Cost-effectiveness analysis of Cadonilimab in first-line treatment of advanced HER2-negative gastric cancer or gastroesophageal junction cancer

Abstract

Background:

Immune checkpoint inhibitors (ICIs), including Pembrolizumab, Nivolumab, Sintilimab, Tislelizumab, and Sugemalimab, have been approved in China as first-line treatments for advanced HER2-negative gastric cancer (GC) and gastroesophageal junction cancer (GEJC). However, the latest COMPASSION 15 study showed that Cadonilimab in combination with chemotherapy provided significant survival benefits.

Objective:

This study aims to evaluate the cost-effectiveness of Cadonilimab plus chemotherapy versus chemotherapy alone and the ICIs approved in China for first-line treatment of advanced HER2-negative gastric cancer or gastroesophageal junction carcinoma (GC/GEJC) from the perspective of Chinese payers.

Design:

The cost-effectiveness analysis.

Methods:

Based on the research data from COMPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303, we constructed a 15-year Markov model to evaluate the cost and health outcomes of Cadonilimab combined with chemotherapy versus chemotherapy alone and other ICIs in advanced HER2-negative GC and GEJC. This evaluation includes total cost, life years (LYs), quality-adjusted life years (QALYs), and incremental cost-effectiveness ratio (ICER).

Results:

Cadonilimab produced 0.73 QALYs (1.10 LYs) at a cost of $26,591. It required an additional investment of $17,826 to gain 0.25 QALYs (0.37 LYs), resulting in an ICER of $72,492.29 per QALY compared to chemotherapy alone. In comparison, other ICIs approved in China—Pembrolizumab, Nivolumab, Sintilimab, Tislelizumab, and Sugemalimab—incurred total costs of $11,735, $13,970, $16,346, $10,765, and $14,857, respectively, generating 0.68 QALYs (1.04 LYs), 0.69 QALYs (1.04 LYs), 0.73 QALYs (1.12 LYs), 0.82 QALYs (1.26 LYs), and 0.81 QALYs (1.25 LYs). Sensitivity analysis revealed that the cost of Cadonilimab, the utility value of progressive disease, and the risk of platelet decline in the Cadonilimab group were the most influential factors affecting the model’s stability. At a willingness-to-pay threshold of $37,386, Cadonilimab is not a cost-effective option for the first-line treatment of advanced GE/GEJC.

Conclusion:

Cadonilimab is not a cost-effective option for the first-line treatment of advanced HER2-negative GC/GEJC. In comparison to other ICIs approved in China, Tislelizumab appears to be a more favorable option.

Keywords

advanced HER2-negative gastric cancer Cadonilimab cost-effectiveness analysis immune checkpoint inhibitors Markov model

Introduction

Gastric cancer (GC) is the fifth most diagnosed cancer and the fifth leading cause of cancer-related deaths globally. According to the International Agency for Research on Cancer, there were an estimated 968,350 new cases and 659,853 deaths due to gastric cancer worldwide in 2022.¹ In China, gastric cancer ranks as the fifth most frequently diagnosed cancer and is the third leading cause of cancer-related deaths. In 2022, there were approximately 358,700 new cases of gastric cancer and about 260,400 associated deaths in China.² Although advances in endoscopic technology have improved the detection rate of early gastric cancer, 36.4% of patients are still diagnosed with advanced stages of the disease, resulting in a 5-year survival rate of only 7%.³

Over the past few decades, the treatment of advanced gastric cancer has remained a significant challenge. The primary management approach has been chemotherapy, which offers limited improvements in overall survival (OS) and presents difficulties in managing chemotherapy-induced toxicity and side effects.^4–7 Recently, targeted therapy has demonstrated efficacy in advanced gastric cancer. For instance, Trastuzumab, a monoclonal antibody targeting HER2, has been approved for first-line treatment of HER2-positive advanced gastric cancer,^8–10 However, its indications still have limitations. With the emergence of immune checkpoint inhibitors (ICIs), the combination of programmed cell death protein 1 (PD-1) inhibitors and chemotherapy has become the standard first-line treatment for advanced gastric cancer (GC) and gastroesophageal junction cancer (GEJC). A phase III clinical trial, KEYNOTE-859 (NCT03675737), demonstrated a median OS of 12.9 months for patients receiving Pembrolizumab and 11.5 months for those receiving placebo in combination with the XELOX regimen (HR = 0.78, p < 0.0001).¹¹ Four-year follow-up data from the phase III clinical trial CHECKMATE-649¹² (NCT02872116) demonstrated that patients with a programmed cell death-ligand 1 combined positive score (PD-L1 CPS) ⩾5 who received Nivolumab had a median OS of 15.5 months (95% CI: 11.9–21.1 months), compared to 9.6 months (95% CI: 8.0–12.1 months) in the control group.¹³ Based on the efficacy data mentioned above, both the 2024 NCCN guidelines and the CSCO guidelines recommend Pembrolizumab and Nivolumab as the first-line treatment for advanced HER2-negative GC and GEJC.^8,9 Furthermore, phase III clinical trials ORIENT-16 (NCT03745170), RATIONALE-305 (NCT03777657), and GEMSTONE-303 (NCT03802591), Sintilimab, Tislelizumab, and Sugemalimab have also demonstrated significant OS and progression-free survival (PFS) benefits.^14–16 Therefore, these agents have been successively approved in China for first-line treatment of advanced HER2-negative gastric cancer or gastroesophageal junction adenocarcinoma (GC/GEJC).⁹

