Cost-efficiency and budget-neutral expanded access modeling of the novel PD-1 inhibitor toripalimab versus pembrolizumab in recurrent or metastatic nasopharyngeal carcinoma

Abstract

Aims

To estimate, in the setting of recurrent or metastatic nasopharyngeal carcinoma (R/M NPC) for an assumed 1,207 incident US cases in 2024, (1) the cost-efficiency of a toripalimab-gemcitabine-cisplatin regimen compared to a similar pembrolizumab regimen; and (2) the budget-neutral expanded access to additional toripalimab cycles and regimens afforded by the accrued savings.

Methods

Simulation modeling utilized two cost inputs (wholesale acquisition cost (WAC) at market entry and an ex ante toripalimab price point of 80% of pembrolizumab average sales price (ASP)) and drug administration costs over 1 and 2 years of treatment with treatment rates ranging from 45% to 90%. In the absence of trial data for pembrolizumab-gemcitabine-cisplatin in R/M NPC, it is assumed that such a regimen would be comparable to toripalimab-gemcitabine-cisplatin in efficacy and safety.

Results

In the models utilizing the WAC, toripalimab saves $2,223 per patient per cycle and $40,014 over 1 year of treatment ($77,805 over 2 years). Extrapolated to the 1,207-patient panel, estimated 1-year savings range from $21,733,702 (45% treatment rate) to $43,467,404 (90% rate). Reallocating these savings permits budget-neutral expanded access to an additional 2,359 (45% rate) to 4,717 (90% rate) toripalimab maintenance cycles or to an additional 126 (45% rate) to 252 (90%) full 1-year toripalimab regimens with all agents. Two-year savings range from $42,259,976 (45% rate) to $84,519,952 (90% rate). Reallocating these efficiencies provides expanded access, ranging from an additional 4,586 (45% rate) to 9,172 (90% rate) toripalimab cycles or to an additional 128–257 full 2-year toripalimab regimens. The ex ante ASP model showed similar results.

Conclusion

This simulation demonstrates that treatment with toripalimab generates savings that enable budget-neutral funding for up to an additional 252 regimens with toripalimab-gemcitabine-cisplatin for one full year, the equivalent of approximately 21% of the 2024 incident cases of R/M NPC in the US.

PLAIN LANGUAGE SUMMARY

An estimated 1,207 patients will be diagnosed with late-stage nasopharyngeal cancer in the US in 2024. Toripalimab is a novel PD-1 inhibitor drug approved by the US Food and Drug Administration on October 27, 2023 as first-line treatment for patients with recurrent or metastatic nasopharyngeal cancer when used in combination with gemcitabine and cisplatin. We conducted economic evaluations of the costs of this toripalimab regimen versus the costs of a similar regimen with the PD-1 inhibitor pembrolizumab. Our simulation models used two pricing scenarios: the wholesale acquisition cost (WAC) or “list price” at market entry and, as no average sales price (ASP) will be available for toripalimab for several quarters, an estimated toripalimab price point of 80% of the pembrolizumab ASP. We compared the savings in each scenario when between 45% and 90% of the 1,207 patients are treated with the toripalimab regimen. We then evaluated how these savings could be re-allocated, on a budget-neutral basis and without requiring extra cash outlays, to provide more patients with access to toripalimab treatment; specifically, how many toripalimab doses and how many full toripalimab regimens could be purchased to provide more patients with treatment. We found that, if 90% of new cases of recurrent or metastatic nasopharyngeal cancer were treated with toripalimab over 1 year, these savings are enough to purchase up to 4,717 additional doses on a budget-neutral basis, which could provide up to an additional 252 newly diagnosed patients with 1 year of treatment with toripalimab. In combination with gemcitabine and cisplatin, toripalimab can markedly improve access to care for patients with recurrent or metastatic nasopharyngeal cancer in a cost-responsible way.

Keywords:

• Toripalimab
• PD-1 inhibitor
• nasopharyngeal carcinoma
• cost-efficiency
• expanded access
• metastatic cancer
• pharmacoeconomics

JEL CLASSIFICATION CODES:

• I11
• I13

Introduction

Nasopharyngeal carcinoma (NPC) is a rare disease with less than 1 new diagnosis per 100,000 population annually per year¹, but with significant variability by gender and ethnicity. Males, persons of Asian heritage, and smokers are disproportionately affected, including in the US, where, moreover, a strong association with Epstein-Barr Virus (EBV) infections has been observed^2–5. Treatment for NPC may include surgery, radiotherapy, and/or platinum-based chemotherapy with or without a programmed death receptor-1 (PD-1)-blocking antibody, depending on disease stage.

