Article Text

Original research
Updated cost-effectiveness analysis of adebrelimab plus chemotherapy for extensive-stage small cell lung cancer in China
  1. Yunchun Long1,
  2. Hao Wang2,
  3. Xianhai Xie3,
  4. Junlin Li3,
  5. Yuan Xu2,
  6. Yujie Zhou4
  1. 1Department of Pharmacy, China Pharmaceutical University Nanjing Drum Tower Hospital, Nanjing, China
  2. 2Department of Pharmacy, Nanjing Drum Tower Hospital, Nanjing, China
  3. 3School of Basic Medicine and Clinical Pharmacy, China Pharmaceutical University, Nanjing, China
  4. 4Department of Respiratory and Critical Care Medicine, Nanjing Drum Tower Hospital, Nanjing, China
  1. Correspondence to Dr Yujie Zhou; yujiezhoum{at}


Objective The CAPSTONE-1 trial demonstrated that adebrelimab-based immunotherapy yielded a favourable survival benefit compared with chemotherapy for patients with extensive-stage small cell lung cancer (ES-SCLC). This study aims to evaluate the cost-effectiveness of this immunotherapy in the treatment of ES-SCLC from a healthcare system perspective in China.

Design The TreeAge Pro software was used to establish a three-state partitioned survival model. Survival data came from the CAPSTONE-1 trial (NCT03711305), and only direct medical costs were included. Utility values were obtained from the published literature. Sensitivity analysis was performed to explore the robustness of the model. The cost-effectiveness of immunotherapy was investigated through scenario and exploratory analyses in various settings.

Outcome measures Total costs, incremental costs, life years, quality-adjusted life-years (QALYs), incremental QALYs and incremental cost-effectiveness ratio (ICER).

Results The basic analysis revealed that the adebrelimab group achieved a total of 1.1 QALYs at a cost of US$65 385, while the placebo group attained 0.78 QALYs at a cost of US$12 741. ICER was US$163 893/QALY. Sensitivity analysis confirmed that the model was robust. Results from scenario and exploratory analyses indicated that the combination of adebrelimab and chemotherapy did not demonstrate cost-effectiveness in any scenario.

Conclusions From the perspective of the Chinese healthcare system, adebrelimab in combination with chemotherapy for the treatment of ES-SCLC was not economical compared with chemotherapy.

  • chemotherapy
  • health economics
  • respiratory tract tumours
  • immunology

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • A partitioned survival model was built using TreeAge Pro to assess the cost-effectiveness of combining adebrelimab with chemotherapy for the treatment of small cell lung cancer in China.

  • Due to the short follow-up time of clinical trials, the traditional approach is to extrapolate long-term survival data using the standard parameter model, and this study not only adopted the standard parameter model but also other more flexible models (Royston/Parmar spline model, mixture cure model and non-mixture cure model).

  • An unavoidable limitation of this study was that there was uncertainty in the extrapolated survival results.


The latest data released by the National Central Cancer Registry of China reveals that lung cancer holds the top position in terms of both new cases (820 000) and fatalities (657 000) among all malignant tumours in China.1 The economic burden of lung cancer will also result in significant financial losses for the country. One study estimates that tracheal, bronchial and lung cancer are projected to cause US$1.2 trillion in economic damages in China from 2020 to 2050.2 Small cell lung cancer (SCLC) accounts for approximately 13%–17% of lung cancer.3 4 The annual incidence of SCLC worldwide is approximately 250 000 cases, with a corresponding mortality rate of about 200 000 deaths.5 Due to atypical symptoms, two-thirds of patients are already in the extensive stage at diagnosis, and the 5-year survival rate is less than 7%.3 4

In the past decades, etoposide+cisplatin/carboplatin (EP) chemotherapy has been widely recommended as the standard first-line treatment, with patients receiving EP chemotherapy demonstrating a median survival of approximately 9–11 months.6 7 The majority of patients exhibit a favourable response to chemotherapy; however, relapses typically manifest within a time frame of 6–12 months, and the emergence of resistance is highly probable.8 Therefore, the search for an emerging treatment approach to effectively address the relapse propensity and drug resistance of SCLC has gained significant attention.