In 2024, the American Association for Cancer Research (AACR) General Assembly announced the results of the phase III clinical study COMPASSION-15 (NCT05008783). This study examined the efficacy of Cadonilimab, a bispecific antibody targeting PD-1 and cytotoxic T lymphocyte antigen-4 (CTLA-4), in combination with a chemotherapy regimen for the first-line treatment of advanced gastric or gastroesophageal junction adenocarcinoma. The results indicated that the combination significantly prolonged OS and PFS in the overall population. OS was extended to 15.0 months compared to 10.8 months (HR = 0.62; 95% CI, 0.50–0.78; p < 0.001), and PFS was extended to 7.0 months compared to 5.3 months (HR = 0.53; 95% CI, 0.44–0.65; p < 0.001). The benefits were even more pronounced in patients with PD-L1 CPS ⩾5 scores, resulting in an OS of not estimable compared to 10.6 months (HR = 0.56; 95% CI, 0.39–0.80; p < 0.001) and a PFS of 6.9 months compared to 5.5 months (HR = 0.51; 95% CI, 0.37–0.70; p < 0.001). In the PD-L1 CPS <5 group, the combination of Cadonilimab significantly prolonged OS to 14.8 months compared to 11.1 months (HR = 0.70; 95% CI, 0.51–0.95; p = 0.011) and PFS to 6.9 months compared to 4.6 months (HR = 0.60; 95% CI, 0.45–0.79; p < 0.001).¹⁷

Although Cadonilimab has demonstrated significant efficacy and safety in advanced GC/GEJC, its economic benefits have not been evaluated. This study aims to evaluate the cost-effectiveness of Cadonilimab combined with chemotherapy, compared to chemotherapy alone and other ICIs approved in China for first-line treatment of advanced HER2-negative GC/GEJC, from the perspective of Chinese payers.

Materials and methods

This study is a cost-effectiveness analysis based on a 15-year Markov model. The inclusion criteria are as follows: (1) phase III Randomized Controlled trials; (2) histologically confirmed advanced or metastatic GC/GEJC; (3) first-line treatment; (4) HER2-negative status; (5) ICIs; and (6) approval by the Chinese CSCO guidelines. Eligible studies include COPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303. In addition, this survey and report fully comply with the CHEERS guidelines¹⁸ (Table S1).

Patient population and intervention

In this study, we used patient cohorts and treatment data from the COPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303 trials. In the ORIENT-16, RATIONALE-305, and GEMSTONE-303 trials, the anti-PD-1 antibodies, Sintilimab (200 mg every 3 weeks), Tislelizumab (200 mg every 3 weeks), and Sugemalimab (1200 mg every 3 weeks), were combined with chemotherapy regimens: XELOX (capecitabine 1000 mg/m² twice daily on days 1–14 and oxaliplatin 130 mg/m² on day 1), FP (5-fluorouracil 800 mg/m² on days 1–5 and cisplatin 80 mg/m²), or CAPEOX (oral capecitabine 1000 mg/m² twice daily on days 1–14 and oxaliplatin 130 mg/m² on day 1). For simplicity, the model uniformly adopted the XELOX regimen. After six cycles, maintenance therapy continued with Cadonilimab or placebo (in the Cadonilimab trial) or with Sintilimab, Tislelizumab, or Sugemalimab (in the ORIENT-16, RATIONALE-305, and GEMSTONE-303 trials) until disease progression or unacceptable toxicity, for a maximum of 2 years. In the KEYNOTE-859 and CheckMate-649 trials, patients received Pembrolizumab (200 mg every 3 weeks) or Nivolumab (360 mg every 3 weeks) plus the XELOX chemotherapy regimen, with treatment capped at 2 years. All patients were hospitalized every 3 weeks, and imaging assessments were performed every 6 weeks to monitor disease progression. Since the original article did not specify a second-line treatment regimen, we recommended one based on the 2024 NCCN and CSCO guidelines. Docetaxel (75–100 mg/m² intravenously every 3 weeks) was used as second-line treatment for patients with disease progression. The patient population was assumed to be male, 64 years old, weighing 65 kg, with a body surface area of 1.72 m², to calculate the drug dose.¹⁹

Model structure

In this study, we used TreeAgePro2022 software (TreeAge Software, Williamstown, MA, USA, https://www.treeage.com) to build a decision tree and a Markov model simulating patients with advanced gastric cancer treated with one of seven regimens (Cadonilimab + XELOX, placebo + XELOX, Pembrolizumab + XELOX, Nivolumab + XELOX, Sintilimab + XELOX, Tislelizumab + XELOX, or Sugemalimab + XELOX). Patients transitioned among three health states—PFS, progressive disease (PD), and death—with initial probabilities of 1, 0, and 0, respectively, to evaluate health and economic outcomes (Figure S1). All patients will receive 35 cycles of treatment and be followed regularly, unless disease progression or severe adverse reactions occur. If the disease progresses, second-line treatment will be administered, while the remaining patients will receive optimal supportive care. In conjunction with the patient medication regimen, we defined the model cycle as a 3-week cycle over a 15-year period (with more than 99% of patients dying) and applied a 5% discount rate^20,21 to predict the total cost, life years (LYs), quality-adjusted life years (QALYs), and incremental cost-effectiveness ratio (ICER) of the treatment strategies. We also evaluated the cost-effectiveness based on the willingness-to-pay (WTP) value. Relevant literature indicates that the WTP value is typically calculated as three times per capita GDP for China,^22,23 estimated at $37,386/QALY for 2024.

