from each of the agents utilized in the base year. The remaining 90% of patients with HEC and 95% of patients with MEC would receive the same CINV prophylaxis agents in the base year and the comparator year. There were no differences assumed in patient characteristics for those receiving NEPA vs non-NEPA agents.

Table 1. Market share.

Product	HEC		MEC
	Base year	Comparator year	Base year	Comparator year
5-HT3 RA
Palonosetron IV	55.0%	49.5%	55.0%	52.2%
Granisetron IV	0.0%	0.0%	0.0%	0.0%
Granisetron SubQ	0.0%	0.0%	0.0%	0.0%
Ondansetron IV	45.0%	40.5%	45.0%	42.8%
NK1 RA
Fosaprepitant IV	95.0%	85.5%	9.5%	4.8%
Aprepitant PO	5.0%	4.5%	0.5%	0.2%
Rolapitant PO	0.0%	0.0%	0.0%	0.0%
Aprepitant IV	0.0%	0.0%	0.0%	0.0%
Combination 5-HT3/NK1 RA
Netupitant/palonosetron PO	0.0%	10.0%	0.0%	5.0%
Total 5-HT3	100.0%	100.0%	100.0%	100.0%
Total NK1	100.0%	100.0%	10.0%	10.0%

Abbreviations. 5-HT₃, serotonin-3; HEC, highly emetogenic chemotherapy; IV, intravenous; MEC, moderately emetogenic chemotherapy; NK₁, neurokinin-1; PO, oral; RA, receptor antagonist; SubQ, subcutaneous.

CINV prophylaxis medication costs

Dose per cycle was obtained from the US Food and Drug Administration (FDA) prescribing information (PI) for each prophylaxis agent; cost per dose was estimated using the Average Sales Price (ASP) plus 6% for intravenous products and wholesale acquisition cost (WAC) plus 3% for oral products (Table 2). Additionally, a dispensing fee and an assumed $50 copayment were included for oral products, and administration costs were included for injectable/infused products. Patients were assumed to have six cycles of chemotherapy and CINV prophylaxis.

Table 2. Prophylaxis cost (2017 USD).

Product	Dose	Cost per dose*	Cost per treatment cycle^a	Plan cost per treated patient^b
5-HT3 RA
Palonosetron IV	0.25 mg	$200.20	$222.81	$1,336.86
Granisetron IV	7.5 mg	$20.10	$42.71	$256.26
Granisetron SubQ	10 mg	$434.40	$457.01	$2,742.06
Ondansetron IV	16 mg	$1.39	$24.00	$144.01
NK1 RA
Fosaprepitant IV	150 mg	$314.70	$337.31	$2,023.86
Aprepitant PO	Tri-Pack	$774.36	$749.46	$4,496.76
Rolapitant PO	180 mg	$595.50	$565.24	$3,391.41
Aprepitant IV	130 mg	$312.70	$335.31	$2,011.86
Combination 5-HT3/NK1 RA
Netupitant/palonosetron PO	1 capsule	$560.00	$528.67	$3,172.02
Steroids
Dexamethasone IV	12 mg/8 mg	$3.50	$26.11	$156.66

^a Incorporates $50 copayment and $1.87 dispensing fee for oral products or $22.61 administration cost for intravenous and subcutaneous products.

^b Assumes six cycles of chemotherapy.

* All costs are as of December 2017.

Abbreviations. 5-HT₃, serotonin-3; IV, intravenous; NK₁, neurokinin-1; PO, oral; RA, receptor antagonist; SubQ, subcutaneous.

Medical costs of CINV events

The medical cost of CINV-related events per cycle for each regimen was calculated as the product of: (a) the rate of patients having CINV events despite prophylaxis; (b) the percentage of CINV events requiring resource use; and (c) the medical cost per CINV event requiring resource use. CINV rates were assumed to be independent of the cycle in which they occurred; therefore, overall CINV risk rates for multiple-cycle chemotherapy were calculated as having cumulative likelihood. The cost of a CINV resource use event was assumed to be consistent, regardless of whether the patient received HEC or MEC.

(a) CINV event rate: The overall CINV event rate was determined using PIs and clinical trial data^18–26. The hierarchy of data sources used was: (1) PI, (2) clinical trial publication supporting the regimen’s FDA approval, and (3) a weighted average of the published clinical studies on the regimen.