In recent years, immune checkpoint inhibitors (ICIs) have shown unique advantages in prolonging survival time and improving the quality of life, bringing new hope to patients with SCLC.9 Both the Impower133 and CASPIAN trials showed that the addition of programmed cell death ligand 1 (PD-L1) inhibitors to EP chemotherapy prolonged overall survival (OS) and progression-free survival (PFS).10 11 Adebrelimab is China’s first self-developed PD-L1 inhibitor. The CAPSTONE-1 study demonstrated that the combination of adebrelimab and chemotherapy resulted in a significant extension of median overall survival (mOS: 15.3 vs 12.8 months), a reduction in the risk of death by 27%, and an impressive 3-year survival rate of 21.1% (compared with 10.5% in the chemotherapy group) at a median follow-up duration of 45.8 months.12 13 Based on this trial, the National Medical Products Administration (NMPA) of China approved adebrelimab in combination with carboplatin and etoposide as first-line treatment for extensive-stage (ES)-SCLC.

Although adebrelimab has demonstrated a promising survival benefit, its economics are unclear. Therefore, this study aims to perform the cost-effectiveness evaluation of adebrelimab combined with chemotherapy in the treatment of ES-SCLC from the perspective of the healthcare system in China to provide reference for national medical insurance decision-making.


This study adhered to the Consolidated Health Economic Evaluation Reporting Standards (online supplemental table 1).14

Patients and treatment

The CAPSTONE-1 trial represented the sole phase 3 clinical trial investigating the efficacy and safety of adebrelimab in combination with chemotherapy for ES-SCLC.12 Consequently, only survival data from the CAPSTONE-1 trial were used in this study. Based on the CAPSTONE-1 study, the model included patients aged 18–75 years who had been diagnosed with ES-SCLC.12 Eligible patients were randomly assigned to the adebrelimab or placebo group and were treated as follows: all patients received four cycles of EP therapy: etoposide (100 mg/m2, days 1–3 per cycle) plus carboplatin (area under curve 5 mg/mL, day1 per cycle), each cycle lasting 21 days. Additionally, the adebrelimab group was treated with adebrelimab (20 mg/kg, day 1 per cycle) until disease progression. In the event of disease progression, patients could receive secondary therapy. According to the CAPSTONE-1 trial, subsequent anticancer therapy was administered to 61.3% of patients in the adebrelimab group and 73.7% of patients in the placebo group following disease progression. However, there was a situation of treatment cross-over with an unclear proportion, and no specific treatment plan was provided. Therefore, in this study, the treatment method was selected according to the guidelines of the Chinese Society of Clinical Oncology (CSCO) and National Comprehensive Cancer Network (NCCN), and the following assumptions were made: the subsequent treatment of all patients was chemotherapy (topotecan: 1.25 mg/m2/ day, days 1–5 per cycle), with best supportive care for those who did not receive subsequent therapy.7

Model structure

The TreeAge Pro software was used to establish a partitioned survival model that includes PFS, progressive disease (PD) and death (figure 1). All patients were assumed to enter the model in the PFS state. Referring to the CAPSTONE-1 trial, the cycle length was set to 21 days.12 The time horizon (TH) was set as 10 years for two primary rationales. First, reconstructed Kaplan-Meier (K-M) survival curves revealed a simulated 10-year mortality rate of 99% among patients. Second, it is well documented in the published literature that the majority of current studies on ES-SCLC typically employ a TH of 10 years.15–17 The willingness-to-pay (WTP) threshold was set as three times of the gross domestic product (GDP) per capita in 2022 (WTP=US$37 155/quality-adjusted life-year (QALY)) according to the Guidelines for Pharmacoeconomic Evaluation in China (2020 version), and the cost and utility values were discounted by an annual discount rate of 5%.18 The model output indicators comprise total cost and incremental cost, life-years (LYs), QALYs and incremental QALYs. The indicator of economic judgement, known as the incremental cost-effectiveness ratio (ICER), quantifies the additional cost that should be paid for each extra QALY obtained.

Figure 1

A three-state partitioned survival model simulating extensive-stage small cell lung cancer (ES-SCLC), (A) decision tree, (B) graph of the partitioned survival model. OS, overall survival; PD, progressive disease; PFS, progression-free survival.

