Cost-effectiveness of PD-L1 testing to guide immunotherapy for patients with advanced non-small-cell lung cancer in China

Lung cancer is the leading cause of morbidity and mortality among all malignancies in China, with approximately 1,060,600 new cases and 733,300 lung cancer-related deaths reported by the National Cancer Center [1]. Non-small-cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer patients, with 70% of NSCLC diagnoses occurring at advanced stages [2,3]. In recent years, immunotherapy, particularly inhibitors targeting programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) checkpoints, has achieved significant breakthroughs in lung cancer treatment, resulting in notable survival benefits. The National Health Commission of the People's Republic of China (NHCPRC) guideline recommends immunotherapy as a first-line treatment for advanced NSCLC patients with high PD-L1 expression levels (tumor proportion score [TPS] ≥50% or tumor cells [TC] ≥50%) [4]. According to Chinese Expert Consensus on Standards of PD-L1 Immunohistochemistry (IHC) Testing for NSCLC to identify patients likely to benefit from PD-1/PD-L1 inhibitors, IHC is an effective and the most widely used method to assess PD-L1 expression in tumor tissue [5].

The clinical implementation of PD-L1 testing is complex. Different PD-1/PD-L1 inhibitors were originally developed with specific IHC assays, leading to a one-drug, one-diagnostic paradigm. This creates challenges for pathology labs regarding workflow and resource management. Different PD-1/PD-L1 inhibitors corresponds to different PD-L1 clones and platforms; thus, each requires specific PD-L1 IHC testing assays. When using assays with different clones to test PD-L1 expression in NSCLC tumor tissue, variations in results can occur due to differences in staining performance and other factors. In addition, the intra-assay consistency of the results can vary significantly depending on the antibody concentrate and assay used [6]. Therefore, the selection of assays can affect the sensitivity and specificity of PD-L1 testing in NSCLC patients, potentially leading to variations in diagnostic accuracy. This, in turn, may affect the selection of treatment regimens, ultimately impacting therapeutic outcomes and overall treatment costs. As such, excellent testing performance is critical for guiding appropriate selections of clinical medication and predicting patient prognosis. The Blueprint study is known as the pivotal investigation into the comparability of various IHC assays in both clinical and real-world settings. The Blueprint study found that while three PD-L1 IHC assays (SP263, 22C3 and 28-8) yielded largely concordant results, one assay (SP142) identified significantly fewer PD-L1-positive tumors than others. Such findings highlight the impact of assay variability on patient selection for immunotherapy and underscore the importance of choosing an appropriate PD-L1 test [7].

In China, several PD-L1 IHC testing assays have been approved for the use in guiding immunotherapy in NSCLC patients. VENTANA PD-L1 (SP263) Assay, Dako PD-L1 IHC 22C3 assay, and Dako 22C3 antibody concentrate will be the major PD-L1 IHC testing assays used as companion or complementary diagnostics for the immunotherapy that targets PD-1/PD-L1 in China market. This study aims to evaluate the cost-effectiveness of these three PD-L1 testing assays and provide evidence-based guidance for their clinical selection, considering testing costs, treatment costs and clinical benefit to patients. This study is, to our knowledge, one of the first to compare different PD-L1 diagnostic assays in terms of cost-effectiveness for NSCLC treatment in China, addressing a critical gap in current literature and providing novel insights for precision medicine strategies.

IHC serves as a reliable and effective method for assessing PD-L1 expression in tumor tissues, providing predictive value for the efficacy of PD-1/PD-L1 inhibitors. This study calculated the costs and health outcomes of different PD-L1 IHC testing assays employed for and guiding immunotherapy in NSCLC patients using a decision tree model, further demonstrating the economic value of IHC assays. The results showed that VENTANA PD-L1 (SP263) Assay enhanced the sensitivity and specificity of PD-L1 testing, thereby reducing misdiagnosis and increasing the proportion of patients accurately diagnosed and treated. Additionally, for patients with advanced NSCLC, VENTANA PD-L1 (SP263) Assay can guide multiple PD-1/PD-L1 immune checkpoint inhibitors (ICIs) monotherapies, offering a broader range of treatment options. When compared with Dako PD-L1 IHC 22C3 assay and Dako 22C3 antibody concentrate which only guide a single immunotherapy treatment, VENTANA PD-L1 (SP263) Assay emerges as a cost-effective option. This finding aligns with other cost-effective studies over the world. For example, a cost-effectiveness study comparing in vitro diagnostic with laboratory developed test showed that although the in vitro diagnostic was slightly more expensive, it was more cost-effective and significantly more accurate for guiding immunotherapy in patients with positive PD-L1 expression. This increased the likelihood of patients receiving appropriate treatment and maximized health benefits [26]. Another cost-effectiveness study in Brazil compared the clinical and economic impact of two PD-L1 assays, SP142 versus 22C3, and found that SP142 assay is likely to improve PFS and generate cost savings [27].