Toripalimab, a novel PD-1 inhibitor, was approved in China in 2018 (Tuoyi, Junshi Biosciences, Shanghai, China) and by the US Food and Drug Administration (FDA) on October 27, 2023 as LOQTORZI (toripalimab-tpzi)ⁱ (Coherus BioSciences, Inc., Redwood City, CA, USA), where it had received breakthrough status because it addresses the unmet medical need in recurrent or metastatic NPC (R/M NPC). In the JUPITER-02 trial, a randomized double-blind phase 3 trial in first-line treatment of R/M NPC, toripalimab 240 mg or placebo was administered in combination with gemcitabine-cisplatin every 3 weeks (Q3W) for up to six cycles; followed by monotherapy with toripalimab or placebo Q3W until disease progression, intolerable toxicity, or completion of 2 years of treatment⁶. Toripalimab demonstrated a median progression-free survival (PFS) benefit of 13.2 months over placebo (21.4 months versus 8.2 months). Over a median follow-up of 36 months, median overall survival in patients treated with toripalimab plus chemotherapy was not evaluable (NE; 95% CI = 38.7 months–NE) compared to 33.7 months (95% CI = 27.0–44.2) for patients who received placebo plus chemotherapy⁶.

Keytruda (pembrolizumab, Merck & Co., Inc., Whitehouse Station, NJ), an FDA-approved PD-1 inhibitor, was studied as monotherapy for advanced or metastatic NPC in two clinical trials. In the KEYNOTE-028 study, a multi-cohort non-randomized (single group assignment) phase 1b trial of previously treated patients across 20 different advanced solid malignancies, the analysis of the 27 patients with advanced or metastatic NPC demonstrated median PFS of 3.7 months (95% CI = 2.1–13.4) by central review and 6.5 months (95% CI = 3.6–13.4) by investigator assessment over a median follow-up of 20 months⁷. The KEYNOTE-122 open-label randomized phase III trial compared pembrolizumab monotherapy versus single-agent non-platinum chemotherapy in platinum pre-treated EBV-positive patients with R/M NPC⁸. Over a median follow-up of 45.1 months, median PFS of 4.1 months (95% CI = 2.1–5.6) for pembrolizumab versus 5.5 months (95% CI = 4.0–8.1) for chemotherapy (HR = 1.28; 95% CI = 0.94–1.75), and a median OS of 17.2 months (95% CI = 11.7–22.9) versus 15.3 months (95% CI = 10.9–18.1) (HR = 0.90; 95% CI = 0.67–1.19) were observed, neither comparison attaining statistical significance. Approved for a number of indications including head and neck squamous cell cancer, the prescribing label for pembrolizumab⁹ does not currently include NPC. The label-specified dose of pembrolizumab in all adult indications is 200 mg Q3W or 400 mg Q6W⁹.

The most recent National Comprehensive Cancer Network (NCCN) guidelines for head and neck cancers (Version 3.2024) list gemcitabine-cisplatin with or without toripalimab as category 1 preferred regimens in first-line (while also referencing, though with indirect evidence, gemcitabine-cisplatin with PD-1 inhibitors such as pembrolizumab or nivolumab as category 2A preferred regimens)¹⁰. Toripalimab is the only NCCN-preferred regimen in subsequent-line treatment, with pembrolizumab being a recommended regimen (category 2B). Inclusion of toripalimab in the US-based NCCN guidelines follows the earlier recommendation of the addition of toripalimab to gemcitabine and cisplatin followed by toripalimab maintenance therapy as first-line treatment in R/M NPC by the European Society for Medical Oncology in 2022¹¹.

At the time of this study, the list price for pembrolizumab was $11,115.04 per dose¹² or $200,071 for 1 year of treatment. The market-entry list price for toripalimab has been set¹³ and corresponds to 80% of the pembrolizumab list price, offering the potential for significant cost-savings. Assuming that a future label expansion for pembrolizumab in combination with chemotherapy (as opposed to monotherapy) may include R/M NPC, we present herein the results of a pharmacoeconomic simulation analysis for the US, on the premise of R/M NPC treatment with pembrolizumab in accordance with the NCCN recommended regimen. Specifically, our simulation analysis (1) compares the cost-efficiency of treatment with toripalimab plus gemcitabine-cisplatin versus a pembrolizumab regimen including gemcitabine-cisplatin in a panel of patients with R/M NPC; (2) determines the number of patients that would need to be treated with the toripalimab regimen instead of the pembrolizumab regimen in order to purchase, budget-neutrally at no additional cost, one additional maintenance cycle of toripalimab and one additional regimen (all agents); and (3) estimates the expanded access to toripalimab treatment, both in terms of additional toripalimab maintenance cycles or full 1- and 2-year toripalimab regimens with all agents, that could be provided on a budget-neutral basis from the cost-savings generated by using toripalimab instead of pembrolizumab.