Survival estimates and model transitions

We used Getdata Graph Digitizer (version 2.20, available at: http://www.getdata-graph-digitizer.com/index.php) to extract survival data from the OS and PFS curves of the COMPASSION-15 KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303 trials. Patient data were then reconstructed using R Studio software (version 4.3.2, available at: https://www.rstudio.com). The log-normal, log-logistic, Weibull, exponential, and Gompertz distributions were selected to calculate the parameters described earlier. The best-fitting parameter distribution curves for the PFS and OS groups were selected based on the Akaike information criterion and Bayesian information criterion (Figure S2 and Table S2). The most suitable Weibull was chosen to fit the data, and the values of the two parameters, scale parameter (λ) and shape parameter (γ), were calculated using the R software. Assuming that the mortality rate during period t is denoted by M, and the mean OS time denoted as S(t), then the cause-specific mortality M at cycle t can be expressed as follows: M = [S(t) – S(t + 1)]/S(t). Based on the fitted parameter, S(t) can be further formulated as follows: exp $(- λ t^{γ})$ $(λ$ >0, $γ$ > >0), the OS rate for each final period is 1 – exp $(λ {[t - u]}^{γ} - λ t^{γ})$ $(λ$ >0, $γ$ >0) where $t$ and $u$ represent different Markov periods. And the PFS rate was estimated by the same methods.

Cost input and utility estimate

The medical costs included in this model encompass drug costs, laboratory and imaging examination costs, hospitalization management costs, adverse reaction management costs, optimal supportive treatment costs, and palliative care costs (Table 1). The price—Cadonilimab, Pembrolizumab, Nivolumab, Sintilimab, Tislelizumab, Sugemalimab—is based on the official list price, while the prices of other drugs are derived from the official pricing at Xiangya Hospital of Central South University. Other costs were obtained from published literature^21,23–26 and were included in the model calculation only when adverse reactions were ⩾grade 3 and their incidence was ⩾5% (all adverse reactions were assumed to occur once within the PFS and PD states). These prices have been adjusted for inflation to reflect current values using the Bureau of Labor Statistics Inflation Calculator (https://www.bls.gov/data/inflation_calculator.htm?ss=P&st_rid=null). All prices are expressed in US dollars (as of March 7, 2025, USD 1 = CNY 7.1705).

Table 1.

Model parameters: baseline values, ranges, and distributions for sensitivity analysis.

Variable	Baseline value	Range		Reference	Distribution
Variable	Baseline value	Minimum	Maximum	Reference	Distribution
Weibull survival model of OS of CX	Scale = 0.027008 Shape = 1.187599	—	—	17	—
Weibull survival model of OS of PX	Scale = 0.019516 Shape = 1.496296
Weibull survival model of PFS of CX	Scale = 0.063116 Shape = 1.123983	—	—	17	—
Weibull survival model of PFS of PX	Scale = 0.056947 Shape = 1.500360
Weibull survival model of OS of PC	Scale = 0.053841 Shape = 1.000616	—	—	11	—
Weibull survival model of PFS of PC	Scale = 0.150120 Shape = 0.781870	—	—	11	—
Weibull survival model of OS of NC	Scale = 0.032500 Shape = 1.152341	—	—	12	—
Weibull survival model of PFS of NC	Scale = 0.072638 Shape = 1.051075	—	—	12	—
Weibull survival model of OS of SNC	Scale = 0.025073 Shape = 1.194991	—	—	15	—
Weibull survival model of PFS of SNC	Scale = 0.098910 Shape = 0.890990	—	—	15	—
Weibull survival model of OS of TC	Scale = 0.041050 Shape = 1.018916	—	—	14	—
Weibull survival model of PFS of TC	Scale = 0.152450 Shape = 0.751340	—	—	14	—
Weibull survival model of OS of SC	Scale = 0.033946 Shape = 1.072142	—	—	16	—
Weibull survival model of PFS of SC	Scale = 0.070440 Shape = 1.101330	—	—	16	—
Risk for main adverse events in the CX group
Risk of decreased platelet count	0.285	0.228	0.342	17	Beta
Risk of decreased neutrophil count	0.151	0.121	0.181	17	Beta
Risk of decreased white blood cell count	0.072	0.058	0.086	17	Beta
Risk of anemia	0.102	0.082	0.122	17	Beta
Risk of hypokalemia	0.059	0.047	0.071	17	Beta
Risk for main adverse events in the PX group
Risk of decreased platelet count	0.250	0.200	0.300	17	Beta
Risk of decreased neutrophil count	0.148	0.118	0.178	17	Beta
Risk of decreased white blood cell count	0.063	0.050	0.076	17	Beta
Risk of anemia	0.125	0.100	0.150	17	Beta
Utility
Utility PFS in first-line treatment	0.797	0.638	0.956	30	Beta
Utility PD in first-line treatment	0.577	0.462	0.692	30	Beta
Disutility
Decreased platelet count	0.023	0.018	0.028	31	Beta
Decreased neutrophil count	0.200	0.160	0.240	31	Beta
Decreased white blood cell count	0.200	0.160	0.240	31	Beta
Anemia	0.020	0.016	0.024	31	Beta
Hypokalemia	0.008	0.0064	0.0096	25	Beta
Diarrhea	0.05	0.4	0.6	30	Beta
Neutropenia	0.07	0.056	0.084	30	Beta
Fatigue	0.09	0.072	0.108	30	Beta
Vomit	0.048	0.0384	0.576	21	Beta
Lipase increase	0.47	0.376	0.564	32	Beta
Drug cost, $/per cycle
Cadonilimab	1556.38	1245.104	1867.656	Official price	Gamma
Oxaliplatin	165.12	132.069	198.144	Local charge	Gamma
Capecitabine	21.66	17.328	25.992	Local charge	Gamma
Docetaxel	28.37	22.696	34.044	Local charge	Gamma
Sugemalimab	493.09	394.472	591.708	Official price	Gamma
Nivolumab	881.80	705.44	1058.16	Official price	Gamma
Pembrolizumab	571.16	456.928	685.392	Official price	Gamma
Sintilimab	301.23	240.984	361.476	Official price	Gamma
Tislelizumab	349.63	279.704	419.556	Official price	Gamma
Expenditures on main adverse events, $
Decreased platelet count	3735.73	2988.584	4482.876	23	Gamma
Decreased neutrophil count	490.15	392.12	588.18	23	Gamma
Decreased white blood cell count	490.15	392.12	588.18	23	Gamma
Anemia	576.3	461.04	691.56	24	Gamma
Hypokalemia	16.93	13.54	20.136	24	Gamma
Diarrhea	410.03	328.024	492.036	25	Gamma
Neutropenia	657.16	525.728	788.592	25	Gamma
Fatigue	3	2.4	3.6	21	Gamma
Vomit	54.62	43.696	65.544	21	Gamma
Lipase increased	0	0	0	—
Imaging and laboratory per cycle	120.79	96.6321	144.948	22	Gamma
Administration per cycle	63.79	51.032	76.548	22	Gamma
Best supportive care	129.68	103.744	155.616	22	Gamma
Terminal care	768.14	614.512	921.768	22	Gamma
Discount rate	0.05	—	—	20, 21	—