Survival estimate

The survival data used in this study were derived from the CAPSTONE-1 trial, which was simulated and extended through mathematical model fitting.12 13 Initially, the Engage Digitizer software ( was used to take the points of survival curves in the clinical trial. Subsequently, using the Guyot method, the survHE package in the R software ( was used to reconstruct individual patient data based on the survival rate, time, sample size, number of risk and other data, transforming it into a data format suitable for survival analysis.19 Finally, the exponential distribution, Weibull distribution, Gompertz distribution, log-logistic distribution and log-normal distribution were used to fit the survival curve, respectively, and the fitting results were shown in online supplemental table 2. Goodness of fit was judged according to visual inspection, Akaike information criterion and Bayesian information criterion, ultimately selecting the log-logistic distribution to fit the survival curves. The fitting parameters and curves are shown in online supplemental table 3 and figure 2, respectively.

Figure 2

Kaplan-Meier (K-M) curve fitting and extrapolation. (A) PFS and OS curves of adebrelimab group, (B) PFS for peer review only and OS curves of placebo group. OS, overall survival; PFS, progression-free survival.

Costs estimate

This cost-effectiveness analysis was conducted from the perspective of the healthcare system, encompassing only direct medical expenses such as drug costs, adverse drug reaction (ADR) treatment costs, hospitalisation costs, follow-up costs and best supportive care costs. Drug-related cost information was sourced from the China Medical Information Network (, using the average listed price in all provinces and cities in the country in 2022. The cost of ADR treatment was calculated by multiplying the expense of drug treatment for severe ADR with the ADR incidence rate. As most adverse events below grade 3 did not require additional clinical intervention, only adverse events above grade 3 were considered in this study. To streamline the model, only adverse events with an incidence of ≥5% in the two groups were included, with this cost calculated only in the first cycle. Hospitalisation, follow-up and best supportive care costs were based on the average price of medical and health services in Jiangsu province. The cost parameters are presented in table 1.

Table 1

Cost and utility parameters

Health utility value estimate

As the CAPSTONE-1 trial did not collect quality of life assessments, and there is currently no study available on the utility value of SCLC, this study adopted utility values of non-SCLC (PFS stage: 0.804, PD stage: 0.321), based on data from a multinational study of the Chinese population.20 Disutility values associated with ADR were also obtained from the published literature. See table 1 for the specific utility values.

Sensitivity analysis

This study conducted a one-way sensitivity analysis by setting the range of parameters changes in the model, which included cost, utility value, discount rate, incidence of ADR, proportion receiving subsequent treatment, weight, body surface area (BSA) and creatinine clearance rate (CCR). The findings were presented using a tornado diagram. The high and low values of drug costs were obtained from the China Medical Information Network. It is noteworthy that, based on the domestic PD-(L)1 inhibitors’ price trend, adebrelimab’s price fluctuation range in this study is set at 20%–100% of the base value.The upper and lower limits of the health utility values were sourced from the relevant literature. The discount rate was 0%–8% according to the Guidelines for Pharmacoeconomic Evaluation in China (2020 version).18 Parameters for which the upper and lower bounds were not determined were adjusted within a range of ±20% from their baseline values (table 1).

The model underwent probabilistic sensitivity analysis to investigate the collective impact of simultaneously varying multiple parameters. The cost parameters followed a gamma distribution, while the utility value and incidence of ADR followed a beta distribution. Weight, BSA and CCR followed a normal distribution. A second-order Monte Carlo simulation was used to randomly sample 1000 times. The resulting simulation data were used to generate a cost-effectiveness scatter plot and cost-effectiveness acceptability curve (CEAC) for analysing the probability of each treatment plan being economically viable under the WTP threshold.

Scenario analysis

Several studies suggest a general decrease in the ICER of economic evaluations as the TH is extended, although a minority of studies indicate an increase in ICER with TH extension.21 To investigate the impact of TH on the economic evaluation of adebrelimab, this study conducted scenario analysis with TH set at 5 and 20 years.