Note that this present paper is part of a larger set of papers reporting on cost-efficiency and expanded access simulations and budget impact analyses of toripalimab in the settings of nasopharyngeal carcinoma¹⁴, nonsquamous non-small cell lung cancer^15,16, and esophageal squamous cell carcinoma^17,18. This is complemented by an introduction to and summary of these studies¹⁹.

Methods

Model

Building on prior cost-efficiency and expanded access studies^20–23, simulation models were developed in Microsoft Excel to estimate the cost-efficiency of and expanded access to toripalimab relative to pembrolizumab for a panel of 1,207 covered cancer patients with R/M NPC from the US payer perspective in 2024. In order, we (1) determined the per-patient savings and total savings for a panel of 1,207 patients generated from treatment with toripalimab instead of pembrolizumab as a function of cost input, treatment rates, and duration of treatment; (2) calculated the number of patients needed to be treated with toripalimab instead of pembrolizumab in order to purchase one additional maintenance cycle of toripalimab or one additional 1- or 2-year toripalimab-gemcitabine-cisplatin regimens; and (3) estimated the number of additional patients who could be treated with toripalimab on a budget-neutral basis from the cost-savings generated from toripalimab treatment in the 1,207-patient panel.

Using publicly available cost data for medication and administration, we first calculated the cumulative per-patient costs of toripalimab or pembrolizumab in combination with gemcitabine-cisplatin per cycle and across 1 and 2 years of treatment. We then computed the per-patient cost-savings of toripalimab over pembrolizumab per cycle and across 1 year (18 cycles) and 2 years of treatment (35 cycles). These calculations were performed for two cost inputs: the “launch” wholesale acquisition cost (WAC) at market entry, which is 80% of the pembrolizumab WAC, and, considering that no average sales price (ASP) will be available for several quarters, an ex ante ASP pegged at 80% of the pembrolizumab ASP. The per-patient savings estimates from treatment with toripalimab were then escalated to a panel of 1,207 patients, which is the estimated 2024 incidence of patients with R/M NPC. We determined the total savings stratified by cost input (WAC and ASP) and treatment rates of using toripalimab instead of pembrolizumab (descending from 90% to 45% in decrements of 5%).

We then estimated the number of patients needed-to-be-treated (NNT) with toripalimab instead of pembrolizumab in order to purchase either one maintenance cycle of toripalimab or one additional regimen of toripalimab-gemcitabine-cisplatin for 1 or 2 years. NNT was calculated as: cost of one unit of toripalimab/(cost of one unit of pembrolizumab − cost of one unit of toripalimab).

Lastly, we divided the estimated per-panel cost-savings by the cost of toripalimab in the 1,207-patient panel to determine the incremental number of additional maintenance cycles of toripalimab that could be purchased and the additional patients that could be treated with 1- and 2-year regimens of toripalimab-gemcitabine-cisplatin on a budget-neutral basis.

Assumptions

The following assumptions apply:

• The toripalimab JUPITER-02 trial⁶ includes first-line combination therapy with gemcitabine-cisplatin while the two pembrolizumab trials^7,8 evaluate monotherapy in previously treated patients. Given the differences in regimens and line of treatment between the pembrolizumab and toripalimab trials, a non-naïve indirect treatment comparison could not be conducted. Lacking comparative clinical evidence between toripalimab in combination with gemcitabine-cisplatin and pembrolizumab in combination with gemcitabine-cisplatin, a cost-effectiveness analysis could likewise not be conducted. However, we assumed that, in the context of R/M NPC, the pembrolizumab treatment regimen would be in combination with chemotherapy as recommended in the NCCN guidelines¹¹, mirroring the approved first-line treatment regimen of toripalimab, which includes gemcitabine and cisplatin. Hence, pembrolizumab in combination with gemcitabine-cisplatin would be an alternative to toripalimab and economic evaluations are therefore warranted.

• This study is a cost-efficiency analysis from the US payer perspective that considers only the cost of medication and drug administration. Thus, not included are: visit costs, patient co-pays, other patient-related costs (time, travel, meals, lodging, …); and indirect, opportunity, or intangible costs.

• A patient-panel of an estimated 1,200 treated R/M NPC patients in the US in 2023 (Coherus, data on file) was adjusted to 1,207 for 2024 using an annual US population growth rate of 0.59²⁴.