(b) CINV events resulting in resource use: The model assumed that not all CINV events involved resource use; some patients experience CINV symptoms, yet do not receive medical treatment for them. Burke et al.⁴ reported that, per cycle, 18.2% of patients receiving HEC and 13.0% of patients receiving MEC had CINV-related resource use (CINV-RU), rates well below the known overall CINV per cycle event rates associated with prophylaxis. To calculate the proportion of CINV events involving CINV-RU in this model, the rates from Burke et al.⁴ were divided by the weighted average CINV event rate for model treatments; this yielded proportions of 76% for HEC and 33.6% for MEC of patients experiencing CINV who also incurred related resource use. We then applied these proportions to the CINV rate per cycle for each prophylaxis regimen, to determine the CINV-RU event rate per cycle for each prophylaxis regimen (Table 3).

Table 3. Prophylaxis regimen-specific CINV event-related medical costs.

Prophylactic regimen^a	CINV event rate/cycle	CINV-related RU^b/cycle	Average CINV medical costs (per treated patient)^c
HEC
PALO IV + FOS IV or APR IV or APR PO	21.5%^21^,^22	16.3%	$1,382
(PALO IV or GRAN SubQ) + ROL PO	29.9%^24	22.7%	$1,922
(OND IV or GRAN IV) + (FOS IV or APR IV or APR PO)	27.0%^19	20.5%	$1,735
OND IV or GRAN IV + ROL PO	29.9%^24	22.7%	$1,922
GRAN SubQ + FOS IV or APR IV	21.5%^23	16.3%	$1,382
NEPA PO	10.4%^25^,^26	7.9%	$668
MEC
PALO IV monotherapy	31.0%^20	10.4%	$882
GRAN SubQ monotherapy	36.7%^23	12.3%	$1,044
OND IV or GRAN IV monotherapy	48.0%^19	16.1%	$1,366

^a All regimens are used in combination with dexamethasone.

^b After adjusting CINV rate per cycle by 76.0% for HEC and 33.6% for MEC, per data on resource use per cycle from Burke et al.⁴

^c Values reflect total costs of CINV events (i.e. not including 5-HT₃ RA or NK₁ RA costs) assuming six cycles per patient.

Abbreviations. APR, aprepitant; CINV, chemotherapy-induced nausea and vomiting; FOS, fosaprepitant; GRAN, granisetron; HEC, highly emetogenic chemotherapy; IV, intravenous; MEC, moderately emetogenic chemotherapy; NEPA, netupitant/palonosetron; OND, ondansetron; PALO, palonosetron; PO, oral; ROL, rolapitant; RU, resource utilization; SubQ, subcutaneous.

(c) CINV-related event costs: The average cost per CINV-RU event was sourced from several studies and databases. In a retrospective claims analysis, Burke et al.⁴ reported that 64.3% of CINV-RU was inpatient, 25.9% ambulatory, and 9.8% ED utilization. The event cost in each setting was obtained from multiple sources, resulting in three scenarios (Table 4):

Table 4. Data sources and inputs for CINV-related medical event costs.

Source	Data year, USD	Inpatient	Ambulatory	Emergency department	Weighted average cost^a (2017 USD)
Burke et al.^4	2005	$7,448	$1,494	$918	$7,778
Rashid et al.^27	2013	$10,074	$91	$250	$7,329
HCUP and other^28^,^31^,^32	2014	$14,284	$240	$1,233	$10,275

^a Relative frequency of HCRU: Inpatient 64.3%, Ambulatory 25.9%, Emergency department 9.8% (Source: Burke et al.⁴).

Abbreviations. CINV, chemotherapy-induced nausea and vomiting; HCUP, Healthcare Cost and Utilization Project; HCRU, healthcare resource use; USD, United States dollars.

1. Burke et al.⁴ estimated inpatient, ambulatory, and ED visits to cost $7,448, $1,494, and $918, respectively, in 2005 US dollars (USD).

2. A chart review study in an integrated healthcare delivery system identified breast cancer patients experiencing adverse events from chemotherapy²⁷, and reported costs for nausea/vomiting of $10,074 (inpatient), $91 (ambulatory), and $250 (ED) in 2013 USD.

3. The Healthcare Cost and Utilization Project (HCUP) by the Agency for Healthcare Research and Quality reported an average charge of $23,063 for a hospital admission related to nausea and vomiting in the year 2014²⁸. Considering a 49% reimbursement rate and an additional 26.4% physician service fee, we derived a cost of $14,284 for each hospitalization^29,30.