AEs, adverse events; CX, Cadonilimab + XELOX; HR, hazard ratio; NC, Nivolumab plus chemotherapy; OS, overall survival; PC, Pembrolizumab plus chemotherapy; PFS, progression-free survival; PX, Placebo + XELOX; SC, Sugemalimab plus chemotherapy; SNC, Sintilimab plus chemotherapy; TC, Tislelizumab plus chemotherapy; —, unreported.

Utility values are used to assess a patient’s health-related quality of life (HRQOL), ranging from 0 (representing death) to 1 (indicating perfect health). Since the COPASSION15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303 studies did not provide measures like EQ-5D, SF-36, or other relevant HRQOL scales,^27–29 we referred to existing literature to gather the required data for our calculations.³⁰ We also included disutility values related to adverse reactions in our analysis.^21,31–33

Sensitivity analysis

Sensitivity analysis was conducted to assess model stability. In the one-way sensitivity analysis, a ±20% variation range was applied to this parameter, and the method described in the literature was used to assess its influence on the ICER. In the probabilistic sensitivity analysis, we conducted 10,00 Monte Carlo simulations on a population of 610 patients to assess the cost-effectiveness of Cadonilimab in combination with XELOX at the $37,386 WTP threshold. In addition, in subgroup analysis, we assessed the cost-effectiveness of different PD-L1 expression levels (PD-L1 CPS <5, PD-L1 CPS ⩾5) under the WTP threshold using survival curves. In other subgroup analyses, because complete Kaplan–Meier curves for PFS or OS were unavailable, we estimated the parameters using the following formulas:

γ_{Cadonilimab} = γ_{Placebo}, λ_{Cadonilimab} = λ_{Placebo} \times HR

^23,34 to calculate cost-effectiveness.

Results

Base case results

The Cadonilimab + XELOX regimen yielded 0.73 QALYs (1.10 life years (LYs)) at a total cost of $26,591, compared with the placebo + XELOX regimen, which yielded 0.48 QALYs (0.73 life years) at $8765. Cadonilimab increased costs by $17,826 for a gain of 0.25 QALYs, resulting in an ICER of $72,492.29 per QALY. At a WTP threshold of $37,386 per QALY, the regimens produced the following outcomes: Pembrolizumab + XELOX: 0.68 QALYs (1.04 LYs) at $11,735; Nivolumab + XELOX: 0.69 QALYs (1.04 LYs) at $13,970; Sintilimab + XELOX: 0.73 QALYs (1.12 LYs) at $16,346; Tislelizumab + XELOX: 0.82 QALYs (1.26 LYs) at $10,765; and Sugemalimab + XELOX: 0.81 QALYs (1.25 LYs) at $14,857; Overall, the Cadonilimab + XELOX regimen did not meet the cost-effectiveness threshold for first-line treatment of advanced gastric cancer (Table 2).

Table 2.

Baseline results.