• Costs refer to the sum of medication cost and medication administration cost.

• Cost inputs and outputs are expressed in 2024 US dollars.

• Costs are estimated per cycle and for 1 year (18 cycles) and 2 years (35 cycles) of treatment as both the toripalimab⁶ and pembrolizumab⁷ trial protocols indicate that treatment continues until progression, toxicity, or for up to 2 years. Full treatment duration is assumed for the 1- and 2-year models; that is, no adjustments are made for discontinuation related to toxicity or progression.

• Toripalimab treatment rates ranged from 90% to 45% in decrements of 5%.

• Medication costs include the WAC²⁵ for toripalimab, pembrolizumab, gemcitabine, and cisplatin and the ASP²⁶ for pembrolizumab, gemcitabine, and cisplatin.

• Since ASP data on toripalimab will not be available until several quarters after launch, and since the launch WAC of toripalimab is 80% of the pembrolizumab WAC, an ex ante toripalimab ASP was estimated at 80% of the pembrolizumab ASP.

• Dosage calculations for gemcitabine and cisplatin assume a body surface area of 1.9 m².

Inputs

Medication and administration cost inputs per cycle are listed in Table 1. The WAC inputs for toripalimab, pembrolizumab, gemcitabine, and cisplatin are the costs as published by Merative Micromedex Red Book²⁶ on December 28, 2023. The most currently available ASP data (for 3Q2024) were obtained from the Centers for Medicare and Medicaid Services (CMS) 1Q2024 Medicare Part B Drug payment limit²⁷ by subtracting the 6% CMS add-on. Pembrolizumab (200 mg) costs are $11,115.04/cycle (WAC) and $10,515.09/cycle (ASP). The toripalimab launch WAC of $8,892.03 for a fixed-dose (240 mg), which is 80% of the pembrolizumab WAC for a fixed-dose (200 mg), was utilized. Toripalimab ASP for a fixed-dose (240 mg), set at 80% of pembrolizumab ASP for a fixed-dose (200 mg), is estimated to be $8,412.08.

Table 1. Costs inputs of medications and administration.

Each Q3W Cycle (up to 35 cycles)	Administration^27	WAC^25	ASP^a
Gemcitabine 1,000 mg/m^2 on days 1 and 8 (cycles 1–6)^b	$389.80	$268.05	$65.44
Cisplatin 80 mg/m^2 on day 1 (cycles 1–6)^b	$402.72	$49.35	$57.92
Pembrolizumab 200 mg on day 1	$322.68	$11,115.04	$10,515.09
Toripalimab 240 mg on day 1	$322.68	$8,892.03	$8,412.08^c

Abbreviations: ASP, average sales price; m², squared meters; mg, milligrams; Q3W, every 3 weeks; WAC, wholesale acquisition cost.

^a Computed as payment limit minus add-on. Data from Centers for Medicare and Medicaid Services ²⁶.

^b Assumes a body surface area of 1.9 m².

^c Toripalimab ASP not available; set at 80% of pembrolizumab.

Administration costs were obtained from the 1Q2024 CMS Hospital Outpatient Prospective Payment System Addendum B for the hospital outpatient setting²⁷. Administration costs are the same for pembrolizumab and toripalimab ($322.68 per cycle).

Results

Cost of treatments

Per-patient costs for the pembrolizumab and toripalimab regimens are presented in Table 2 for each cycle in combination with chemotherapy, each maintenance monotherapy cycle, and 1 and 2 years of treatment with all agents. Utilizing the WAC cost input, the per-patient per-cycle treatment cost (including drugs and administration) for one cycle of pembrolizumab plus chemotherapy is $12,548 while the cost per-patient per-cycle for pembrolizumab maintenance monotherapy is $11,438. The costs for a 1-year regimen (18 cycles, all agents) and a 2-year regimen (35 cycles) of pembrolizumab are $212,538 and $406,980, respectively. Comparative costs for toripalimab regimens (including WAC drug cost and administration) are $10,325 per cycle in combination with chemotherapy, $9,215 for one maintenance monotherapy cycle, $172,524 for a full 1-year regimen (all agents) and $329,174 for a 2-year regimen.

Table 2. Per-patient costs and savings of toripalimab over pembrolizumab.