   • For this scenario, ambulatory cost was estimated to be $240 (2014 USD) per visit based on average revenue per established patient ($1,500) in the National Practice Benchmark for Oncology report, adjusted by 16% (i.e. evaluation and monitoring revenue as a percentage of total revenue)³¹. ED costs for this scenario were assumed to equal the median charge of $1,233 (2014 USD) for ED visits on 10 conditions from a nationally representative database³².

The three cost scenarios, each weighted by Burke’s percentage of setting, were averaged and adjusted to 2017 USD. The average cost ($8,461 per CINV event with resource use) was used in the model.

Model analysis

The model estimated the budget impact to the health plan and incremental per member per month (PMPM) and per patient costs of including NEPA on the health plan’s formulary. The total cost to the health plan was composed of prophylaxis medication costs plus CINV-RU medical costs for each regimen. The prophylaxis medication cost included acquisition costs of CINV prophylaxis plus administration or dispensing fees exclusive of patient copayments. The medical cost of CINV-RU for each regimen was calculated as the product of: (1) the rate of patients with CINV, (2) the proportion of CINV episodes requiring resource use, and (3) the medical cost per CINV-RU event. Total costs to the health plan were further influenced by population size, mix of market share, and the number of chemotherapy cycles.

Sensitivity analysis

A univariate sensitivity analysis was used to explore the effect of variability in the model parameters on the budget impact. Costs were increased or decreased by 10% from the current cost (except palonosetron, which was decreased by 65% to reflect patent expiration), the market share of NEPA in the comparator year was halved or doubled, and the number of cycles of chemotherapy was varied by one; the remaining parameters were increased or decreased by 10 percentage points (within 0–100%). Separate scenario analyses including additional recent NK₁ agents were also conducted.

Results

Based on the epidemiological data, in a hypothetical 1.15 million-member health plan with 150,000 Medicare beneficiaries, 8,130 patients were estimated to receive active treatment for cancer, including 4,210 receiving infused chemotherapy. Within the infused chemotherapy population, 716 patients received HEC and 1,305 patients MEC, for a total of 2,021 patients eligible to receive CINV prophylaxis with 5-HT₃ RA ± NK₁ RA + dex. After adjusting for guideline compliance rates, 738 patients were estimated to receive 5-HT₃ RA + NK₁ RA + dex (triple prophylaxis) and 1,257 patients 5-HT₃ RA + dex, with 26 patients receiving only dex monotherapy.

Comparing the scenarios to evaluate the impact of adding NEPA showed an $25,130 (0.7%) increase in CINV prophylaxis cost of from the base year ($3,493,630 vs $3,518,760). Medical cost associated with CINV events decreased $586,197 (3.9%), from $15,118,639 to $14,532,442. Together these produced a net cost saving of $561,067 (3.0%) for the total CINV cost of care for the health plan ($18,612,269 vs $18,051,202), equivalent to savings of $0.04 PMPM (Table 5), $760.25 per patient receiving triple prophylaxis, or $5,011 per patient moved to NEPA.

Table 5. Base case results (2017 USD).

Product	Base year	Comparator year	Difference
Drug costs	$3,493,630	$3,518,760	$25,130
CINV medical costs	$15,118,639	$14,532,442	−$586,197
Total costs	$18,612,269	$18,051,202	−$561,067

Abbreviations. CINV, chemotherapy-induced nausea and vomiting; USD, United States dollars.

Patients receiving palonosetron had the lowest total CINV cost, including prophylaxis cost and CINV medical cost, among the 5-HT₃ RA + dex regimens, while patients receiving NEPA incurred the lowest total cost among the triple prophylaxis regimens. The low cost for NEPA was driven principally by low CINV event rates (Table 6; Figure 1). The average prophylaxis cost per patient was highest for granisetron + rolapitant + dex ($6,290) and lowest for ondansetron + dex ($301); prophylaxis cost per patient for NEPA + dex was $3,329. CINV medical costs were highest for regimens with rolapitant ($11,531) and lowest for NEPA + dex ($4,011). For the health plan, CINV prophylaxis and medical costs represented 19% and 81% of total CINV cost of care, respectively, in the baseline period.

Figure 1. Per-patient CINV total costs (2017 USD) by CINV prophylaxis regimen, with prophylaxis and CINV medical cost breakout. Abbreviations. APR, aprepitant; CINV, chemotherapy-induced nausea and vomiting; FOS, fosaprepitant; GRAN, granisetron; IV, intravenous; MEC, moderately emetogenic chemotherapy; NEPA, netupitant/palonosetron; OND, ondansetron; PALO, palonosetron; PO, oral; ROL, rolapitant; SubQ, subcutaneous; USD, United States dollars.