Exploratory analysis

Exploratory analysis 1: Apart from adebrelimab, serplulimab is the sole domestically approved drug by the NMPA for first-line treatment of ES-SCLC.22 Despite a marginal disparity in survival benefit between adebrelimab and serplulimab was marginal (mOS:15.3 vs 15.4 months), there existed a substantial discrepancy in terms of pricing (price per cycle: US$2975 vs US$2362).23 Therefore, a cost-effectiveness analysis of adebrelimab plus EP versus serplulimab plus EP was performed in this study. The comparability of the two drugs was based on the similarity in key patient characteristics between the CAPSTONE-1 and ASTRUM-005 trials, including median age, sex ratio, smoking status, Eastern Cooperative Oncology Group Performance Status, and PD-L1 expression level (online supplemental table 4). Due to the absence of head-to-head clinical trials of adebrelimab against serplulimab, the survival rate of adebrelimab was used as the baseline data, and HRs derived from network meta analysis (NMA) were employed to adjust the survival data of the serplulimab group. First, the ‘gemtc’ package in R software was used for conducting Bayesian NMA. The deviance information criterion (DIC) was employed to assess the goodness of fit between the fixed and random effects models. If the difference in DIC values between these two models was less than 5, it indicated a consistent degree of fit, and preference was given to the model with the lower I2 value. Conversely, if the difference exceeded 5, priority was assigned to the model with the smaller DIC value. The initial value was determined using three Markov chains, and the number of iterations was set to 500 000. Based on the HRs of OS and PFS reported in the CAPSTONE-1 and ASTRUM-005 trial, the DIC differences between the fixed and random effects models for PFS and OS were 0.002 and 0.01, respectively, with an I2 value of 50%. Due to the large I2 value, a random effects model was employed in this study. The HRPFS and HROS for the serplulimab group compared with the adebrelimab group were found to be 0.71 and 0.88, respectively (table 1). Then, the fitting parameters (scale parameter (λ) and shape parameter (γ)) of the best fit (log-logistic distribution) of the PFS and OS curves for the serplulimab group were transformed by the following formula, referring to the method of Hoyle et al: γSEP= γcontrol drug, λSEP= λcontrol drug * HR.24 Finally, the survival rate of the serplulimab group was calculated based on the transformed parameters.

Exploratory analysis 2: The standard parametric model (SPM) is frequently the preferred approach inferring survival data to be used in economic models.25 But the limitation of the SPM lies in its ability to capture the risk function of a specific shape only, and when dealing with highly complex actual risk functions, the fitting effect of the SPM tends to be inadequate.26 27 Therefore, the survival curve was fitted using more flexible models in this study, including the Royston/Parmar spline model (R/PSM), mixture cure model (MCM) and non-MCM (NMCM).


Base-case results

The results of the basic analysis are shown in table 2. The adebrelimab group achieved a higher gain in QALYs compared with the placebo group, with an increment of 0.32 QALYs (1.1 vs 0.78 QALYs), at a cost of US$52 644 (US$65 385 vs US$12 741). However, considering the current WTP, the combination of adebrelimab and chemotherapy was not deemed economically viable when compared with chemotherapy alone due to its significantly higher ICER of US$163 893/QALY.

Table 2

The results of basic analysis, scenario analysis and exploratory analysis

Sensitivity analysis

The one-way sensitivity analysis showed that altering the parameters within a predetermined range had no impact on the conclusions (figure 3). The ICER was most significantly influenced by the price of adebrelimab. Variables such as weight, as well as utility values at the PFS and PD stages had a moderate impact on the model. Other parameters had minimal effects on the model.

Figure 3

Tornado diagram of one-way sensitivity analysis. C_HOS, cost of hospitalisation; EV, expected value; ICER, incremental cost-effectiveness ratio; PD, progressive disease; PFS, progression-free survival; WTP, willingness to pay.

The CEAC figure (online supplemental figure 1) suggested that adebrelimab group might not be a cost-effective treatment option when the WTP was below US$60 000/QALY. At WTP levels of US$105 000/QALY and US$135 000/QALY, the probabilities of the adebrelimab group being considered cost-effective is estimated to be 50% and 90%, respectively.

It can be seen from the scatter plot (online supplemental figure 2) that the ICER values of 1000 simulations all fall in the area above the WTP line, indicating that adebrelimab coupled with chemotherapy is not economical in comparison with chemotherapy.

Scenario analysis

The scenario analysis findings demonstrated a gradual decrease in the ICER as the TH was extended. When the TH was set at 5 years, the ICER was US$199 496/QALY. However, with a simulation horizon of 20 years, the ICER decreased to US$148 166/QALY.

Exploratory analysis

Exploratory analysis 1: The results (table 2) demonstrated that despite the adebrelimab group yielding a higher QALYs, it also incurred a correspondingly higher cost compared with the serplulimab group (1.1 vs 1 QALYs, US$65 385 vs US$50 251). The calculated ICER was US$155 956/QALY, significantly surpassing the WTP.

Exploratory analysis 2: The findings of fitting and extrapolating OS curves using flexible models are illustrated in figure 2. The ICER calculated using both the mixture and NMCMs exhibited relatively close values, amounting to US$141 186/QALY and US$141 479/QALY, respectively. Meanwhile, the ICER determined through employment of the R/PSM was estimated at US$146 264/QALY.