	Per chemo cycle^a	Per maintenance cycle^b	1 year of treatment^c	2 years of treatment^d
	Costs
	Wholesale acquisition cost
Pembrolizumab	$12,548	$11,438	$212,538	$406,980
Toripalimab	$10,325	$9,215	$172,524	$329,174
	Average Sales Price
Pembrolizumab	$11,754	$10,838	$200,575	$384,817
Toripalimab^e	$9,651	$8,735	$162,721	$311,212
	Savings
	Wholesale acquisition cost
Toripalimab vs pembrolizumab	$2,223	$2,223	$40,014	$77,805
	Average Sales Price
Toripalimab vs pembrolizumab	$2,103	$2,103	$37,854	$73,606

All estimates are rounded to the nearest dollar.

^a Chemo cycles are cycles 1–6 and include PD-1 inhibitor + gemcitabine + cisplatin.

^b Maintenance cycles are cycles 7+ and include PD-1 inhibitor monotherapy.

^c A 1-year regimen includes 18 cycles of PD-1 inhibitor with gemcitabine + cisplatin in cycles 1–6 followed by PD-1 inhibitor monotherapy.

^d A 2-year regimen includes 35 cycles of PD-1 inhibitor with gemcitabine + cisplatin in cycles 1–6 followed by PD-1 inhibitor monotherapy.

^e Toripalimab average sales price not available; set at 80% of pembrolizumab average sales price.

Using the ASP input, treatment (including drugs and administration) with pembrolizumab costs $11,754 per-patient per-cycle in combination with chemotherapy, $10,838 for one maintenance monotherapy cycle, $202,575 per patient for a 1-year regimen and $384,817 for a 2-year regimen. Corresponding costs for toripalimab regimens (ex ante ASP drug cost and administration) are $9,651 per chemotherapy cycle and $8,735 per monotherapy maintenance cycle, while a full 1-year regimen costs $162,721 and a 2-year regimen $311,212.

Savings from toripalimab over pembrolizumab

The potential per-patient savings from treating with toripalimab instead of pembrolizumab when using the WAC cost input are $2,223 per cycle for both chemotherapy cycles and maintenance cycles (Table 2). Extrapolated to 1 year and 18 cycles of treatment for one patient with toripalimab (all agents), savings of $40,014 per patient are achievable while savings of $77,805 over 2 years and 35 cycles of treatment are estimated. Savings based on the ex ante ASP model are $2,103 per cycle (either in combination with chemotherapy or maintenance) with toripalimab cost set at 80% of pembrolizumab ASP. This translates into savings of $37,854 per patient over 1 year of treatment and $73,606 for a 2-year regimen.

Panel savings over 1 and 2 years using the two cost inputs and over ten treatment rates (from 90% to 45% in 5% decrements) are presented in Table 3. When between 45% of the 1,207-patient panel (n = 543) and 90% (n = 1,086) are treated with the toripalimab regimen (all agents) instead of pembrolizumab for 1 year, total savings utilizing the WAC are estimated to range from $21,733,702 to $43,467,404. WAC savings over 2 years of treatment range from $42,259,976 (at a treatment rate of 45%) to $84,519,952 (90% treatment rate). In the ex ante ASP analysis at a price point of 80% of pembrolizumab ASP, savings over 1 year of treatment range from $20,560,585 (45% treatment rate) to $41,121,169 (90% treatment rate). Corresponding savings over 2 years range from $39,978,914 to $79,957,829. If only 50% of the panel (n = 604) are treated with toripalimab instead of pembrolizumab for 1 year, estimated savings range from $22,845,094 (80% of ASP) to $24,148,558 (WAC); 2-year savings range from $44,421,016 to $46,955,529, respectively.

Table 3. Total savings from toripalimab over pembrolizumab in a panel of 1,207 patients.

		Wholesale acquisition cost	Average sales price^a
Treatment rate	Patients	1 year of treatment
90%	1,086	$43,467,404	$41,121,169
85%	1,026	$41,052,548	$38,836,660
80%	966	$38,637,692	$36,552,150
75%	905	$36,222,836	$34,267,641
70%	845	$33,807,981	$31,983,132
65%	785	$31,393,125	$29,698,622
60%	724	$28,978,269	$27,414,113
55%	664	$26,563,413	$25,129,603
50%	604	$24,148,558	$22,845,094
45%	543	$21,733,702	$20,560,585
Treatment rate	Patients	2 years of treatment
90%	1,086	$84,519,952	$79,957,829
85%	1,026	$79,824,399	$75,515,727
80%	966	$75,128,846	$71,073,626
75%	905	$70,433,293	$66,631,524
70%	845	$65,737,740	$62,189,422
65%	785	$61,042,187	$57,747,321
60%	724	$56,346,634	$53,305,219
55%	664	$51,651,082	$48,863,118
50%	604	$46,955,529	$44,421,016
45%	543	$42,259,976	$39,978,914

All estimates are rounded to the nearest dollar.