Parameters	LYs	QALYs	Total cost $	ICER $/LY^a	ICER $/QALY^b	WTP $/QALY
All patients
Cadonilimab plus XELOX	1.10	0.73	26,591	47,495.05	72,492.29	37,386
Placebo plus XELOX	0.73	0.48	8765	—	—	—
PD-L1 CPS ⩾5%
Cadonilimab plus XELOX	1.43	0.93	29,274	28,995.59	45,928.89	37,386
Placebo plus XELOX	0.72	0.48	8686	—	—	—
PD-L1 CPS <5%
Cadonilimab plus XELOX	1.07	0.70	25,587	56,921.74	86,205.72	37,386
Placebo plus XELOX	0.78	0.51	8974	—	—	37,386
Sugemalimab plus chemotherapy	1.25	0.81	14857	NA	NA	37,386
Nivolumab plus chemotherapy	1.04	0.69	13970	NA	NA	37,386
Pembrolizumab plus chemotherapy	1.04	0.68	11735	NA	NA	37,386
Sintilimab plus chemotherapy	1.12	0.73	16346	NA	NA	37,386
Tislelizumab plus chemotherapy	1.26	0.82	10765	NA	NA	37,386

Compared to Placebo plus XELOX ($/LY).

Compared to Placebo plus XELOX ($/QALY).

ICER, incremental cost-effectiveness ratio; LY, life-year; NA, not applicable; QALY, quality-adjusted life-year; WTP, willingness-to-pay; XELOX, Capecitabine 1000 mg/m², twice daily, orally from days 1 to 14, every 3 weeks; Oxaliplatin 130 mg/m², intravenously every 3 weeks.

Sensitivity analysis results

One-way sensitivity analysis revealed that the cost of Cadonilimab, the utility value associated with disease progression, and the risk of Cadonilimab-induced thrombocytopenia were key factors affecting model stability (Figure 1). Among these variables, the cost of Cadonilimab was the most influential. As the per-cycle cost of Cadonilimab ranged from $1245 to $1867, the ICER varied from $62,219.72/QALY to $82,764.37/QALY. The per-cycle costs of oxaliplatin, docetaxel, and hypokalemia management had minimal impact on model outcomes. Furthermore, variations in ICER values within their respective ranges had negligible effects on the overall model outcomes. In the PD-L1 CPS ⩾5% subgroup, the most influential factors affecting model stability were the utility value of PD, the cost of Cadonilimab, and the risk of platelet decline in the Cadonilimab group (Figure 2). In the PD-L1 CPS <5% subgroup, the cost of Cadonilimab, the risk of platelet declines in the Cadonilimab group, and the utility value of PFS were the most influential factors affecting model stability (Figure 3). During a comprehensive comparison of multiple clinical trials, such as COMPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303, we conducted a sensitivity analysis to assess the net monetary benefits (NMB). The results reveal that the utility value of PD, the utility value of PFS, and the cost associated with decreased platelet count are the three most influential factors affecting model stability. Conversely, the costs of Cadonilimab, Sugemalimab, Nivolumab, and Pembrolizumab have minimal impact on the robustness of the model (Figure 4).

Figure 1.

The one-way sensitivity analyses for Cadonilimab plus chemotherapy (all patients).

Figure 2.

The one-way sensitivity analyses for Cadonilimab plus chemotherapy (PD-L1 CPS ⩾5%).

Figure 3.

The one-way sensitivity analyses for Cadonilimab plus chemotherapy (PD-L1 CPS <5%).

Figure 4.

The one-way sensitivity analyses for all groups (COPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303).

In the COMPASSION-15 trial, probability sensitivity analysis showed that, at a WTP threshold of $37,386 per QALY, the probability of Cadonilimab in combination with XELOX being cost-effective for the Chinese population was 27% (Figure S3). The cost-effectiveness acceptability curve further suggested that chemotherapy alone might be a more cost-effective treatment option (Figure 5). In comparison with other ICIs approved in China, the acceptability curve indicates that Tislelizumab is a relatively better option for first-line treatment of advanced gastric cancer (Figure 6).

Figure 5.

Acceptable curve plots (COMPASSION-15).

Figure 6.

Acceptable curve plots (COMPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303).

Subgroup analysis

In the subgroup with PD-L1 CPS levels, the ICER exceeds the WTP threshold. When PD-L1 CPS ⩾5%, Cadonilimab + XELOX and placebo + XELOX produced 0.93 QALYs (1.43 LYs) and 0.48 QALYs (0.72 LYs), respectively. The ICER was $45,928.89 per QALY ($28,995.59 per LY). When PD-L1 CPS <5%, Cadonilimab + XELOX and placebo + XELOX produced 0.70 QALYs (1.07 LYs) and 0.51 QALYs (0.78 LYs), respectively, with ICERs of $86,205.12 per QALY ($56,921.74 per LY) (Table 2). This indicates that Cadonilimab was not a cost-effective option regardless of PD-L1 CPS levels (Figure 5), a conclusion also supported in other subgroups (Table S3).

Discussion

Gastric cancer is one of the most common malignancies in China. Due to the large population, the number of gastric cancer cases in China accounts for nearly half of the global total.³⁵ Statistical data indicate that approximately 35.1% of patients survive more than 5 years after diagnosis,³⁶ showing that gastric cancer remains a major health threat in China. Furthermore, the economic burden of gastric cancer treatment is significant. In 2017, total cancer treatment expenditure in China was 304.8 billion yuan, with cancer-related hospitalization costs reaching 241.9 billion yuan. Gastric cancer ranked among the top three cancers for hospitalization costs from 2005 to 2017.³⁷ In tertiary hospitals, gastric cancer had the highest inpatient costs compared to other cancers.³⁵ Therefore, when a treatment program provides significant benefits, evaluating its economic impact is crucial to help patients and clinicians make informed clinical decisions.