The combination of PD-L1 inhibitors with chemotherapy has been shown to significantly improve survival in patients with ES-SCLC, as demonstrated by international IMpower133 and CASPIAN studies. However, it should be noted that these two studies had limited inclusion of Chinese patients, resulting in a lack of data specifically pertaining to this patient population.10 11 The CAPSTONE-1 study, in contrast to the two preceding studies, included a comprehensive cohort of Chinese patients.12 Consequently, the findings from the CAPSTONE-1 study robustly demonstrate that the combination of adebrelimab and chemotherapy confers substantial survival advantages for Chinese patients with ES-SCLC.

The evaluation of novel therapies plays a crucial role in the allocation of limited medical resources and the management of escalating healthcare costs. Consequently, this study assessed the cost-effectiveness of adebrelimab coupled with chemotherapy as a first-line treatment for ES-SCLC. The results demonstrated that the combination therapy was not cost-effective. The findings of this study align with the cost-effectiveness analyses conducted on most introduced PD-(L)1 inhibitors, indicating that these innovative agents do not present an economically feasible choice for cancer treatment.28–31 This is attributed to both the lower WTP and the high cost of immunotherapy. The average ICER of antitumour drugs was found to be approximately US$138 582/QALY, with a WTP range between US$100 000/QALY and US$150 000/QALY. Conversely, the average ICER of other drugs amounted to around US$49 913/QALY, while the average WTP ranged from US$50 00/QALY to US$100 000/QALY.32 The Guidelines for Pharmacoeconomic Evaluation in China (2020 version) recommend setting the WTP at 1–3 times per capita GDP. However, adhering to this recommendation may pose challenges in assessing the economic value of innovative drugs. Consequently, some scholars argue that different WTP values should be adopted based on specific drug characteristics, with orphan drugs warranting a higher WTP compared with common drugs.32

The one-way sensitivity analysis revealed a significant impact of the adebrelimab price on the ICER. Even with an 80% reduction in the price of adebrelimab, the ICER (US$49 115/QALY) remained higher than the WTP (US$37 155/QALY). Camrelizumab and sintilimab, independently developed PD-1 inhibitors in China, initially had price tags of US$2861.48/200 mg and US$1132.74/100 mg, respectively. However, following negotiations with medical insurance providers, the prices were significantly reduced to US$423.15/200 mg and US$156.08/100 mg, respectively, representing a reduction of over 85%. Therefore, we reasonably believe that with the development of China’s economy, the probability of adebrelimab being cost-effective will be greatly increased in the future.

The TH of model is a crucial influencing factor of ICER, yet national guidelines lack clarity regarding the necessary duration for conducting health economic assessments. Notably, the WHO recommends that cost-effectiveness analysis should assess the value of all interventions over a 10-year TH.33 But for some malignant tumours with rapid progression and short survival time, it would not be consistent with clinical practice if the model were to select a longer TH. Moreover, extrapolation would also increase uncertainty in the long-term estimation of K-M curves. Similarly, if the model chooses a study with a short TH, the overall survival rate in the study will not fall to a low proportion by the time the study reaches the end point, and the model results will not reflect the lifetime economics of the treatment. Therefore, scenario analysis explored the influence of different TH on ICER. The results showed that the longer the TH, the larger incremental cost and incremental QALY, but the increase of incremental QALY can make up for the lack of incremental cost increase, resulting in lower ICER. This may be due to the fact that ICIs not only bring immediate effects but also entail that the anticancer effect persists even after the end of immunotherapy, which is the ‘delayed effect’ of immunotherapy.34

Previously, Wang et al conducted a meta-analysis of the cost-effectiveness of immunotherapy for ES-SCLC and found that adebrelimab+EP (AEP) and serplulimab+EP (SEP) were cost-effective compared with EP.35 Unfortunately, Wang et al did not directly conduct a cost-effectiveness comparison of AEP against SEP. However, in the real medical environment, stakeholders (doctors, patients, medical insurance managers, etc) are often faced with the choice between drugs with the same mechanism of action. In view of this situation, this study performed a cost-effectiveness study of AEP against SEP, which showed that AEP was not economical compared with SEP. Probably because the incremental QALY was small (0.03 QALYs), the ICER was calculated as incremental cost divided by incremental QALY, and a small survival benefit gap could easily lead to a higher ICER value.