^a Toripalimab average sales price not available; set at 80% of pembrolizumab average sales price.

Number-needed-to-treat

The number of patients needed to be treated (NNT) with toripalimab instead of pembrolizumab to purchase one additional maintenance phase cycle of toripalimab or one additional regimen (all agents for either 1 or 2 years) are listed in Table 4. Utilizing the WAC cost input, the NNT over 1 year to purchase one additional monotherapy maintenance cycle is 0.23 patient; over 2 years, the NNT to purchase one additional maintenance cycle is 0.12 patient. The NNT over 1 year to purchase a full year regimen (all agents) is 4.31 patients and the NNT over 2 years is 4.23 patients.

Table 4. Number of patients needed-to-be-treated (NNT) with toripalimab instead of pembrolizumab to purchase one additional toripalimab maintenance cycle or one additional regimen (all agents) on a budget-neutral basis.

	Wholesale acquisition cost	Average sales price^a
	NNT over 1 year to purchase additional treatment
One additional toripalimab maintenance cycle^b	0.23	0.23
One additional regimen (all agents 1 year)	4.31	4.30
	NNT over 2 years to purchase additional treatment
One additional toripalimab maintenance cycle^b	0.12	0.12
One additional regimen (all agents 2 years)	4.23	4.23

Abbreviation: NNT, number needed to treat.

^a Toripalimab average sales price not available; set at 80% of pembrolizumab average sales price.

^b Maintenance cycle includes toripalimab monotherapy.

Similarly, when using the ex ante ASP price point, the NNT over 1 year to purchase one additional monotherapy maintenance cycle is 0.23 patient; over 2 years, the NNT to purchase one additional maintenance cycle is 0.12 patient. The NNT over 1 year to purchase a full year regimen (all agents) is 4.30 patients while the NNT over 2 years is 4.23 patients.

Expanded access to toripalimab

Savings generated from treating with toripalimab instead of pembrolizumab enable budget-neutral expanded access to additional treatment with toripalimab (Table 5). Using the WAC-based model, the savings from treating a 1,207-patient panel over 1 year permit purchasing between 2,359 additional toripalimab maintenance cycles (at a 45% treatment rate) and 4,717 additional toripalimab maintenance cycles (at a 90% treatment rate) at no additional cost. Or, these same savings can purchase, on a budget-neutral basis, between 126 (45% treatment rate) and 252 (90% treatment rate) full 1-year regimens (all agents). Over 2 years, between 4,586 and 9,172 additional toripalimab maintenance cycles or 128 to 257 full 2-year regimens can be provided at no additional cost, respectively.

Table 5. Expanded access to additional toripalimab that could be purchased on a budget-neutral basis from savings gained from treatment with toripalimab over pembrolizumab in a panel of 1,207 patients.

	1 year of treatment
	Wholesale acquisition cost		Average sales price^a
Treatment rate	Expanded access to additional toripalimab maintenance cycles^b	Expanded access to additional regimens (all agents 1 year)	Expanded access to additional toripalimab maintenance cycles^b	Expanded access to additional regimens (all agents 1 year)
90%	4,717	252	4,708	253
85%	4,455	238	4,446	239
80%	4,193	224	4,185	225
75%	3,931	210	3,923	211
70%	3,669	196	3,662	197
65%	3,407	182	3,400	183
60%	3,145	168	3,139	168
55%	2,883	154	2,877	154
50%	2,621	140	2,615	140
45%	2,359	126	2,354	126
	2 years of treatment
	Wholesale acquisition cost		Average sales price^a
Treatment rate	Expanded access to additional toripalimab maintenance cycles^b	Expanded access to additional regimens (all agents 2 years)	Expanded access to additional toripalimab maintenance cycles^b	Expanded access to additional regimens (all agents 2 years)
90%	9,172	257	9,154	257
85%	8,663	242	8,645	243
80%	8,153	228	8,137	228
75%	7,644	214	7,628	214
70%	7,134	200	7,120	200
65%	6,624	185	6,611	186
60%	6,115	171	6,103	171
55%	5,605	157	5,594	157
50%	5,096	143	5,086	143
45%	4,586	128	4,577	128

^a Toripalimab average sales price not available; set at 80% of pembrolizumab average sales price.

^b Maintenance cycles include toripalimab monotherapy.