The COMPASSION-15 study showed that the combination of Cadonilimab and XELOX chemotherapy improved OS in the entire population, with more pronounced OS benefits when PD-L1 CPS ⩾5%. However, this study used a Markov model to analyze cost-effectiveness and concluded that Cadonilimab was not cost-effective. Subgroup analysis further showed that for patients with PD-L1 CPS ⩾5%, the treatment regimen was also not cost-effective. Furthermore, the probability of Cadonilimab being cost-effective at a WTP threshold of $37,386 was 43.7%. When PD-L1 CPS <5%, the probability of Cadonilimab being cost-effective at the WTP threshold of $37,386 dropped to 22.6%. Sensitivity analysis showed that the cost of Cadonilimab per cycle was the most influential factor in the entire population affecting the model. When the cost of Cadonilimab was reduced by 70%, the total cost and QALYs generated by the Cadonilimab + XELOX chemotherapy regimen were $17,750 and 0.73 QALYs, respectively. Compared to the placebo + XELOX chemotherapy regimen, the total cost increased by $8985, while generating an additional 0.25 QALYs. The resulting ICER was $36,538.97 per QALY. At a WTP threshold of $37,386, Cadonilimab is a cost-effective option for first-line treatment of advanced gastric cancer. In patients with PD-L1 CPS ⩾5%, when the cost of Cadonilimab per cycle is reduced by 30%, the Cadonilimab + XELOX regimen results in a total cost of $25,196 and generates 0.93 QALYs. Compared to the alternative treatment, this regimen incurs an additional cost of $16,510 and provides an incremental gain of 0.45 QALYs. The resulting ICER is $36,831.90 per QALY, indicating that Cadonilimab is cost-effective at this threshold.

During the sensitivity analyses of the clinical trials COMPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and GEMSTONE-303, a more appropriate approach involved introducing NMB as a key metric. Traditionally, cost-effectiveness ratio sensitivity analyses involve pairwise comparisons. When multiple intervention regimens are considered, selecting a common reference group, typically the standard treatment, is necessary. Subsequently, the ICERs of other regimens relative to this reference are calculated individually, followed by the sensitivity analysis. This method restricts the clear illustration of the comparative performance of all regimens across various WTP thresholds in a single graphical representation. Transforming the analysis through the formula: NMB = Effectiveness * λ – Cost (where λ represents WTP) shifts it from a ratio-based to a linear framework. For any given WTP threshold, the NMB of each treatment strategy can be directly calculated, identifying the strategy with the highest NMB as the most cost-effective at that threshold. Among the parameters influencing NMB, the utility values for PD and PFS, along with the costs associated with decreased platelet count, are the three most influential factors affecting model stability. Specifically, higher utility values for both PD and PFS enhance cost-effectiveness. When the utility value of PD ranges from 0.4616 to 0.6924, the corresponding NMB varies from $16,511 to $23,364. Similarly, when the utility value of PFS ranges from 0.6376 to 0.9564, the corresponding NMB varies from $17,014 to $22,861. In addition, lower costs associated with decreased platelet count enhance cost-effectiveness. When the costs associated with a decline in platelet count range from $2988 to $4482, the corresponding NMB varies from $19,252 to $20,623. By contrast, the costs of Cadonilimab, Sugemalimab, Nivolumab, Pembrolizumab, and Sintilimab had minimal impact on the results of the sensitivity analysis. This is because these treatments were not identified as cost-effective options compared to Tislelizumab in the model.

PD-1/PD-L1 antibodies are widely used in the first-line treatment of gastric cancer, supported by data from phase III clinical trials such as KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and Gemstone-303. Consequently, Pembrolizumab, Nivolumab, Sintilimab, Tislelizumab, and Sugemalimab have been successively approved by the Chinese CSCO guidelines⁹ for the first-line treatment of HER2-negative advanced GC/GEJC. In this study, we evaluated the cost-effectiveness of Cadonilimab compared to other ICIs. Our results indicate that Cadonilimab is not cost-effective compared with other ICIs. Tislelizumab emerges as the most cost-effective option. We also performed a linear analysis of incremental NMB across various WTP thresholds, which further confirmed this conclusion (Figure 7). Pembrolizumab is more cost-effective than Nivolumab, consistent with the findings of Zhu et al.²³ However, Chinese studies^30,38,39 have found neither Nivolumab nor Pembrolizumab to be cost-effective, which does not contradict our results. Our analysis compares Pembrolizumab primarily with Cadonilimab, not with chemotherapy alone as in the KEYNOTE-859 and CheckMate-649 trials. Likewise, the reported cost-effectiveness of Sintilimab in China⁴⁰ aligns with our study’s conclusions.

Figure 7.

Linear analysis chart of NMB.