In addition, our study found that the drug cost of immunotherapies and chemotherapy significantly impacted the ICER results. Similar observations were reported in several cost-effectiveness studies on the first-line treatment of advanced NSCLC. Studies from Swedish and USA demonstrated that immunotherapy is cost-effective as the first-line treatment for metastatic NSCLC expressing high levels of PD-L1 when comparing pembrolizumab with chemotherapy [28,29]. A study in Hong Kong, China, also found immunotherapy cost-effective under a threshold of three-times the per capita GDP [30]. However, these studies did not consider the impact of PD-L1 testing on the health outcome results. A study by Wan et al. considered both the PD-L1 testing results and the differences in the selection of subsequent treatment plans, and compared the cost-effectiveness of different testing strategies in both China and the United States. The study showed that compared with chemotherapy, the pembrolizumab-chemotherapy combination treatment of NSCLC is not cost-effective in the US and Chinese healthcare systems at the WTP threshold of $100,000/QALY for the US and $27,351/QALY for China, but using PD-L1 test for patient selection could improve the cost-effective probabilities of immunotherapy for NSCLC [31].

A unique feature of our study is that when using different PD-L1 testing assays, the corresponding immunotherapy selection option varies. After careful matching between the PD-L1 testing assay and subsequent immunotherapy, our study indicated that the VENTANA PD-L1 (SP263) Assay testing assay is more cost-effective compared with other testing assays.

The study was subject to the following limitations. First, the accuracy data for the VENTANA PD-L1 (SP263) Assay, Dako PD-L1 IHC 22C3 assay and Dako 22C3 antibody concentrate were derived from research reports of NordiQC. There was a notable absence of representative domestic statistical data in China, which should be further enriched in future research. Furthermore, false positive and false negative data for each of the three testing methods were not publicly reported, and were therefore estimated based on the aggregated proportion of false-positive and false-negative rates reported in the NordiQC’s reports, combined with the acceptable rates of the three testing methods, respectively. Our analysis may also oversimply the multifactorial nature of diagnostic accuracy, as inter-laboratory reproducibility and inter-observer variability are critical factors in clinical practice. These were not explicitly modeled, representing a limitation. To mitigate the uncertainty of data estimation, expert interviews were conducted in this study to verify the value, but further evaluation can be performed when relevant data becomes available in the future.

Second, advanced NSCLC patients with false positive results are treated with atezolizumab or pembrolizumab monotherapy. According to clinical experts interviews, patients with low PD-L1 expression response poorly to targeted treatment. Therefore, the actual efficacy of these treatments in patients was assumed to be equivalent to that of chemotherapy in the study due to limited data availability. This may potentially lead to an underestimation of the efficacy of immunotherapy in patients with false-positive results. Although the number of affected patients was small (<2%), more data are needed in the future to further validate the study results.

Finally, this study evaluated the costs and health outcomes to patients over a one-year time horizon. As it used the proportion of patients correctly diagnosed and treated as a measurement of health outcomes, it may not comprehensively reflect the long-term clinical benefits to patients. Future research could dive deeper into the cost-effectiveness in long-term scenarios (e.g., 3 years, 5 years, or life time).

Effective testing of biomarkers to accurately identify the population that will benefit is a key aspect of precision treatment. PD-L1 has been widely used as an efficacy predictive biomarker to guide the use of ICIs in the NSCLC treatment [29]. VENTANA PD-L1 (SP263) Assay, as a cost-effective PD-L1 assay, can be used to identify the group that will benefit most from immunotherapy and to predict patient prognosis [5,6].

In conclusion, this study demonstrated that VENTANA PD-L1 (SP263) Assay was a cost-effective PD-L1 assay to guide immunotherapy in advanced NSCLC patients, offering higher QALYs at lower costs compared with the Dako PD-L1 IHC 22C3 assay and Dako 22C3 antibody concentrate. From a policy perspective, the finding suggests that healthcare decision-makers in China should consider encouraging the use of accurate, validated assays as it yields better patient outcomes while at lower costs. For instance, clinical guidelines and hospital protocols may be updated to prefer the VENTANA PD-L1 (SP263) Assay for PD-L1 testing in advanced NSCLC. Standardizing PD-L1 testing around a high-performance, cost-effective assay can enhance the overall value of NSCLC treatment and ensure patients are accurately identified for the most appropriate therapies.

Despite limitations related to data availability and assumptions regarding false positive cases, VENTANA PD-L1 (SP263) Assay presents an economically viable option by providing a broader range of treatment choices. This study served as a scientific foundation for selecting cost-effective diagnostics and ICIs monotherapy plans for NSCLC patients in China, thereby advancing the development of precision diagnosis and treatment of NSCLC in China. Ultimately, this work supports the advancement of precision diagnosis and sustainable cancer care in China.

F Zhou and Y Li were responsible for study methodology, data interpretation, manuscript writing and providing medical advisory. J Ming was responsible for the study execution, literature review, data acquisition and analysis. C Zhou and Y Chen were responsible for study design, quality control, providing medical and health economics advisory, manuscript review and final approval.

The authors thank the oncologists and pathologists participating in this study for providing important suggestions.

This study was funded by Roche Diagnostics (Shanghai) Co., Ltd. The authors have received no other financial and/or material support for this research or the creation of this work apart from that disclosed.

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No funded writing assistance was utilized in the production of this manuscript.

The authors state this article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.

All data generated and analyzed during this study are included in this manuscript. Datasets are available through the corresponding author on reasonable request.

This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/4.0/