Table 6. Per patient costs (2017 USD) by CINV regimen.

Prophylactic regimen	Prophylaxis cost^a	CINV medical cost^a	Total cost^a
5-HT3 RA + dexamethasone
PALO IV monotherapy	$1,494	$5,293	$6,787
GRAN SubQ monotherapy	$2,899	$6,266	$9,165
OND IV monotherapy	$301	$8,196	$8,496
GRAN IV monotherapy	$413	$8,196	$8,609
5-HT3 RA + NK1 RA + dexamethasone
PALO IV + FOS IV	$3,517	$8,292	$11,809
PALO IV + APR PO	$5,990	$8,292	$14,282
PALO IV + ROL PO	$4,885	$11,531	$16,416
PALO IV + APR IV	$3,505	$8,292	$11,797
GRAN SubQ + FOS IV	$4,923	$8,292	$13,214
GRAN SubQ + ROL PO	$6,290	$11,531	$17,822
OND IV + FOS IV	$2,325	$10,413	$12,737
OND IV + APR PO	$4,797	$10,413	$15,210
OND IV + ROL PO	$3,692	$11,531	$15,223
OND IV + APR IV	$2,313	$10,413	$12,725
GRAN IV + FOS IV	$2,437	$10,413	$12,850
GRAN IV + APR PO	$4,910	$10,413	$15,323
GRAN IV + ROL PO	$3,804	$11,531	$15,336
GRAN IV + APR IV	$2,425	$10,413	$12,838
NEPA PO	$3,329	$4,011	$7,340

^a Assumes six cycles of chemotherapy and prophylaxis per patient.

Abbreviations. 5-HT₃, serotonin-3; APR, aprepitant; CINV, chemotherapy-induced nausea and vomiting; FOS, fosaprepitant; GRAN, granisetron; IV, intravenous; NEPA, netupitant/palonosetron; NK₁, neurokinin-1; OND, ondansetron; PALO, palonosetron; PO, oral; RA, receptor antagonist; ROL, rolapitant; SubQ, subcutaneous; USD, United States dollars.

Within the expected ranges of key input parameters, total costs in the univariate sensitivity analysis were consistently lower in the comparator year compared to the base year (Figure 2), favoring NEPA. Results were most sensitive to CINV event rates, the market share of prophylaxis regimens, the proportion of Medicare beneficiaries in the health plan, and the proportions of patients receiving HEC and MEC. A 10-percentage-point increase (worsening) in CINV event rate with NEPA continued to produce cost savings ($129,000). Adding NEPA would continue to be cost-saving (>$420,000), even with decreased CINV event rates for palonosetron and fosaprepitant, or with reduced palonosetron costs associated with generic availability. Separate scenario analyses, including additional recent NK₁ agents, did not materially change the results.

Figure 2. Sensitivity analysis results. Abbreviations. APR, aprepitant; CINV, chemotherapy-induced nausea and vomiting; FOS, fosaprepitant; HEC, highly emetogenic chemotherapy; IV, intravenous; MEC, moderately emetogenic chemotherapy; NEPA, netupitant/palonosetron; NK₁, neurokinin-1; PO, oral.

Discussion

The primary objective of this analysis was to evaluate the budget impact of NEPA for the treatment of CINV from a US payer perspective over 1 year, assuming 10% and 5% NEPA use among patients with HEC and MEC, respectively. Despite the higher prophylaxis medication cost of NEPA compared to generically available agents, results from our analysis suggest that adding NEPA to the payer formulary is cost-saving due to its efficacy at preventing CINV and thereby avoiding high CINV-associated medical costs. The expected cost saving to the payer was over $560,000 ($0.04 PMPM), amounting to over $5,000 per patient moved to NEPA.

The sensitivity analysis confirmed the robustness of the results and continued to predict cost savings with varied parameters. Particularly, with patent expirations on several agents (e.g. palonosetron, fosaprepitant) included in the model, lower-cost generic drugs may be available for CINV prophylaxis. However, regardless of cost reductions from generic prophylaxis agents, NEPA would still be cost-saving, as the savings it yields—driven by lower CINV-related medical events and costs—far outweigh its incremental costs of prophylaxis. The opportunity and need for medical cost savings are substantial, representing over 80% of the total costs of care for CINV.