This study has several limitations: First, the data are derived from the COMPASSION-15, KEYNOTE-859, CheckMate-649, ORIENT-16, RATIONALE-305, and Gemstone-303 trials, and any biases in those trials may influence our findings. Second, because these trials did not report health utility values by treatment group, we used utility estimates from external studies, which may introduce bias. Third, for adverse reaction cost estimation, we excluded grade 1–2 events with an incidence below 5%, affecting direct medical costs. However, one-way sensitivity analyses showed that this exclusion did not compromise model stability. Fourth, in the absence of head-to-head trials, we relied on indirect comparisons to evaluate the drugs’ cost-effectiveness. This approach may introduce heterogeneity due to differences in trial cohorts. However, the enrolled populations share similar baseline characteristics, which mitigates this heterogeneity. Sixth, this study is based on data from China’s healthcare system and evaluates the cost-effectiveness of Cadonilimab as a first-line immunotherapy for advanced GC and GEJC from the perspective of Chinese healthcare payers. Consequently, the findings and implications of this study are specifically applicable within the context of China’s healthcare system. Finally, clinical guidelines recommend multiple second-line therapy options after disease progression. In our analysis, we assumed all patients received docetaxel, which may not reflect real-world practice where clinicians individualize treatment based on each patient’s condition. However, one-way sensitivity analysis showed that variations in second-line treatment costs had a negligible effect on the ICER.

Conclusion

In conclusion, from the perspective of Chinese payers, Cadonilimab is not cost-effective compared to chemotherapy alone or to other ICIs approved in China, whereas Tislelizumab represents the most cost-effective option among all treatments evaluated.

Supplemental Material

sj-docx-1-tag-10.1177_17562848251381143 – Supplemental material for Cost-effectiveness analysis of Cadonilimab in first-line treatment of advanced HER2-negative gastric cancer or gastroesophageal junction cancer

Supplemental material, sj-docx-1-tag-10.1177_17562848251381143 for Cost-effectiveness analysis of Cadonilimab in first-line treatment of advanced HER2-negative gastric cancer or gastroesophageal junction cancer by Kaiqi Zhu, Mengyao Qin, Zhaoyi Pan and Jin Huang in Therapeutic Advances in Gastroenterology

Footnotes

Acknowledgements

None.

Declarations

ORCID iD

Kaiqi Zhu

Supplemental material

Supplemental material for this article is available online.

References

Bray

Laversanne

Sung

, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024; 74: 229–263.

Zheng

Chen

Han

, et al. Cancer incidence and mortality in China, 2022. Zhonghua Zhong Liu Za Zhi 2024; 46: 221–231.

NCI. Cancer stat facts: stomach cancer, https://seer.cancer.gov/statfacts/html/stomach.html (2025).

Al-Batran

Hartmann

Probst

, et al. Phase III trial in metastatic gastroesophageal adenocarcinoma with fluorouracil, leucovorin plus either oxaliplatin or cisplatin: a study of the Arbeitsgemeinschaft Internistische Onkologie. J Clin Oncol 2008; 26: 1435–1442.

Kang

Shin

, et al. Capecitabine/cisplatin versus 5-fluorouracil/cisplatin as first-line therapy in patients with advanced gastric cancer: a randomised phase III noninferiority trial. Ann Oncol 2009; 20: 666–673.

Luo

Wang

, et al. Phase II trial of XELOX as first-line treatment for patients with advanced gastric cancer. Chemotherapy 2010; 56: 94–100.

Shah

Janjigian

Stoller

, et al. Randomized multicenter phase II study of modified Docetaxel, Cisplatin, and Fluorouracil (DCF) versus DCF plus growth factor support in patients with metastatic gastric adenocarcinoma: a study of the US gastric cancer consortium. J Clin Oncol 2015; 33: 3874–3879.

National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology(NCCN Guidelines®): Gastric Cancer Version 1, https://www.nccn.org/professionals/physician_gls/pdf/gastric_basic.pdf (2024, accessed 7 March 2024).

CSCO. Guidelines of Chinese Society of Clinical Oncology (CSCO): gastric cancer, http://www.csco.org.cn/cn/index.aspx (2024).

10.

Bang

Van Cutsem

Feyereislova

, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet 2010; 376: 687–697.

11.

Rha

Yanez

, et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for HER2-negative advanced gastric cancer (KEYNOTE-859): a multicentre, randomised, double-blind, phase 3 trial. Lancet Oncol 2023; 24: 1181–1195.

12.

Janjigian

Shitara

Moehler

, et al. First-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trial. Lancet 2021; 398: 27–40.

13.

Shen

Bai

Lin

, et al. First-line (1L) nivolumab (NIVO) plus chemotherapy (chemo) vs chemo in patients (pts) with advanced gastric cancer, gastroesophageal junction cancer, and esophageal adenocarcinoma (GC/GEJC/EAC): CheckMate 649 Chinese subgroup analysis 4-year (yr) follow-up. J Clin Oncol 2024; 42: 318.

14.

Qiu

Kato

, et al. Tislelizumab plus chemotherapy versus placebo plus chemotherapy as first line treatment for advanced gastric or gastro-oesophageal junction adenocarcinoma: RATIONALE-305 randomised, double blind, phase 3 trial. BMJ 2024; 385: e078876.

15.

Jiang

Pan

, et al. Sintilimab plus chemotherapy for unresectable gastric or gastroesophageal junction cancer: the ORIENT-16 randomized clinical trial. JAMA 2023; 330: 2064–2074.

16.

Zhang

Wang

, et al. First-line sugemalimab plus chemotherapy for advanced gastric cancer: the GEMSTONE-303 randomized clinical trial. JAMA 2025; 333(15): 1305–1314.

17.

Shen

Zhang

, et al. First-line cadonilimab plus chemotherapy in HER2-negative advanced gastric or gastroesophageal junction adenocarcinoma: a randomized, double-blind, phase 3 trial. Nat Med 2025; 31(4): 1163–1170.