Using the ex ante ASP price point, the savings from treating 90% of the panel with toripalimab over 1 year enable purchasing 4,708 additional maintenance cycles of toripalimab. Alternatively, these same savings can provide a full 1-year regimen of toripalimab (all agents) to 253 patients. At a treatment rate of 50%, 2,615 additional maintenance cycles can be purchased; or 140 additional patients can be treated for a full year (all agents) on a budget-neutral basis. Over 2 years, savings enable expanded access to 9,154 additional toripalimab maintenance cycles if 90% of patients are treated with toripalimab instead of pembrolizumab. Alternatively, these same savings provide a full 1-year regimen of toripalimab (all agents) to 257 patients. At a treatment rate of 50%, 5,086 additional maintenance cycles can be purchased; or 143 additional patients can be treated for a full year (all agents) on a budget-neutral basis.

Discussion

Our economic simulation analyses of treating patients with R/M NPC with toripalimab versus pembrolizumab demonstrate that, across scenarios including two cost inputs for toripalimab (launch WAC and an ex ante ASP set at 80% of pembrolizumab ASP) and 10 toripalimab versus pembrolizumab treatment rates (45% to 90%), toripalimab treatment consistently generates cost-efficiencies over pembrolizumab. The ensuing savings can be applied to purchase additional maintenance cycles of toripalimab or full regimens (all agents) and thus provide expanded access to patients with R/M NPC on a budget-neutral basis.

In the simulations using the WAC, per-patient-per-cycle savings with toripalimab amount to $2,223 over pembrolizumab therapy, which translates to $40,014 in savings for 1 year of treatment and $77,805 in savings for 2 years of treatment. Escalating these efficiencies to the 1,207 annual incident patients in the US underscores the large savings generated by treatment with lower-priced toripalimab. For instance, if toripalimab were to attain 90% market share of the eligible R/M NPC patients, savings rounded to the nearest million for 1 year (18 cycles) are $43 million, while over 2 years (35 cycles) savings reach $85 million. At only 50% market share, savings are estimated to be $24 million for 1 year of treatment and $47 million for 2 years.

To determine how savings can be re-allocated to purchasing additional toripalimab treatment, either maintenance cycles or 1- and 2-year regimens with all agents, we estimated the NNT with toripalimab instead of pembrolizumab to purchase one additional maintenance cycle of toripalimab and to purchase full regimens for 1 and 2 years at different price points. For instance, with an NNT over 1 year of 0.23 patient in order to purchase one additional toripalimab monotherapy maintenance cycle, four maintenance cycles can be provided budget neutrally for every R/M NPC patient treated. Further, it takes approximately 4.3 patients to be treated with toripalimab instead of pembrolizumab for 1 year to provide one additional patient with the toripalimab regimen for a full year (all agents) on a budget-neutral basis.

The NNT results accentuate the relative feasibility of providing toripalimab treatment to even more R/M NPC patients. The additional maintenance cycles of toripalimab that can be purchased on a budget-neutral basis from the savings provide expanded access to many additional patients. For instance, if half of the 1,207 incident patients with R/M NPC in 2024 are treated for 1 year with toripalimab priced at the WAC, a total of 2,618 additional maintenance cycles of toripalimab can be purchased or an additional 140 patients can be treated for 1 year with all agents. This increases the number of toripalimab treated patients from 604 to 744, potentially raising the treatment rate of toripalimab from 50.0% to 61.6%.

As the toripalimab with gemcitabine-cisplatin regimen has the potential to assure additional access to treatment, it is important to frame this medical treatment within comprehensive care for patients with R/M NPC. As advocated by De Felice et al.,²⁸ this treatment should be integrated into a multidisciplinary model that is specific to head and neck cancer and assures patient assessment, treatment planning, delivery of services, rehabilitation, and survivorship – all delivered by an all-inclusive team of well-educated and committed health professionals. This team should be supported by robust, remotely accessible, and quality technology that spans the clinical center.