Commonly used statistical methods for survival analysis include non-parametric methods, semiparametric methods and parametric methods (SPMs, etc). The SPM is most popularly applied in the economic evaluation of antitumour drugs.25 However, in recent years, advancements in innovative therapies have facilitated the potential cure of cancer. There are still some patients who survive in the late follow-up of clinical trials, and the survival curve has a non-zero plateau tailing phenomenon, which is inconsistent with the assumption that patients in SPM must have an endpoint event (disease progression or death), resulting in the inability of SPM to identify non-disease-related deaths. Consequently, several methods have been developed to refine the extrapolation of survival curve fitting, and the R/PSM, MCM, NMCM are some of the new methods.36 Therefore, exploratory analysis 2 used different survival models to fit and extrapolate survival curves. The results indicated that the ICERs for the R/PSM and SPM were higher, amounting to US$146 264/QALY) and US$163 893/QALY, respectively. The larger ICERs may be attributed to the fact that these two models underestimate the therapeutic efficacy of immunotherapy.36 When the survival data are mature and there are cured patients, the MCM and NMCM can fit the survival curve more accurately, so they are considered to be more in line with clinical practice, and the ICER obtained by using these two models would be more accurate.36 37 Unfortunately, due to the limited duration of follow-up time and immature survival data of the CAPSTONE-1 trial, it is not feasible to judge which model yielded more accurate results in this study. However, both the basic analysis and exploratory analysis 2 consistently demonstrated that adebrelimab plus chemotherapy did not demonstrate cost-effectiveness irrespective of the modelling approach employed. As the SPM is the most commonly used model in economic evaluation of antitumour drugs, and most published cost-effectiveness studies of immunosuppressive therapies for ES-SCLC have also used the SPM, it is still recommended that the results obtained with the SPM should be adopted.25

To the best of our knowledge, this study represents the inaugural cost-effectiveness analysis using 3-year survival data derived from adebrelimab for ES-SCLC. Previously, You et al also conducted a study on the economic evaluation of adebrelimab plus chemotherapy and reported that this combination therapy was deemed cost-effective (ICER: US$25 914/QALY), which was contradictory to our results.38 The existence of this phenomenon can be attributed to two primary factors. First, the price of adebrelimab was not disclosed during their economic evaluation, and they estimated the cost (US$25.77/100 mg) based on that of sintilimab. However, recent published data reveal that adebrelimab is priced at US$1372.93/600 mg, which is nearly nine times higher than the previously assumed cost. Second, You et al used 2-year survival data instead of the most up-to-date 3-year survival data. The pivotal parameters in pharmacoeconomic evaluation encompass cost and survival data. The utilisation of more precise data in this study compared with that employed by You et al enhances the reliability of the results.

There are also some limitations in this study. First, owing to the short follow-up time of the CAPSTONE-1 trial, we had to extrapolate long-term survival data based on the K-M curves, and there was uncertainty in the results obtained via extrapolation. Second, the current lack of international research on the utility value of SCLC necessitates the use of health utility values for non-SCLC. However, one-way sensitivity analysis demonstrated that altering the utility values did not alter the results. Third, due to the lack of a specific treatment plan after disease progression in the CAPSTONE-1 trial, subsequent anticancer therapy was selected based on the CSCO and NCCN guidelines, thus disregarding individual variances. Fourth, there a was cross-over after disease progression, but there was a lack of sufficient information to assess the impact of such crossover on ICER. Finally, the results may be slightly influenced by certain underlying assumptions in this study. Notwithstanding the aforementioned limitations, the results of the sensitivity analysis have demonstrated the robustness of the model, thereby indicating that the findings remain valuable.


From the perspective of the Chinese healthcare system, adding adebrelimab to chemotherapy is not cost-effective.

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.

Ethics statements

Patient consent for publication

Ethics approval

Not required.


The authors would like to thank Editage ( for English language editing.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • YL and HW contributed equally.

  • Contributors Conception and design: YL and HW. Administrative support: YX and YZ. Provision of study materials or patients: YX and YZ. Collection and assembly of data: YL, HW, XX and JL. Data analysis and interpretation: YL, HW, XX and JL. Manuscript writing and final approval of the manuscript: YL, XX and JL. All authors contributed to the article and approved the submitted version. YZ accepts full responsibility for this work as guarantor.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.