With rising attention on opportunities to improve cancer treatment outcomes while reducing costs, decision-makers are increasingly open to evaluating medical cost offsets to potential pharmacy expenditure increases. Nevertheless, the short-term allure of pharmacy cost reductions and the opaque or undocumented impact of medical costs often necessitate that a sizeable risk/reward ratio be demonstrated to support any pharmacy “investment”. The >10:1 ratio of medical savings to pharmacy incremental spend seen with NEPA may be sufficiently meaningful to provoke consideration from decision-makers, particularly when aligned with recent Centers for Medicare and Medicaid Services (CMS) clinical initiatives highlighting the benefits of reducing avoidable chemotherapy toxicities including nausea and emesis^8,33.

Previous studies have also suggested cost savings with NEPA. Restelli et al.³⁴ developed a Markov model from the Italian National Health Service perspective to evaluate the budget impact of NEPA on a 5-year time horizon. The authors did not separate CINV events from resource use, but estimated CINV episode costs and added costs for injection site reactions and anorexia. Using this costing method, the authors found NEPA was likely to lead to save €63.7 million over 5 years for the estimated chemotherapy cycles in Italy. Another study developed a three-state (complete response, complete protection, and incomplete response) Markov model from the UK National Health Service perspective based on the NEPA clinical trial results³⁵. Other agents considered in the model were aprepitant + palonosetron or palonosetron alone. Over 5 years, the model showed a cumulative cost saving of nearly £14 million. While the actual monetary values should not be compared head-to-head across different countries, both the Italian and English studies were consistent with our finding that the lower rates of CINV-associated events with NEPA led to overall cost savings, regardless of additions to prophylaxis costs.

Limitations

This study has several limitations. First, the epidemiological data used for calculating the target population were not available from a single source. Therefore, multiple data sources were used to calculate the target population in a step-wise fashion using a number of assumptions. Additionally, the market shares used in the model may not be generalizable to a particular payer, and the budget impact may change if market shares change or additional drugs are approved and used. To mitigate this limitation, we reported per patient costs by treatment regimen.

Next, medical costs were sourced from claims-based evidence, which may under- or over-estimate CINV-associated resource use and costs due to potential coding errors or missing data. Nevertheless, the inclusion of multiple sources, all of which present substantially similar depictions of cost despite the use of various sources and methods, is believed to mitigate this limitation. Additionally, assigning costs to CINV events is difficult since CINV could be one of multiple factors driving resource utilization, and not all CINV events result in resource utilization. To address these limitations, conservative estimates were used as default values, which may result in under-statement of the costs associated with CINV events.

Another limitation was that the rates of CINV events per cycle were often determined in separate trials, a common practice in budget impact models, yet one involving indirect comparisons that may not account for differences in study populations. The calculation of the proportion of CINV events requiring resource use was based on a weighted average of CINV rates for model treatments, and thus varied based on the market shares used in the model. A sensitivity analysis showed that, while the rates impact the numeric values of the cost savings, the topline results remain consistently in favor of NEPA.

This analysis does not consider the potential benefit of avoided impairment of chemotherapy dose or duration due to fewer CINV events. Patient copayments, which vary based on individuals’ plan coverage, were assumed to be $50; the univariate sensitivity analysis, showing varying copayments from $0 to $100, had a minimal impact on the overall budget (result not reported). Finally, the costs of rescue medication and side-effects associated with the use of prophylaxis were not included in the model. However, the costs of rescue medication are modest, and the resource use associated with side-effects are typically low; therefore, the economic impact of not including these costs is expected to be insignificant¹¹.

Conclusion

In conclusion, the model shows that adding NEPA to a payer or practice formulary results in a net decrease in the total budget due to a substantial reduction in CINV event-related resource utilization and medical costs, and increase in pharmacy costs <1%, saving over $5,000 per patient moved to NEPA. With mounting cost pressure in oncology, and the creation of US government mandates measuring avoidable hospitalization for toxicities including nausea and emesis, payers and practices may find enhancement of CINV prophylaxis to be a worthwhile clinical and economic investment.

Transparency

Declaration of financial/other relationships

GB is an employee of Helsinn Therapeutics (US), Inc. SP, SC, and MB are employees of Pharmerit International, which has received consultancy fees from Helsinn Therapeutics (US), Inc. for this study. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Previous presentations

This study was presented at the AMCP Managed Care & Specialty Pharmacy Annual Meeting, Boston, MA, April 23–26, 2018, as a poster abstract.

Acknowledgements

None reported.

Additional information

Funding

The study was funded by Helsinn Therapeutics (US), Inc.