Our analysis has limitations while also pointing to areas of future research. It focuses on the financial impact in terms of medication and administration costs when treating R/M NPC patients with a toripalimab in combination with gemcitabine-cisplatin regimen versus pembrolizumab in combination with gemcitabine-cisplatin, when it is not known whether the manufacturer of the latter may seek approval in the R/M NPC indication in combination with chemotherapy. As noted above, the KEYNOTE-028 and KEYNOTE-122 trials evaluated pembrolizumab monotherapy and reported lower survival outcomes than the JUPITER-02 toripalimab trial. Accordingly, indirect treatment comparisons, including a naïve comparison, could not be performed. This precludes economic evaluations that consider clinical benefit such as cost-effectiveness and cost-utility analyses. To parallel the FDA-approved toripalimab regimen evaluated in the JUPITER-02 trial, our analysis assumed conservatively a combination regimen of pembrolizumab, gemcitabine, and cisplatin which is a NCCN-recommended first-line regimen for R/M NPC. While there may be possible unknown differences in efficacy and safety as well as patient or provider preferences between the toripalimab and pembrolizumab regimens, this economic evaluation demonstrates the cost-efficiency of toripalimab to payers. We used publicly available cost inputs including WAC and ASP, the former reflecting the list price and the latter more reflective of real-world pricing dynamics as ASP is net of any price concessions that are privately negotiated between a manufacturer and purchasers (e.g. volume discounts, prompt pay discounts, cash discounts, free goods, chargebacks, rebates, etc. with the exception of the Medicaid drug rebate program [MDRP] and certain federal discounts and rebates). The ex ante ASP model which estimates the toripalimab ASP to be 80% of the pembrolizumab ASP will need to be replicated in the future with actual toripalimab ASPs, when available and stable, to incorporate the effect of rebates, discounts, and other incentives. Toripalimab treatment rates were static, ranging from 90% to 45% in decrements of 5%. Post launch and over time, the actual market share achieved may be impacted by competitor action, regulatory decisions, and emerging clinical evidence (trial and/or real-world evidence) that could affect treatment adoption rates and overall cost-savings. Other assumptions of this simulated modeling may not reflect real-world treatment patterns post launch. For example, while both the toripalimab and pembrolizumab trial protocols and labels stipulate treatment durations of up to 2 years, actual treatment duration may vary and should be taken into account when interpreting results of potential savings. We simulated only expanded access to the toripalimab regimen, but the efficiencies generated could also be applied to other treatments in other tumor types on a budget-neutral basis. If a trial of a pembrolizumab plus chemotherapy regimen were to be conducted, economic analyses of differential clinical benefit relative to differential cost should be performed. Future research utilizing real-world data of actual toripalimab utilization in clinical practice is required to confirm the exact cost-savings and expanded access that can be achieved in the US health care system.

Conclusion

Using an estimated panel of 1,207 incident NPC patients in 2024, two cost inputs for toripalimab (launch WAC and an ex ante ASP price point of 80% of pembrolizumab ASP) and 10 toripalimab-to-pembrolizumab treatment rates (45% to 90%), treating patients with lower-priced toripalimab in combination with gemcitabine and cisplatin instead of a similar pembrolizumab regimen can generate marked savings. Using the toripalimab launch WAC, if 90% of R/M NPC patients are treated with the toripalimab regimen for 1 year, savings of $43 million can be reallocated to provide access to 4,717 additional toripalimab maintenance cycles or 252 additional 1-year regimens with all agents on a budget-neutral basis. If treated over 2 years, savings of $85 million provide access to 9,172 additional toripalimab maintenance cycles or 257 additional patients could be treated with all agents for 2 years. This simulation demonstrates that toripalimab generates savings that enable the budget-neutral funding for additional toripalimab-gemcitabine-cisplatin regimens for the equivalent of approximately 21% of the 2024 incident cases of R/M NPC in the US.

Transparency

Declaration of funding

This study was sponsored by Coherus BioSciences, Inc., Redwood City, CA, USA.

Declaration of financial/other interests

MP is an employee of Coherus BioSciences, Inc., the study sponsor. KM and IA are owners of Matrix45, which was contracted by Coherus BioSciences, Inc. to conduct this study. By company policy, owners and employees are prohibited from owning equity in sponsor organizations (except through mutual funds or other independently administered collective investment instruments) or contracting independently with client organizations. IA serves on a Speakers Bureau for Coherus BioSciences, Inc., the study sponsor. IA is Editor-in-Chief of the Journal of Medical Economics. He was not involved in any editorial decisions regarding this manuscript.

Author contributions

All authors met ICMJE and COPE criteria and contributed substantively to the study as follows: Study concept: KM, IA; Study design: KM, IA; Model development and simulations: KM, IA; Review and interpretation of results: KM, IA, MP; Development of manuscript: KM, IA, MP; Review of the manuscript for intellectual content: KM, IA, MP.

Reviewer disclosures

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Previous presentations

An earlier version of this analysis was presented as a poster at AMCP-Nexus 2022 and published as an abstract (MacDonald K, Walden P, Geller R, Abraham I. Cost-efficiency and budget-neutral expanded access modeling of PD-1 inhibitor toripalimab plus gemcitabine-cisplatin over a simulated parallel pembrolizumab regimen for treatment of recurrent or metastatic naso-pharyngeal cancer. J Manag Care Spec Pharm 2022;28(Suppl 10a):S14.

Notes

i Hereafter referred to as toripalimab.