18.

Husereau

Drummond

Augustovski

, et al. Consolidated Health economic evaluation reporting standards 2022 (CHEERS 2022) statement: updated reporting guidance for health economic evaluations. Value Health 2022; 25: 3–9.

19.

Lang

Zhang

, et al. Cost-effectiveness analysis of tislelizumab plus chemotherapy versus placebo plus chemotherapy as first-line treatment for advanced gastric or gastroesophageal junction adenocarcinoma: perspectives from the United States and China. Front Pharmacol 2024; 15: 1461571.

20.

Chen

Jing

, et al. Cost-effectiveness analysis of sugemalimab in combination with chemotherapy as first-line treatment in Chinese patients with metastatic NSCLC. Lung Cancer 2022; 174: 157–164.

21.

Chen

Sun

Zhang

, et al. Cost-effectiveness analysis of Tislelizumab vs Sorafenib as the first-line treatment of unresectable hepatocellular carcinoma. PLoS One 2024; 19: e0295090.

22.

Zhu

Liu

Ding

, et al. Pembrolizumab plus chemotherapy as first-line treatment for advanced esophageal cancer: a cost-effectiveness analysis. Adv Ther 2022; 39: 2614–2629.

23.

Zhu

Liu

Zhu

, et al. Immune checkpoint inhibitors plus chemotherapy for HER2-negative advanced gastric/gastroesophageal junction cancer: a cost-effectiveness analysis. Therap Adv Gastroenterol 2023; 16: 17562848231207200.

24.

Han

Zeng

Tian

, et al. First-line immunotherapy combinations for recurrent or metastatic nasopharyngeal carcinoma: an updated network meta-analysis and cost-effectiveness analysis. Head Neck 2023; 45: 2246–2258.

25.

Liu

Wang

Chen

, et al. Cost-effectiveness of sintilimab plus chemotherapy versus chemotherapy alone as first-line treatment of locally advanced or metastatic oesophageal squamous cell carcinoma. Front Immunol 2023; 14: 1092385.

26.

Zhu

Han

Cost-effectiveness analysis of Pembrolizumab versus chemotherapy as first-line treatment for mismatch-repair-deficient (dMMR) or microsatellite-instability-high (MSI-H) advanced or metastatic colorectal cancer from the perspective of the Chinese health-care system. BMC Health Serv Res 2023; 23: 1083.

27.

Rabin

de Charro

EQ-5D: a measure of health status from the EuroQol Group. Ann Med 2001; 33: 337–343.

28.

Brazier

Harper

Jones

, et al. Validating the SF-36 health survey questionnaire: new outcome measure for primary care. BMJ 1992; 305: 160–164.

29.

Kaplan

Hays

RD.

Health-related quality of life measurement in public health. Annu Rev Public Health 2022; 43: 355–373.

30.

Lang

Deng

, et al. Cost-effectiveness analysis of pembrolizumab plus chemotherapy versus placebo plus chemotherapy for HER2-negative advanced gastric/gastroesophageal junction cancer in the Chinese healthcare system. Expert Rev Pharmacoecon Outcomes Res 2024; 24: 1027–1042.

31.

Ding

Liao

, et al. Cost-effectiveness analysis of atezolizumab plus chemotherapy in the first-line treatment of metastatic non-squamous non-small cell lung cancer. Adv Ther 2020; 37: 2116–2126.

32.

Long

Yao

, et al. Updated cost-effectiveness analysis of tislelizumab in combination with chemotherapy for the first-line treatment of advanced gastric cancer or gastroesophageal junction adenocarcinoma. Front Oncol 2024; 14: 1477722.

33.

Tang

, et al. Cost-effectiveness of nivolumab plus ipilimumab as first-line treatment for American patients with unresectable malignant pleural mesothelioma. Front Public Health 2022; 10: 947375.

34.

Hoyle

Green

Thompson-Coon

, et al. Cost-effectiveness of temsirolimus for first line treatment of advanced renal cell carcinoma. Value Health 2010; 13: 61–68.

35.

Yan

Lei

, et al. Stomach cancer burden in China: epidemiology and prevention. Chin J Cancer Res 2023; 35: 81–91.

36.

Zeng

Chen

Zheng

, et al. Changing cancer survival in China during 2003-15: a pooled analysis of 17 population-based cancer registries. Lancet Glob Health 2018; 6: e555–e567.

37.

Cai

Chen

Wang

, et al. Contemporary trends on expenditure of hospital care on total cancer and its subtypes in China during 2008-2017. Chin J Cancer Res 2021; 33: 627–636.

38.

Jiang

Wang

LXW

. Cost-effectiveness analysis of nivolumab plus standard chemotherapy versus chemotherapy alone for the first-line treatment of unresectable advanced or metastatic gastric cancer, gastroesophageal junction cancer, and esophageal adenocarcinoma. Int J Clin Pharm 2022; 44: 499–506.

39.

Lang

Lin

, et al. Pembrolizumab alone or in combination with chemotherapy versus chemotherapy for advanced gastric cancer: a cost-effectiveness analysis. Cancer Med 2023; 12: 18447–18459.

40.

Xiang

, et al. Cost-effectiveness analysis of first-line sintilimab plus chemotherapy vs. Chemotherapy alone for unresectable advanced or metastatic gastric or gastroesophageal junction cancer in China. Front Pharmacol 2024; 15: 1411571.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.52 MB